# Beyond the Basics ### About this export | Field | Value | | --- | --- | | **content_type** | course | | **platform** | contentstack-academy | | **source_url** | https://www.contentstack.com/academy/courses/beyond-the-basics | | **language** | en | | **product_area** | lytics | | **learning_path** | standalone | | **course_id** | beyond-the-basics | | **slug** | beyond-the-basics | | **version** | 2026-03-01 | | **last_updated** | 2026-04-28 | | **status** | published | | **keywords** | ["lytics"] | | **summary_one_line** | Dive into more advanced topics to optimize your use of Lytics and unlock new activations. | | **total_duration_minutes** | 28 | | **lessons_count** | 5 | | **video_lessons_count** | 5 | | **text_lessons_count** | 0 | | **linked_learning_path** | standalone | | **linked_assessment_ref** | LMS_UNCONFIGURED_COURSE_ASSESSMENT | | **markdown_file_url** | /academy/md/courses/beyond-the-basics.md | | **generated_at** | 2026-04-28T06:55:34.936Z | | **intended_audience** | [] | | **prerequisites** | [] | | **related_courses** | [] | > **Academy MD v3** — companion `.md` for Ask AI. Quizzes and graded assessments are **LMS-only**; this file never contains answer keys. ## Course Overview | Metadata | Value | | --- | --- | | Catalog duration | 27m 59s | | Released (if known) | 2026-03-01 | | Product area | lytics | ### Description Dive into more advanced topics to optimize your use of Lytics and unlock new activations. ### Learning objectives 1. Follow each lesson in order. 2. Practice in a training stack using placeholders **YOUR_STACK_API_KEY** and **YOUR_DELIVERY_TOKEN** in local `.env` files only. 3. Validate API responses against the official documentation. ### Topics covered lytics ## Course structure ```text beyond-the-basics/ ├── 01-lytics-data-flow · video · 120s ├── 02-lytics-javascript-tag · video · 363s ├── 03-jobs-and-authorizations · video · 193s ├── 04-lookalike-models · video · 110s ├── 05-lytics-query-language · video · 893s ``` ## Lessons ### Lesson 01 — Lytics Data Flow #### Video details #### At a glance - **Title:** Lytics Data Flow - Data Transformation - **Duration:** 2m - **Media link:** https://cdn.jwplayer.com/previews/FQmwlMCw - **Publish date (unix):** 1751879785 #### Streaming renditions - application/vnd.apple.mpegurl - audio/mp4 · AAC Audio · 114318 kbps - video/mp4 · 180p · 246p · 147073 kbps - video/mp4 · 270p · 370p · 164152 kbps - video/mp4 · 360p · 494p · 185037 kbps - video/mp4 · 406p · 556p · 193660 kbps #### Timed text tracks (delivery) - **thumbnails:** `https://cdn.jwplayer.com/strips/FQmwlMCw-120.vtt` #### Transcript The next place that we're going to look in this process is going to be the LQL. LQL stands for Linux Query Language. It's a transformative language that we use to map your data as it's coming in onto the user profile. This can be very powerful in that it can enrich the downstream tools with additional data that they may not have had beforehand. And what this can also do is allow you to merge data that you wouldn't have otherwise been able to merge onto the user profile. An example of this would be if you had a first name field that existed in multiple different systems upstream of Linux. We can take those multiple fields and actually map them to the same place on the user profile. What this will allow us to do then is create a more educated user profile. And when we use the term educated user profile, what we mean is that it has more sources feeding into it and therefore is less specific to one type of channel or one type of user behavior. The reason that this is super, super important is that's going to give us a more colorful picture of what this user is doing and what different channels they're interacting with throughout your ecosystem. Finally, the last place in this data channel or data process is going to be our audiences. Once the user profiles have been created and mapped to, they will be able to go into audiences that can be built out in the UI of Linux. An audience is just going to be a group of users that have a certain set of criteria associated to them. So if we're going to think about creating an audience of users that have all interacted with a certain ad or a certain pop-up or a certain form, that's going to be the sort of criteria that we're going to want to set on the audience itself. Then anybody that's interacted with that information will fall into this audience, and that audience can then be sent downstream for activation. #### Subtitles (WebVTT) ```webvtt WEBVTT 1 00:00:00.000 --> 00:00:05.000 The next place that we're going to look in this process is going to be the LQL. 2 00:00:05.000 --> 00:00:08.000 LQL stands for Linux Query Language. 3 00:00:08.000 --> 00:00:12.000 It's a transformative language that we use to map your data 4 00:00:12.000 --> 00:00:15.000 as it's coming in onto the user profile. 5 00:00:15.000 --> 00:00:20.000 This can be very powerful in that it can enrich the downstream 6 00:00:20.000 --> 00:00:24.000 tools with additional data that they may not have had beforehand. 7 00:00:24.000 --> 00:00:28.000 And what this can also do is allow you to merge data that you wouldn't have 8 00:00:28.000 --> 00:00:32.000 otherwise been able to merge onto the user profile. 9 00:00:32.000 --> 00:00:37.000 An example of this would be if you had a first name field that existed in 10 00:00:37.000 --> 00:00:40.000 multiple different systems upstream of Linux. 11 00:00:40.000 --> 00:00:44.000 We can take those multiple fields and actually map them to the same place 12 00:00:44.000 --> 00:00:46.000 on the user profile. 13 00:00:46.000 --> 00:00:51.000 What this will allow us to do then is create a more educated user profile. 14 00:00:51.000 --> 00:00:54.000 And when we use the term educated user profile, 15 00:00:54.000 --> 00:00:59.000 what we mean is that it has more sources feeding into it 16 00:00:59.000 --> 00:01:03.000 and therefore is less specific to one type of channel 17 00:01:03.000 --> 00:01:05.000 or one type of user behavior. 18 00:01:05.000 --> 00:01:08.000 The reason that this is super, super important is that's going to give us 19 00:01:08.000 --> 00:01:12.000 a more colorful picture of what this user is doing and what different 20 00:01:12.000 --> 00:01:16.000 channels they're interacting with throughout your ecosystem. 21 00:01:16.000 --> 00:01:20.000 Finally, the last place in this data channel or data process 22 00:01:20.000 --> 00:01:22.000 is going to be our audiences. 23 00:01:22.000 --> 00:01:26.000 Once the user profiles have been created and mapped to, 24 00:01:26.000 --> 00:01:30.000 they will be able to go into audiences that can be built out in the UI 25 00:01:30.000 --> 00:01:31.000 of Linux. 26 00:01:31.000 --> 00:01:35.000 An audience is just going to be a group of users that have a certain set 27 00:01:35.000 --> 00:01:38.000 of criteria associated to them. 28 00:01:38.000 --> 00:01:41.000 So if we're going to think about creating an audience of users that have 29 00:01:41.000 --> 00:01:46.000 all interacted with a certain ad or a certain pop-up or a certain form, 30 00:01:46.000 --> 00:01:49.000 that's going to be the sort of criteria that we're going to want to set 31 00:01:49.000 --> 00:01:51.000 on the audience itself. 32 00:01:51.000 --> 00:01:55.000 Then anybody that's interacted with that information will fall into this 33 00:01:55.000 --> 00:02:00.000 audience, and that audience can then be sent downstream for activation. ``` ```transcript [00:00] The next place that we're going to look in this process is going to be the LQL. [00:05] LQL stands for Linux Query Language. [00:08] It's a transformative language that we use to map your data [00:12] as it's coming in onto the user profile. [00:15] This can be very powerful in that it can enrich the downstream [00:20] tools with additional data that they may not have had beforehand. [00:24] And what this can also do is allow you to merge data that you wouldn't have [00:28] otherwise been able to merge onto the user profile. [00:32] An example of this would be if you had a first name field that existed in [00:37] multiple different systems upstream of Linux. [00:40] We can take those multiple fields and actually map them to the same place [00:44] on the user profile. [00:46] What this will allow us to do then is create a more educated user profile. [00:51] And when we use the term educated user profile, [00:54] what we mean is that it has more sources feeding into it [00:59] and therefore is less specific to one type of channel [01:03] or one type of user behavior. [01:05] The reason that this is super, super important is that's going to give us [01:08] a more colorful picture of what this user is doing and what different [01:12] channels they're interacting with throughout your ecosystem. [01:16] Finally, the last place in this data channel or data process [01:20] is going to be our audiences. [01:22] Once the user profiles have been created and mapped to, [01:26] they will be able to go into audiences that can be built out in the UI [01:30] of Linux. [01:31] An audience is just going to be a group of users that have a certain set [01:35] of criteria associated to them. [01:38] So if we're going to think about creating an audience of users that have [01:41] all interacted with a certain ad or a certain pop-up or a certain form, [01:46] that's going to be the sort of criteria that we're going to want to set [01:49] on the audience itself. [01:51] Then anybody that's interacted with that information will fall into this [01:55] audience, and that audience can then be sent downstream for activation. ``` #### Lesson text Get an overview of how data flows in and out of Lytics, and the data transformations that occur inside Lytics. ## Lytics Data Flow ### Introduction #### What will I learn? * What are the main ingress and egress methods? * What data transformations take place inside of Lytics? * What are Events, LQL, and Audiences? This guide will give you an overview of the main steps the process of data flowing in and out of Lytics. **What's Ingress and Egress?** * Ingress: Data going into Lytics AKA data import * Egress: Data going out of Lytics AKA data export ### Data Ingress The first step is getting your customer data into Lytics. In this video (2.5 min), we will cover the main data ingress methods at a high-level. #### Methods to send data to Lytics Events can be ingested into Lytics through a few methods. * SFTP * CSV or JSON * Batch process - can run hourly, daily, monthly * Also can be used for a one-time backfill of historical data * Real-time APIs * Lytics JavaScript Tag - real-time connector, capturing onsite user behavior * Custom APIs (Collect endpoints) - powerful option for custom connections and setting the frequency needed for your use case * Bulk API * Separate channel to not interrupt the real-time data coming into Lytics #### Knowledge Check **What data formats are supported by the SFTP method? Select all that apply.** A. CSV B. Free Form C. JSON D. XML Answer: A, C **The Lytics JavaScript Tag constantly pushes data into Lytics in real time.** A. True B. False Answer: A ### Data Transformation Once your customer data has been ingested into Lytics, the next step is to **transform** that data so that it can be activated on for your marketing use cases. In the second part of the "Lytics Data Flow - Data Transformation" video (2 min), we'll give a brief overview of the data transformation process that occurs in Lytics: * Lytics Query Language (LQL) * User Profiles * Audiences #### Data Transformation in Lytics **Lytics Query Language (LQL)**: LQL is a transformative language that maps data as it comes into Lytics. LQL enables 2 key capabilities of the Lytics CDP:  * **Enrich** data - provide additional data to your downstream tools that wasn't available beforehand. * **Merge** data - combining multiple sources of data onto a single user profile. **User Profiles**: Lytics builds "educated" user profiles, meaning: * There are more data sources feeding into a profile (such as web, email, CRM, etc.) * Not limited to a specific channel (such as email) * Not limited to a specific type of user behavior (such as email opens) **Audiences**: Audiences are group of users that have certain set of criteria associated with them. * For example, users that have interacted with an ad, email, web form, etc. * Can be sent downstream for activation **What does LQL stand for?** A. Laughing Quite Loudly B. Lytics Query Language C. Learning Query Language Answer: B **Which of the following are benefits of LQL? Select all that apply.** A. Map incoming data into user profiles B. Enrich data with additional information from other sources C. Remove duplicated fields across data sources D. Merge data from multiple sources into an individual user profile Answer: A, B, D ### Data Egress Once your customer data has been transformed and enriched, you can then send user profile data and audiences from Lytics to your marketing tools for activation. This last section of the "Lytics Data Flow - Data Egress" video (3 min) will highlight a few of the main data egress options: * Website personalization * Email personalization * Enriching data in your database or CRM #### Personalize the web experience based on user behavior * Main channel, used the most * Real-time, based on audience behavior * For example, you can surface different information for first time visitors vs. frequent users #### Email customers the right content at the right time * Evaluate and serve the right campaigns to users based on user behavior and what's relevant to their interests. * Send email data (opens, clicks) back into Lytics to further refine audiences #### Enrich other data sources for downstream activations * Send user profiles with rich, behavioral data from Lytics to your CRM, database, or data lake * Opens up new opportunities for activation on your customer data that wouldn't have been possible otherwise **Match the data egress option to its example use case**. Data enrichment Show a modal relevant for a first-time site visitor Web Personalization Send a product newsletter based on the user's previous purchase Email Personalization Add cross-channel information to previously siloed data sources ## Summary and Next Steps ### Recap As a quick recap, in this Lytics Data Flow we guide we covered: * Data Ingress - different methods and different frequencies for sending data to Lytics. * Data Transformation - how Lytics merges and enriches data into actionable user profiles and audiences. * Data Egress - important downstream tools that you can activate on your customer data. #### Start with the RIGHT data Your first instinct may be to send ALL of your data to Lytics. However, our most successful implementations start with just the RIGHT data to execute their key use cases. ### More Resource #### Academy Courses * Lytics JavaScript Tag * Lytics Query Language #### Documentation * [How does Lytics work?](https://learn.lytics.com/documentation/product/features/getting-started/how-does-lytics-work) * [Onboarding Web Data](https://learn.lytics.com/documentation/product/features/data-onboarding-and-management/onboarding-web-data) * [Integrated Marketing Tools](https://learn.lytics.com/documentation/product/features/data-onboarding-and-management/integrated-marketing-tools) #### Key takeaways - Connect **Lytics Data Flow** back to your stack configuration before moving to the next module. - Capture one concrete artifact (screenshot, Postman call, or code snippet) that proves the step works in your environment. - Re-read the delivery versus management boundary for anything you changed in the entry model. ### Lesson 02 — Lytics JavaScript Tag #### Video details #### At a glance - **Title:** Lytics JavaScript Tag - **Duration:** 6m 3s - **Media link:** https://cdn.jwplayer.com/previews/d4p068ys - **Publish date (unix):** 1751881581 #### Streaming renditions - application/vnd.apple.mpegurl - video/mp4 · 180p · 248p · 150480 kbps #### Timed text tracks (delivery) - **thumbnails:** `https://cdn.jwplayer.com/strips/d4p068ys-120.vtt` #### Transcript In this video we're going to cover the JS tag and how to audit the data that's coming in from the JS tag in Linux. So the first place that we're going to start is talking about how data comes in from the web and how to install that JS tag on your site. So there are going to be a couple different ways to install the JS tag so that that data passes in to our data streams within Linux. The first way to install the method is through a tag manager such as Google Tag Manager or Teleum. There's usually going to be the Linux tag in the store itself where you can just select the tag, put your account ID in there, and then install the tag. However, if you have a more custom installation that requires a little bit more finesse to get that information in, you also have the option to manually upload the script onto the web page and add those additional features such as if you have a single page app or if you want to pass additional information in for when that tag loads. Additionally, if you have any extra information that you want to pass to Linux outside of what passes in the default data layer or in the default tag itself, you have options using the JS tag dot send functionality to get that information to pass into Linux. So downstream there's the default information that can pass into the data streams into what we call the default stream, which is just a stream of data coming from the web. You also have the option, as I mentioned before, to use the JS tag dot send method to get additional information into Linux. So both of those can be really, really useful when bringing data into Linux. So the next place that we're going to look or talk about is the data streams. The reason that we start there is that's the first place the data is going to hit when it comes into Linux. So this data stream is going to be the raw values that we see coming in from the web. This doesn't necessarily mean that it's going to be the only the values that we want to bring in map to the profiles, but any values that are available from the JS tag. That could include your default data layer, a more custom data layer that's been set up in your tag, or any custom sends and JSON payloads that are being sent using the JS tag dot send method. In the data streams tag in Linux, you're able to see the raw fields, the frequency of which they come in, and any sort of cardinality with that field. This gives us a ton of information on if that field's coming in, how often that field is coming in, and then if there's any additional information associated to that field. You can also see sample information in the raw data stream of a sample set of what we can expect to see in that value. This is going to be super helpful for data auditing because it will allow us to see if there's any sort of garbage data coming in, information that's not super valuable, nulls, special characters that we don't want to see in that field. This is sort of a window into that raw information. The next place that the data is going to flow is the data schema. So the data scheme is important because it's going to be only fields that we've explicitly mapped in the LQL itself. LQL, as a reminder, stands for a Lytics Query Language. In the Lytics Query file, we are going to pull in information that we want to map to the profile and only information we want to map to the profile. What that means is we may be receiving raw information in the data stream that never gets mapped and never gets surfaced in the UI. That is acceptable and can happen. However, that information will not show up in the data schema. The data schema can be super useful because it's going to talk to you about a couple different things. It's going to tell you what fields we've mapped in the LQL, the data types for those fields, how many users have those fields, if those fields are used in any audiences, and then also if there's any overlap between the different sources and those fields. So for instance, if we have something like an email, that's going to be seen throughout the different sources and that will show up in the data schema as something that's frequently used across different sources. This will also give you a health score or a percentage of how much of your data is actually shared across multiple sources. The final destination of your data, once it's come into the data schema, is going to be the user fields tab in Linux. The reason that the user fields tab is really important is it's going to show you only fields that have showed up in user profiles themselves. If it's in the data schema but not in the user fields, that means that we've mapped the fields in the LQL but we have yet to see them on any user profiles. The user fields are going to show you the type of data that's showing up in that field and then also how much of that data is showing up in that field. It will also show you through visuals what sort of data shows up more frequently in those fields using different types of graphs. Depending on the field, whether it be a Boolean, a String, or an Int, the type of visuals are going to be a little bit different. So that's something to note. From the user fields tab, you can also create audiences. So if you find that there's a field there that you're really really interested in sort of hashing out what it is and how many users have it, how often it's seen, then you can directly go to the audience builder with that field and find out that information. Something really important to remember about both the data schema and the user fields is that they're going to be able to break down per source. So you can see per stream what information is available and what data has been mapped or has also showed up on the user fields themselves. #### Subtitles (WebVTT) ```webvtt WEBVTT 1 00:00:00.000 --> 00:00:05.000 In this video we're going to cover the JS tag and how to audit the data that's 2 00:00:05.000 --> 00:00:08.920 coming in from the JS tag in Linux. So the first place that we're going to 3 00:00:08.920 --> 00:00:15.080 start is talking about how data comes in from the web and how to install that JS 4 00:00:15.080 --> 00:00:20.000 tag on your site. So there are going to be a couple different ways to install the JS 5 00:00:20.000 --> 00:00:25.640 tag so that that data passes in to our data streams within Linux. The first way 6 00:00:25.640 --> 00:00:30.240 to install the method is through a tag manager such as Google Tag Manager or 7 00:00:30.240 --> 00:00:35.240 Teleum. There's usually going to be the Linux tag in the store itself where you 8 00:00:35.240 --> 00:00:39.320 can just select the tag, put your account ID in there, and then install the tag. 9 00:00:39.320 --> 00:00:44.000 However, if you have a more custom installation that requires a little bit 10 00:00:44.000 --> 00:00:49.680 more finesse to get that information in, you also have the option to manually 11 00:00:49.680 --> 00:00:54.880 upload the script onto the web page and add those additional features such as if 12 00:00:54.880 --> 00:00:59.680 you have a single page app or if you want to pass additional information in 13 00:00:59.680 --> 00:01:06.880 for when that tag loads. Additionally, if you have any extra information that you 14 00:01:06.880 --> 00:01:11.240 want to pass to Linux outside of what passes in the default data layer or in 15 00:01:11.240 --> 00:01:17.680 the default tag itself, you have options using the JS tag dot send functionality 16 00:01:17.680 --> 00:01:23.200 to get that information to pass into Linux. So downstream there's the default 17 00:01:23.200 --> 00:01:27.640 information that can pass into the data streams into what we call the default 18 00:01:27.640 --> 00:01:32.280 stream, which is just a stream of data coming from the web. You also have the 19 00:01:32.280 --> 00:01:38.040 option, as I mentioned before, to use the JS tag dot send method to get additional 20 00:01:38.040 --> 00:01:43.480 information into Linux. So both of those can be really, really useful when 21 00:01:43.480 --> 00:01:48.400 bringing data into Linux. So the next place that we're going to look or talk 22 00:01:48.400 --> 00:01:52.920 about is the data streams. The reason that we start there is that's the first 23 00:01:52.920 --> 00:01:57.000 place the data is going to hit when it comes into Linux. So this data stream is 24 00:01:57.000 --> 00:02:01.600 going to be the raw values that we see coming in from the web. This doesn't 25 00:02:01.600 --> 00:02:05.520 necessarily mean that it's going to be the only the values that we want to 26 00:02:05.520 --> 00:02:10.080 bring in map to the profiles, but any values that are available from the JS 27 00:02:10.080 --> 00:02:14.720 tag. That could include your default data layer, a more custom data layer that's 28 00:02:14.720 --> 00:02:19.760 been set up in your tag, or any custom sends and JSON payloads that are being 29 00:02:19.800 --> 00:02:25.880 sent using the JS tag dot send method. In the data streams tag in Linux, you're 30 00:02:25.880 --> 00:02:31.840 able to see the raw fields, the frequency of which they come in, and any sort of 31 00:02:31.840 --> 00:02:38.000 cardinality with that field. This gives us a ton of information on if that 32 00:02:38.000 --> 00:02:42.040 field's coming in, how often that field is coming in, and then if there's any 33 00:02:42.040 --> 00:02:46.360 additional information associated to that field. You can also see sample 34 00:02:46.360 --> 00:02:51.760 information in the raw data stream of a sample set of what we can expect to see 35 00:02:51.760 --> 00:02:56.600 in that value. This is going to be super helpful for data auditing because it 36 00:02:56.600 --> 00:03:00.480 will allow us to see if there's any sort of garbage data coming in, information 37 00:03:00.480 --> 00:03:05.080 that's not super valuable, nulls, special characters that we don't want to see in 38 00:03:05.080 --> 00:03:10.800 that field. This is sort of a window into that raw information. The next place that 39 00:03:10.800 --> 00:03:15.800 the data is going to flow is the data schema. So the data scheme is important 40 00:03:15.800 --> 00:03:19.800 because it's going to be only fields that we've explicitly mapped in the LQL 41 00:03:19.800 --> 00:03:26.600 itself. LQL, as a reminder, stands for a Lytics Query Language. In the Lytics 42 00:03:26.600 --> 00:03:31.800 Query file, we are going to pull in information that we want to map to the 43 00:03:31.800 --> 00:03:37.360 profile and only information we want to map to the profile. What that means is we 44 00:03:37.360 --> 00:03:41.720 may be receiving raw information in the data stream that never gets mapped and 45 00:03:41.760 --> 00:03:47.320 never gets surfaced in the UI. That is acceptable and can happen. However, that 46 00:03:47.320 --> 00:03:52.400 information will not show up in the data schema. The data schema can be super 47 00:03:52.400 --> 00:03:56.640 useful because it's going to talk to you about a couple different things. It's 48 00:03:56.640 --> 00:04:00.120 going to tell you what fields we've mapped in the LQL, the data types for 49 00:04:00.120 --> 00:04:05.560 those fields, how many users have those fields, if those fields are used in any 50 00:04:05.560 --> 00:04:10.280 audiences, and then also if there's any overlap between the different sources and 51 00:04:10.280 --> 00:04:14.600 those fields. So for instance, if we have something like an email, that's 52 00:04:14.600 --> 00:04:18.440 going to be seen throughout the different sources and that will show up 53 00:04:18.440 --> 00:04:22.960 in the data schema as something that's frequently used across different 54 00:04:22.960 --> 00:04:27.920 sources. This will also give you a health score or a percentage of how much of 55 00:04:27.920 --> 00:04:33.120 your data is actually shared across multiple sources. The final destination 56 00:04:33.120 --> 00:04:36.960 of your data, once it's come into the data schema, is going to be the user 57 00:04:36.960 --> 00:04:42.040 fields tab in Linux. The reason that the user fields tab is really important is 58 00:04:42.040 --> 00:04:45.720 it's going to show you only fields that have showed up in user profiles 59 00:04:45.720 --> 00:04:51.480 themselves. If it's in the data schema but not in the user fields, that means 60 00:04:51.480 --> 00:04:56.600 that we've mapped the fields in the LQL but we have yet to see them on any user 61 00:04:56.600 --> 00:05:01.160 profiles. The user fields are going to show you the type of data that's showing 62 00:05:01.160 --> 00:05:05.840 up in that field and then also how much of that data is showing up in that 63 00:05:05.840 --> 00:05:10.560 field. It will also show you through visuals what sort of data shows up more 64 00:05:10.560 --> 00:05:15.200 frequently in those fields using different types of graphs. Depending on 65 00:05:15.200 --> 00:05:19.160 the field, whether it be a Boolean, a String, or an Int, the type of visuals 66 00:05:19.160 --> 00:05:23.520 are going to be a little bit different. So that's something to note. From the 67 00:05:23.520 --> 00:05:27.840 user fields tab, you can also create audiences. So if you find that there's a 68 00:05:27.840 --> 00:05:32.200 field there that you're really really interested in sort of hashing out what 69 00:05:32.200 --> 00:05:37.120 it is and how many users have it, how often it's seen, then you can directly go 70 00:05:37.120 --> 00:05:41.680 to the audience builder with that field and find out that information. Something 71 00:05:41.680 --> 00:05:46.240 really important to remember about both the data schema and the user 72 00:05:46.240 --> 00:05:51.200 fields is that they're going to be able to break down per source. So you can see per 73 00:05:51.200 --> 00:05:57.160 stream what information is available and what data has been mapped or has 74 00:05:57.160 --> 00:06:01.480 also showed up on the user fields themselves. ``` ```transcript [00:00] In this video we're going to cover the JS tag and how to audit the data that's [00:05] coming in from the JS tag in Linux. So the first place that we're going to [00:08] start is talking about how data comes in from the web and how to install that JS [00:15] tag on your site. So there are going to be a couple different ways to install the JS [00:20] tag so that that data passes in to our data streams within Linux. The first way [00:25] to install the method is through a tag manager such as Google Tag Manager or [00:30] Teleum. There's usually going to be the Linux tag in the store itself where you [00:35] can just select the tag, put your account ID in there, and then install the tag. [00:39] However, if you have a more custom installation that requires a little bit [00:44] more finesse to get that information in, you also have the option to manually [00:49] upload the script onto the web page and add those additional features such as if [00:54] you have a single page app or if you want to pass additional information in [00:59] for when that tag loads. Additionally, if you have any extra information that you [01:06] want to pass to Linux outside of what passes in the default data layer or in [01:11] the default tag itself, you have options using the JS tag dot send functionality [01:17] to get that information to pass into Linux. So downstream there's the default [01:23] information that can pass into the data streams into what we call the default [01:27] stream, which is just a stream of data coming from the web. You also have the [01:32] option, as I mentioned before, to use the JS tag dot send method to get additional [01:38] information into Linux. So both of those can be really, really useful when [01:43] bringing data into Linux. So the next place that we're going to look or talk [01:48] about is the data streams. The reason that we start there is that's the first [01:52] place the data is going to hit when it comes into Linux. So this data stream is [01:57] going to be the raw values that we see coming in from the web. This doesn't [02:01] necessarily mean that it's going to be the only the values that we want to [02:05] bring in map to the profiles, but any values that are available from the JS [02:10] tag. That could include your default data layer, a more custom data layer that's [02:14] been set up in your tag, or any custom sends and JSON payloads that are being [02:19] sent using the JS tag dot send method. In the data streams tag in Linux, you're [02:25] able to see the raw fields, the frequency of which they come in, and any sort of [02:31] cardinality with that field. This gives us a ton of information on if that [02:38] field's coming in, how often that field is coming in, and then if there's any [02:42] additional information associated to that field. You can also see sample [02:46] information in the raw data stream of a sample set of what we can expect to see [02:51] in that value. This is going to be super helpful for data auditing because it [02:56] will allow us to see if there's any sort of garbage data coming in, information [03:00] that's not super valuable, nulls, special characters that we don't want to see in [03:05] that field. This is sort of a window into that raw information. The next place that [03:10] the data is going to flow is the data schema. So the data scheme is important [03:15] because it's going to be only fields that we've explicitly mapped in the LQL [03:19] itself. LQL, as a reminder, stands for a Lytics Query Language. In the Lytics [03:26] Query file, we are going to pull in information that we want to map to the [03:31] profile and only information we want to map to the profile. What that means is we [03:37] may be receiving raw information in the data stream that never gets mapped and [03:41] never gets surfaced in the UI. That is acceptable and can happen. However, that [03:47] information will not show up in the data schema. The data schema can be super [03:52] useful because it's going to talk to you about a couple different things. It's [03:56] going to tell you what fields we've mapped in the LQL, the data types for [04:00] those fields, how many users have those fields, if those fields are used in any [04:05] audiences, and then also if there's any overlap between the different sources and [04:10] those fields. So for instance, if we have something like an email, that's [04:14] going to be seen throughout the different sources and that will show up [04:18] in the data schema as something that's frequently used across different [04:22] sources. This will also give you a health score or a percentage of how much of [04:27] your data is actually shared across multiple sources. The final destination [04:33] of your data, once it's come into the data schema, is going to be the user [04:36] fields tab in Linux. The reason that the user fields tab is really important is [04:42] it's going to show you only fields that have showed up in user profiles [04:45] themselves. If it's in the data schema but not in the user fields, that means [04:51] that we've mapped the fields in the LQL but we have yet to see them on any user ``` #### Lesson text Learn about the Lytics JavaScript tag and how it passes information from your website into Lytics user profiles. ## Video Tutorial ### What will I learn? * What default data and events are collected? * How raw event inputs are sent through the [Data Streams](https://learn.lytics.com/documentation/product/features/data-onboarding-and-management/data-streams). * How LQL mapping creates the [Data Schema](https://learn.lytics.com/documentation/product/features/data-onboarding-and-management/schema-audit). * How [User Fields](https://learn.lytics.com/documentation/product/features/data-onboarding-and-management/user-fields) are stored in the user profile. Watch the "Lytics JavaScript Tag" video (6 mins) for an introduction to the Lytics JavaScript tag and how it passes information from your website into Lytics user profiles. ### LQL Mapping The video briefly mentions LQL mapping functions. See the Lytics Query Language course for more details. ## Installing the Lytics JS Tag To find the Lytics JavaScript tag in the Lytics UI, click your Account Name at the bottom of the lefthand navigation menu and click [JavaScript Snippet](https://app.lytics.com/connect?view=v3) from the expanded menu. Make sure to use Version 3 of the Lytics tag. See the documentation for more information:  * [Installation & Configuration](https://learn.lytics.com/documentation/product/features/lytics-javascript-tag/using-version-3/installation-configuration) * [Version 3 documentation](https://learn.lytics.com/documentation/product/features/lytics-javascript-tag/version-3-improvements) You'll probably find it easier to install the Lytics JS Tag using a Tag Manager, like Google Tag Manager. See details at: [Working with Tag Managers](https://learn.lytics.com/documentation/product/features/lytics-javascript-tag/working-with-tag-managers) ## Knowledge Check Where can raw data fields be seen in the Lytics UI? Data Streams **Map the section of the data flow with the data points involved** Data Schema Includes mapped data on user profiles Data Stream Includes mapped fields, even if no data User Fields Includes raw values coming in from the web **Which data stream contains web data collected by the Lytics JavaScript Tag?** A. Experiences stream B. Default stream C. Custom stream Answer: B **Where can you find the percentage of user data being used in audiences?** A. Data Streams B. User Fields C. Schema Audit Answer: C **New fields added via the jstag.send method will automatically be added to a user's profile.** A. True B. False Answer: False - raw event data still needs to be mapped using Lytics Query Language (LQL) through the Data Schema first. ## More Resources ### Documentation Here are some recommended resources to continue learning about the Lytics JS Tag. * [Lytics JavaScript Tag introduction](https://learn.lytics.com/documentation/product/features/lytics-javascript-tag/introduction) * [Onboarding Web Data](https://learn.lytics.com/documentation/product/features/data-onboarding-and-management/onboarding-web-data) * [Collecting Data with V3 Tag](https://learn.lytics.com/documentation/product/features/lytics-javascript-tag/using-version-3/collecting-data) * [Troubleshooting: Verifying Data is sent to Lytics](https://learn.lytics.com/documentation/product/features/lytics-javascript-tag/troubleshooting#verify-data-is-being-sent-to-lytics) * [Installing Lytics Image Pixel](https://learn.lytics.com/documentation/product/features/data-onboarding-and-management/onboarding-web-data#lytics-image-pixel) #### Key takeaways - Connect **Lytics JavaScript Tag** back to your stack configuration before moving to the next module. - Capture one concrete artifact (screenshot, Postman call, or code snippet) that proves the step works in your environment. - Re-read the delivery versus management boundary for anything you changed in the entry model. ### Lesson 03 — Jobs and Authorizations #### Video details #### At a glance - **Title:** Jobs And Authorizations - **Duration:** 3m 13s - **Media link:** https://cdn.jwplayer.com/previews/igYBULhs - **Publish date (unix):** 1751882661 #### Streaming renditions - application/vnd.apple.mpegurl - audio/mp4 · AAC Audio · 113851 kbps - video/mp4 · 180p · 200p · 145281 kbps - video/mp4 · 270p · 300p · 163767 kbps - video/mp4 · 360p · 400p · 185365 kbps - video/mp4 · 406p · 450p · 198596 kbps - video/mp4 · 540p · 600p · 239894 kbps #### Timed text tracks (delivery) - **thumbnails:** `https://cdn.jwplayer.com/strips/igYBULhs-120.vtt` #### Transcript Long, long ago when we first started integrating third-party marketing tools into Lytx, we considered the providers themselves to be the first-class citizens. Look at all these providers! Wow! But as we grew and learned how our customers were using these integrations, what they wanted to do and how, we realized we needed to invert that original setup. Let's begin with authorizations, which are basically secret handshakes that Lytx and third-party tools use to recognize that it's safe to pass data to one of them. On the authorization dashboard seen here, you'll see a filterable, sortable, and searchable list of all of your authorizations. We now have labels in additions to descriptions to help you find the one you're looking for. We also include the health status of the authorization for those providers that we can run daily health checks against. Not all providers have support for that. The authorization summary gives you an at-a-glance view of the authorization's details and usage. From here, you can edit an authorization and also delete it, but you can only delete an authorization if it is not actively in use, which this one is, and that's why we get this error message. Below the activity charter, two tables, one for the associated jobs and another for associated experiences, since imported experiences still require authorization. While we were streamlining the flow for jobs and authorizations, we also extracted some of the technical jargon. We used to call our jobs works. The types of jobs were called workflows. Everything that existed before still exists, but we want both the language and your experience working in the Lytx application to feel more natural. Jobs are configured rule sets that determine the specific data passed between Lytx and a third-party tool. This can be data imported into Lytx, say, activity data, or exported from Lytx, such as audiences. On the jobs dashboard, you can see a list of all of your jobs, once again in a filterable, sortable, and searchable format. Note that the search picks up things from the name, the authorization, and the provider. Like authorizations, jobs now have names and descriptions to help you identify them quickly. The job summary contains at-a- glance details, including links to the authorization that it's using, and if it's supporting an imported experience, a link to that. The details tab also displays the configuration details for the job, and the logs tab will show you recent events most useful for jobs that are running continually. Thanks for watching this run-through of jobs and authorizations. #### Subtitles (WebVTT) ```webvtt WEBVTT 1 00:00:00.000 --> 00:00:04.600 Long, long ago when we first started integrating third-party marketing tools 2 00:00:04.600 --> 00:00:08.280 into Lytx, we considered the providers themselves to be the first-class 3 00:00:08.280 --> 00:00:13.480 citizens. Look at all these providers! Wow! But as we grew and learned how our 4 00:00:13.480 --> 00:00:17.440 customers were using these integrations, what they wanted to do and how, we 5 00:00:17.440 --> 00:00:24.800 realized we needed to invert that original setup. Let's begin with 6 00:00:24.800 --> 00:00:28.200 authorizations, which are basically secret handshakes that Lytx and 7 00:00:28.200 --> 00:00:32.360 third-party tools use to recognize that it's safe to pass data to one of them. On 8 00:00:32.360 --> 00:00:40.480 the authorization dashboard seen here, you'll see a filterable, sortable, and 9 00:00:40.480 --> 00:00:47.480 searchable list of all of your authorizations. We now have labels in 10 00:00:47.480 --> 00:00:53.440 additions to descriptions to help you find the one you're looking for. We also 11 00:00:53.440 --> 00:00:57.440 include the health status of the authorization for those providers that 12 00:00:57.440 --> 00:01:02.280 we can run daily health checks against. Not all providers have support for that. 13 00:01:02.280 --> 00:01:07.560 The authorization summary gives you an at-a-glance view of the authorization's 14 00:01:07.560 --> 00:01:13.880 details and usage. From here, you can edit an authorization and also delete it, but 15 00:01:13.880 --> 00:01:18.560 you can only delete an authorization if it is not actively in use, which this one 16 00:01:18.560 --> 00:01:23.480 is, and that's why we get this error message. Below the activity charter, two 17 00:01:23.480 --> 00:01:29.760 tables, one for the associated jobs and another for associated experiences, since 18 00:01:29.760 --> 00:01:35.920 imported experiences still require authorization. While we were streamlining 19 00:01:35.920 --> 00:01:40.160 the flow for jobs and authorizations, we also extracted some of the technical 20 00:01:40.160 --> 00:01:45.600 jargon. We used to call our jobs works. The types of jobs were called workflows. 21 00:01:45.600 --> 00:01:50.400 Everything that existed before still exists, but we want both the language and 22 00:01:50.400 --> 00:01:55.200 your experience working in the Lytx application to feel more natural. Jobs 23 00:01:55.200 --> 00:02:00.040 are configured rule sets that determine the specific data passed between Lytx 24 00:02:00.040 --> 00:02:05.520 and a third-party tool. This can be data imported into Lytx, say, activity data, or 25 00:02:05.520 --> 00:02:11.800 exported from Lytx, such as audiences. On the jobs dashboard, you can see a list of 26 00:02:11.800 --> 00:02:16.920 all of your jobs, once again in a filterable, 27 00:02:17.920 --> 00:02:29.000 sortable, and searchable format. Note that the search picks up things from the name, 28 00:02:29.000 --> 00:02:38.320 the authorization, and the provider. Like authorizations, jobs now have names and 29 00:02:38.320 --> 00:02:43.520 descriptions to help you identify them quickly. The job summary contains at-a- 30 00:02:43.600 --> 00:02:50.440 glance details, including links to the authorization that it's using, and if 31 00:02:50.440 --> 00:02:59.480 it's supporting an imported experience, a link to that. The details tab also 32 00:02:59.480 --> 00:03:04.120 displays the configuration details for the job, and the logs tab will show you 33 00:03:04.120 --> 00:03:09.560 recent events most useful for jobs that are running continually. Thanks for 34 00:03:09.560 --> 00:03:13.960 watching this run-through of jobs and authorizations. ``` ```transcript [00:00] Long, long ago when we first started integrating third-party marketing tools [00:04] into Lytx, we considered the providers themselves to be the first-class [00:08] citizens. Look at all these providers! Wow! But as we grew and learned how our [00:13] customers were using these integrations, what they wanted to do and how, we [00:17] realized we needed to invert that original setup. Let's begin with [00:24] authorizations, which are basically secret handshakes that Lytx and [00:28] third-party tools use to recognize that it's safe to pass data to one of them. On [00:32] the authorization dashboard seen here, you'll see a filterable, sortable, and [00:40] searchable list of all of your authorizations. We now have labels in [00:47] additions to descriptions to help you find the one you're looking for. We also [00:53] include the health status of the authorization for those providers that [00:57] we can run daily health checks against. Not all providers have support for that. [01:02] The authorization summary gives you an at-a-glance view of the authorization's [01:07] details and usage. From here, you can edit an authorization and also delete it, but [01:13] you can only delete an authorization if it is not actively in use, which this one [01:18] is, and that's why we get this error message. Below the activity charter, two [01:23] tables, one for the associated jobs and another for associated experiences, since [01:29] imported experiences still require authorization. While we were streamlining [01:35] the flow for jobs and authorizations, we also extracted some of the technical [01:40] jargon. We used to call our jobs works. The types of jobs were called workflows. [01:45] Everything that existed before still exists, but we want both the language and [01:50] your experience working in the Lytx application to feel more natural. Jobs [01:55] are configured rule sets that determine the specific data passed between Lytx [02:00] and a third-party tool. This can be data imported into Lytx, say, activity data, or [02:05] exported from Lytx, such as audiences. On the jobs dashboard, you can see a list of [02:11] all of your jobs, once again in a filterable, [02:17] sortable, and searchable format. Note that the search picks up things from the name, [02:29] the authorization, and the provider. Like authorizations, jobs now have names and [02:38] descriptions to help you identify them quickly. The job summary contains at-a- [02:43] glance details, including links to the authorization that it's using, and if [02:50] it's supporting an imported experience, a link to that. The details tab also [02:59] displays the configuration details for the job, and the logs tab will show you [03:04] recent events most useful for jobs that are running continually. Thanks for [03:09] watching this run-through of jobs and authorizations. ``` #### Lesson text Walk through setting up jobs and authorizations to connect Lytics to your marketing tools.[](https://learn.lytics.com/documentation/product/features/data-onboarding-and-management/onboarding-web-data#lytics-image-pixel) ## Introduction ### Integrations Overview **Note:** On January 10, 2023, we upgraded our UI with a new, refreshed interface. All of the underlying functionality is the same, but you will notice that things look a little different from this Academy guide. The most notable change is that the navigation menu has moved from the top of the app to the left side. We appreciate your patience as we work on updating our Academy. In this training guide, we will demonstrate how to set up jobs and authorizations step-by-step. Jobs are fundamental to moving your first-party data between Lytics and your marketing tools. To start, watch the short overview video "Jobs and Authorizations" (3 mins). ### Objectives and Key Terms #### Managing your data connections in Lytics The Data tab in Lytics is central to setting up and managing the ongoing flow of data between Lytics and your other marketing tools. What used to be labeled as the "**Integrations**" tab now lives within 2 new sections under the **Data** tab: **Jobs** and **Authorizations.** ![Jobs - Authorizations Nav.png](https://images.contentstack.io/v3/assets/bltebc53cfaf0dd6403/bltd91798a2a2383c49/686b9df7b6f4b47e96082834/Jobs_-_Authorizations_Nav.png) By the end of this training, you will have learned - How to create a job - How to create an authorization - What's new & improved? ## Key Terms * **Jobs:** responsible for moving your first-party data between Lytics and your marketing tools. * 3 main types: imports, exports, and enrichments (more on this in the next section). * Formerly referred to as "works" and "workflows." * **Authorizations:** responsible for connecting your provider tools to Lytics. * **Provider**: third-party tool that you are connecting with Lytics. * **Sources:** send data from a provider to Lytics. * **Destinations:** send data from Lytics to a provider. **Match the term to its definition** Provider How you move data between Lytics and other tools Authorizations How you allow other tools to connect to Lytics Jobs Third-party tool that you are connecting with Lytics ## Creating Jobs ### Job Types #### Three Main Job Types A provider tool may support one or more job types within Lytics. There are a variety of ways you can send data to and from Lytics, but they fall into 3 main categories: * **Import jobs**: Ingest data from a source tool into Lytics.  * Result in data coming into Lytics to populate user profiles. * **Export jobs:** Send data from Lytics to a destination tool. * Result in user profiles or audience membership being sent to your channel tools for campaign activation. * **Enrichment jobs:** Use a third-party service to enhance and enrich existing user profiles within Lytics.  * Take existing profile data and expand on it by pulling in more information. Many providers, such as **Mailchimp**, offer an import and export audiences option. ![mailchimp-job-type.png](https://images.contentstack.io/v3/assets/bltebc53cfaf0dd6403/bltf4d8083e08623956/686b9e903687732f96eec3b9/mailchimp-job-type.png) **Amazon Web Services** is an example of a provider with many job types including several different tools under the AWS umbrella - Kinesis, Pinpoint, S3, etc. ![aws-job-types.png](https://images.contentstack.io/v3/assets/bltebc53cfaf0dd6403/bltc4e076777850d748/686b9e914b4fe965bf5c4f12/aws-job-types.png) **Note:** For a few providers such as **Google Tag Manage**r, you will see **Other** listed as the job type. For more info on the connection types and techniques available, see our [Integrated Marketing Tools](https://learn.lytics.com/documentation/product/features/data-onboarding-and-management/integrated-marketing-tools) documentation. **True or false: All integrations support import and export jobs.** A. True B. False Answer: False - Many providers support import and export jobs but not all. You can see which job types are supported in app and on Learn Lytics. ### Jobs Dashboard The Jobs Dashboard gives an overview of your existing jobs and their status, and it's where you'll go to create a new job. #### How to create a job Check out the "Jobs Dashboard - Create A Job" video (3.5 mins) for a quick walk through of creating a job and an authorization together in a single flow. Regardless of the job type, you'll follow these steps: 1. Choose the provider. 2. Choose the job type. 3. Choose the authorization. 4. Configure your job. ![jobs-wizard.png](https://images.contentstack.io/v3/assets/bltebc53cfaf0dd6403/blt51edc03d0040b5f7/686b9fb54b4fe98df75c4f23/jobs-wizard.png)For more info, see our [Jobs Dashboard](https://learn.lytics.com/documentation/product/features/data-onboarding-and-management/jobs/jobs-dashboard) documentation. **Which of the following is NOT a step in the job creation flow?** A. Choose Provider B. Choose Authorization C. Create Custom Job Type D. Configure Job Answer: C ### Job Summary When you click on a job from the dashboard, you will be taken to the job's summary page. Here, you'll find important metadata about the job, such as the status, owner, creation date, and associated authorization, and the configuration details. ![lytics-job-summary-example.png](https://images.contentstack.io/v3/assets/bltebc53cfaf0dd6403/bltb0631065fb8f196c/686ba0fff1c7c2d1154ec68f/lytics-job-summary-example.png) #### Edit the name and description of a job  While creating a new job or editing an existing one, you can add a custom label and description. This is particularly important when you have numerous jobs of the same type running. #### Job Activity Metrics See how many user profiles were added, removed, or omitted on an hourly, daily, or weekly basis. #### Job Status and Logs Read up on job statuses and checking the logs in our [Job Summary](https://learn.lytics.com/documentation/product/features/data-onboarding-and-management/jobs/job-summary) documentation.  **For an existing job, which of the following can you edit? Select all that apply.** A. Name (label) B. Job Type C. Description D. Provider Answer: A, C - If you want to change the provider or job type, you should simply start a new job. **The Activity chart on the Job Summary displays which of the following metrics?** A. Potential reach, total reach, converted B. Profiles added, profiles removed, and profiles omitted C. Data fields, data source, data streams D. Audience size, conversions, conversion rate Answer: B ## Creating Authorizations ### Authorizations Dashboard The Authorizations Dashboard gives an at-a-glance view of your existing authorizations and their usage. ![Lytics\_authorizations\_dashboard.png](https://images.contentstack.io/v3/assets/bltebc53cfaf0dd6403/bltf35e98764b7c09c4/686ba21828e271aa3aa3b482/Lytics_authorizations_dashboard.png) #### Authorization health indicator Lytics checks the status of your authorizations automatically on a daily basis. For an authorization to marked as "**healthy**," it must be **valid** and **active**. [Learn More](https://learn.lytics.com/documentation/product/features/data-onboarding-and-management/authorizations/authorization-summary#what-determines-authorization-health?)  #### How to create an authorization There are two ways you can create an authorization: **Option 1**  From the Authorizations Dashboard, click **Create New Authorization** and complete the steps: 1. Choose the provider. 2. Select the authorization method. 3. Complete the configuration. **Option 2**  As shown in the video in the previous section, you can also create a new authorization **inline of the job creation flow.** This allows you to not interrupt your process of starting a job in order to make a new authorization first. ![auth-creation-in-job-wizard.png](https://images.contentstack.io/v3/assets/bltebc53cfaf0dd6403/blt6d58d1161a4d4755/686ba21802334227e192e5ee/auth-creation-in-job-wizard.png) If you go this route, you'll see two steps added to the job flow that you'll complete before configuring your job. #### Authorization Methods Note that some integration providers only have one authorization method, but others offer multiple methods such as API keys, OAuth, etc. Certain methods enable different job types, so be sure to select the option that supports your use case. For more info, see our [Authorizations Dashboard](https://learn.lytics.com/documentation/product/features/data-onboarding-and-management/authorizations/authorizations-dashboard) documentation. **A certain authorization method may be required to enable one job type vs. another.** A. True B. False Answer: True - Certain authorization methods enable different jobs such as real-time audience exports vs. bulk audience exports. If you are unsure which method to use, you can find more information in the integration documentation for your provider. ### Authorization Summary When you click on an authorization from the dashboard, you will be taken to its summary page. Similar to the job summary, you'll find important metadata about the authorization such as the status, owner, creation date, etc. ![authorizations-summary-example.png](https://images.contentstack.io/v3/assets/bltebc53cfaf0dd6403/blt2d34fee5ed4eaa19/686ba3ef321163ea84005610/authorizations-summary-example.png) #### Authorization activity metrics See how many API requests have been sent on an hourly, daily, or weekly basis. #### Edit the names and descriptions of authorizations Add a custom label and description while creating an authorization or editing an existing one. #### Change the authorization associated with a job You can replace or update an authorization. There are many reasons why an authorization may need to be changed such as incorrect credentials, employee associated with an authorization leaves, etc.  **NOTE:** You must select an authorization that supports the existing job configuration (e.g. authorizing for the specific account the job is targeting). #### Delete authorizations directly via the UI Keep your account current by deleting any invalid or unused authorizations.  **NOTE:** You can only delete an authorization if there are no active jobs using it. For more info, see our [Authorization Summary](https://learn.lytics.com/documentation/product/features/data-onboarding-and-management/authorizations/authorization-summary) documentation. **What metric is shown in the authorization activity chart?** API Requests #### Knowledge Check **Users can change the authorization associated with an existing job.** A. True B. False Answer: True - As long as the authorization method supports the job type selected, you can change the authorization or add a new one. **You can delete an authorization that is associated with a job when the job status is \_\_\_\_\_? Select all that apply.** A. Running B. Sleeping C. Failed D. Paused E. Completed Answer: C, E ## Wrap-Up ### Summary and More Resource **Nice job, learning about jobs!** For a quick summary of what we covered in this guide: * Why we made this series of improvements to reduce friction points. * How to create a job in the new, streamlined flow. * How to create an authorization - on its own or as part of creating a job. #### More Resources The docs referenced throughout this guide: * [Integrated Marketing Tools](https://learn.lytics.com/documentation/product/features/data-onboarding-and-management/integrated-marketing-tools)  * [Jobs Dashboard](https://learn.lytics.com/documentation/product/features/data-onboarding-and-management/jobs/jobs-dashboard) * [Job Summary](https://learn.lytics.com/documentation/product/features/data-onboarding-and-management/jobs/job-summary) * [Authorizations Dashboard](https://learn.lytics.com/documentation/product/features/data-onboarding-and-management/authorizations/authorizations-dashboard) * [Authorization Summary](https://learn.lytics.com/documentation/product/features/data-onboarding-and-management/authorizations/authorization-summary) For a recap of the series of improvements Lytics made to managing integrations: * [Streamlining Jobs Management](https://learn.lytics.com/product-updates/streamlining-jobs-management) * [Authorizations Get an Upgrade](https://learn.lytics.com/product-updates/authorizations-get-an-upgrade) * [Managing Integrations Made Easier](https://learn.lytics.com/product-updates/managing-integrations-made-easier) #### Key takeaways - Connect **Jobs and Authorizations** back to your stack configuration before moving to the next module. - Capture one concrete artifact (screenshot, Postman call, or code snippet) that proves the step works in your environment. - Re-read the delivery versus management boundary for anything you changed in the entry model. ### Lesson 04 — Lookalike Models #### Video details #### At a glance - **Title:** Lookalike Models - **Duration:** 1m 50s - **Media link:** https://cdn.jwplayer.com/previews/oXPDbF1w - **Publish date (unix):** 1751897371 #### Streaming renditions - application/vnd.apple.mpegurl - audio/mp4 · AAC Audio · 113573 kbps - video/mp4 · 180p · 200p · 152187 kbps - video/mp4 · 270p · 300p · 168241 kbps - video/mp4 · 360p · 400p · 185580 kbps - video/mp4 · 406p · 450p · 201175 kbps - video/mp4 · 540p · 600p · 244015 kbps #### Timed text tracks (delivery) - **thumbnails:** `https://cdn.jwplayer.com/strips/oXPDbF1w-120.vtt` #### Transcript Welcome to the laboratory, your hub for getting hands-on with all things data science inside of Linux. Here we will focus on bringing you unprecedented access to industry-leading machine learning and AI tools, all of which are self-serviceable. It really is like having a team of data scientists at your disposal. What if you knew which users were going to buy a product or unsubscribe before an action even took place? Now with lookalike models you can do just that by comparing those who have reached a particular goal, the target audience, to those who have not, your source. We can then build a predictive audience of those who are very likely to reach that goal in the future. The UI comes packed with a variety of existing and new features as well as improvements upon existing ones. Auto-tuning for example acts as an easy button ensuring even those with little experience can successfully build great models to power predictive audiences. We've improved both visuals and navigation to ensure you fully understand how models are being used and what makes each one unique. Clear and actionable debugging information ensures there's no mystery and always a next step. Finally, accuracy and reach help you find the perfect balance to support your desired use case. Lookalike models apply to many marketing use cases at any stage of the funnel. Locate customers who are likely to churn, find those most likely to make a purchase, and nurture those that show clear signs of becoming high-value users. And so much more. Get started today by creating your first model. #### Subtitles (WebVTT) ```webvtt WEBVTT 1 00:00:00.000 --> 00:00:04.960 Welcome to the laboratory, your hub for getting hands-on with all things data 2 00:00:04.960 --> 00:00:09.820 science inside of Linux. Here we will focus on bringing you unprecedented 3 00:00:09.820 --> 00:00:14.400 access to industry-leading machine learning and AI tools, all of which are 4 00:00:14.400 --> 00:00:18.880 self-serviceable. It really is like having a team of data scientists at your 5 00:00:18.880 --> 00:00:24.760 disposal. What if you knew which users were going to buy a product or unsubscribe 6 00:00:24.760 --> 00:00:29.720 before an action even took place? Now with lookalike models you can do just 7 00:00:29.720 --> 00:00:35.200 that by comparing those who have reached a particular goal, the target audience, to 8 00:00:35.200 --> 00:00:40.040 those who have not, your source. We can then build a predictive audience of 9 00:00:40.040 --> 00:00:46.560 those who are very likely to reach that goal in the future. The UI comes packed 10 00:00:46.560 --> 00:00:50.840 with a variety of existing and new features as well as improvements upon 11 00:00:50.840 --> 00:00:56.960 existing ones. Auto-tuning for example acts as an easy button ensuring even 12 00:00:56.960 --> 00:01:01.280 those with little experience can successfully build great models to power 13 00:01:01.280 --> 00:01:07.840 predictive audiences. We've improved both visuals and navigation to ensure you 14 00:01:07.840 --> 00:01:12.680 fully understand how models are being used and what makes each one unique. 15 00:01:12.680 --> 00:01:17.440 Clear and actionable debugging information ensures there's no mystery 16 00:01:17.440 --> 00:01:24.400 and always a next step. Finally, accuracy and reach help you find the perfect 17 00:01:24.400 --> 00:01:31.040 balance to support your desired use case. Lookalike models apply to many 18 00:01:31.040 --> 00:01:36.200 marketing use cases at any stage of the funnel. Locate customers who are likely 19 00:01:36.200 --> 00:01:42.080 to churn, find those most likely to make a purchase, and nurture those that show 20 00:01:42.080 --> 00:01:48.120 clear signs of becoming high-value users. And so much more. Get started today by 21 00:01:48.120 --> 00:01:51.400 creating your first model. ``` ```transcript [00:00] Welcome to the laboratory, your hub for getting hands-on with all things data [00:04] science inside of Linux. Here we will focus on bringing you unprecedented [00:09] access to industry-leading machine learning and AI tools, all of which are [00:14] self-serviceable. It really is like having a team of data scientists at your [00:18] disposal. What if you knew which users were going to buy a product or unsubscribe [00:24] before an action even took place? Now with lookalike models you can do just [00:29] that by comparing those who have reached a particular goal, the target audience, to [00:35] those who have not, your source. We can then build a predictive audience of [00:40] those who are very likely to reach that goal in the future. The UI comes packed [00:46] with a variety of existing and new features as well as improvements upon [00:50] existing ones. Auto-tuning for example acts as an easy button ensuring even [00:56] those with little experience can successfully build great models to power [01:01] predictive audiences. We've improved both visuals and navigation to ensure you [01:07] fully understand how models are being used and what makes each one unique. [01:12] Clear and actionable debugging information ensures there's no mystery [01:17] and always a next step. Finally, accuracy and reach help you find the perfect [01:24] balance to support your desired use case. Lookalike models apply to many [01:31] marketing use cases at any stage of the funnel. Locate customers who are likely [01:36] to churn, find those most likely to make a purchase, and nurture those that show [01:42] clear signs of becoming high-value users. And so much more. Get started today by [01:48] creating your first model. ``` #### Lesson text Gain an in-depth understanding of Lytics Lookalike Models and how to build Predictive Audiences that drive engagement and conversions. ## Introducing Lookalike Models ### Overview Note: On January 10, 2023, we upgraded our UI with a new, refreshed interface. All of the underlying functionality is the same, but you will notice that things look a little different from this Academy guide. The most notable change is that the navigation menu has moved from the top of the app to the left side. We appreciate your patience as we work on updating our Academy. Lytics makes it easy to build Lookalike Models and Predictive Audiences that drive engagement and conversions. ### What will I learn? * What are Lytics Lookalike Models? How can I use them? * Where do Lookalike Models live in Lytics? * How to start making models * How to understand your model performance and what to do next * How to create Predictive Audiences ### How are Lytics Lookalike Models different from other tools? You may be familiar with "Lookalike Audiences" on Facebook or "Similar Audiences" on Google. Lytics Lookalike Models are different in a few key ways: * Based on your own first-party, cross-channel data (no walled garden) * You can build and validate your own custom models (no black box) * Quick and easy interface for marketers to make models (no data science team required) * Provide real-time predictions - updating user scores for dynamic and effective targeting ### Lytics Laboratory Lytics provides a **data science workbench for your marketing teams** in the **Laboratory** section of the Lytics UI. This is where you'll go to build **Lookalike Models**. ![Click Laboratory > UI Models.png](https://images.contentstack.io/v3/assets/bltebc53cfaf0dd6403/bltb24a9f4465cfcc2c/686bd5514e325563d15687ae/Click_Laboratory_UI_Models.png) Check out the short video "Lookalike Models" (1.5 min). #### What are common use cases for Lookalike Models? * Optimize early stage funnel. Identify users least likely to progress to **cast an intelligently wide net**. * Unknown users to known users * Optimize late stage funnel. Identify users most likely to convert to **optimize higher touch experiences**. * Known Users to Purchasers * One time Purchasers to Repeat Purchasers * Nurture users who are likely to become high lifetime value (LTV) customers * Reduce churn by identifying known users who are likely to churn Learn more: [Lytics Laboratory documentation](https://learn.lytics.com/documentation/product/features/laboratory/introduction) **Which of the following are benefits of Lytics Lookalike Models? (select all that apply)** A. Models continuously update allowing real-time predictions B. You can import existing models from other tools into Lytics C. Custom models can be built from scratch D. Models are built using your own first-party data Answer: A, C, D ### Core Concepts Before learning how to build Lookalike Models in Lytics, it's important to understand the key terms that will be used throughout:  * **Model**: the output of customer data and a ML algorithm that can provide predictions on the data. * **Source Audience**: the group of users you want to reach with your marketing messages to get them to convert on a particular marketing goal.  * For example: “unknown users”. * Lytics calculates a score from 0 to 1 representing the user's likelihood to convert to the target audience. * **Target Audience**: the group of users that represents the desired outcome for users in the source audience (have converted in the past). * For example: “users with email addresses”. * **Predictive Audience**: the output audience(s) that are built using the predictive score from a Lookalike Model. For more definitions, see our [Glossary](https://learn.lytics.com/content/lytics-glossary). #### Segmentation Strategies: Manual vs. Machine Check out the [Lookalike Models Overview](https://learn.lytics.com/documentation/product/features/laboratory/lookalike-models/overview#segmentation-strategies:-manual-vs.-machine) for an example of how Lookalike Models work compared to a manual segmentation strategy. **Match the Lytics term to its definition.** Model Output of data and an algorithm that provides predictions Target Audience Users that have yet to convert on your marketing goal Source Audience Users that have converted in the past Predictive Audience Users available to target, using the predictive score output from a Lookalike Model ## Building Lookalike Models ### Lookalike Model Builder The Lookalike Model Builder provides an interface for marketers to quickly build custom models in Lytics. Navigate to the Lookalike Models dashboard and find the **Create New Model** button at the top right to get started. ![click-create-new-model.png](https://images.contentstack.io/v3/assets/bltebc53cfaf0dd6403/blt49a93af1fddafc74/686be0df47548f406f569af6/click-create-new-model.png) ### Configuration Options You can use a number of basic and advanced configuration options to build your model: * Basic options provide an intelligent starting point, making it quick and easy to get started.  * Advanced options give granular control to those who have more experience with model building. For most use cases, building a model with the basic configuration parameters is sufficient. The **only required parameters** are: * Source audience * Target audience  Selecting the right audiences is **very important for building a usable model**. We'll touch on this more in a later section. For descriptions and examples of all configuration options, see [Model Builder documentation](https://learn.lytics.com/documentation/product/features/laboratory/lookalike-models/model-builder).  **Which of the following basic configuration parameters are required?** A. Custom Model Name B. Source Audience C. Auto Tune D. Target Audience Answer: B, D **Models built with Auto Tune can also include advanced configuration parameters.** A. True B. False Answer: True ### Model Dashboard The Lookalike Model dashboard provides visibility of the health, status, and usage of your Lookalike Models. It lists all models in your account allowing you to search, sort, and view the summary, configuration, and diagnostic information for each model created. ![lytics-lookalike-model-summary.png](https://images.contentstack.io/v3/assets/bltebc53cfaf0dd6403/blt9aae40f62f0e263e/686be6318f61ad7bdddc66d6/lytics-lookalike-model-summary.png) Here's a quick rundown of the main sections: #### Model health chart Shows the total number of models for an account, and how many are considered **healthy** or unhealthy. A model is considered healthy if it can make accurate predictions. #### Model status  Displays how many of your models have been **activated**. An activated model is a model currently generating a score that is being written to user profiles, meaning the model's predictions can be used for audience segmentation and targeting. #### Model usage  Shows the percentage of all your audiences that are using activated models. #### Model table Allows you to search for any model in your account. Click any model to open its [summary](https://learn.lytics.com/documentation/product/features/laboratory/lookalike-models/model-summary) for more detailed information about that model. See the [documentation](https://learn.lytics.com/documentation/product/features/laboratory/lookalike-models/model-dashboard) for more details.  **Note: Lookalike Models Activation Limit** - You can only have 5 Lookalike Models active at a time, per the default account setting. If you're interested in a higher limit, please contact your Account Manager for more information. **To use your model's predictions in audience segmentation and targeting, the model must be \_\_\_\_\_\_\_?** A. Recently created B. Activated C. Top performing Answer: B ### Model Summary The Lookalike Model Summary surfaces valuable information and metrics about the performance and usage of your model. #### Understanding Accuracy and Reach At the top of each Model Summary page, you'll see two bars indicating the model's accuracy and reach scores. ![accuracy-reach.png](https://images.contentstack.io/v3/assets/bltebc53cfaf0dd6403/blt62f4dd47b1e7fbb7/686bec9c77a15576584c156d/accuracy-reach.png) * **Accuracy**: the precision of a model’s predictions * **Reach**: the relative size of a Lookalike Model’s addressable audience As a general principle, you cannot optimize for both accuracy and reach in a Lookalike Model. There are inherent tradeoffs between the two. * Optimize for **reach** for targeting users in earlier stages of your funnel * Reach more users, while being less precise * Optimize for **accuracy** for targeting users in later stages of your funnel * Be more precise, but reach less users ![accuracy-reach-graph.png](https://images.contentstack.io/v3/assets/bltebc53cfaf0dd6403/blt71c82eed35bbc7b9/686bec9d167482a55e1b16f9/accuracy-reach-graph.png) When balancing the trade off between accuracy and reach, consider the sum of accuracy and reach to determine a model’s fitness to be used: good models have an accuracy and reach sum around 10, excellent models will sum to more than 10, fair/poor models will sum to less than 10. Refer to the [Accuracy vs. Reach](https://learn.lytics.com/documentation/product/features/laboratory/lookalike-models/accuracy-vs-reach) documentation **Interpreting your model's results** In the rest of the Model Summary, you can see how the source and target audiences overlap, the important features that contribute to the model's predictions, what audiences are using the model, and any diagnostic and troubleshooting messages. ![model-summary.png](https://images.contentstack.io/v3/assets/bltebc53cfaf0dd6403/blt33a305dda3cad71c/686bec9d588d4685e5838125/model-summary.png) **Model predictions** Displays the size of your source and target audiences and charts the predictions for those audiences. The shape of the graph is most important, which indicates the amount of overlap between your source and target audiences. **Model features importance** Indicates which information the model is using to make predictions. More specifically, the chart lists the relative importance of features among users likely to convert from the source to the target audience. **Model usage** Displays a list of audiences that are utilizing this model for targeting and a button to create a new Predictive Audience with this model. **Diagnostics & troubleshooting** Provides messages pertaining to the performance and status of your Lookalike Model. These messages are categorized as warnings, errors, or information, and may include suggestions on how to improve your models. Refer the [Model Summary](https://learn.lytics.com/documentation/product/features/laboratory/lookalike-models/model-summary) documentation **Check your understanding of the tradeoffs between accuracy and reach.** High reach Better for earlier stages of your funnel Low accuracy and low reach Better for later stages of your funnel High accuracy Not a good model, shouldn't be used for predictions[](https://learn.lytics.com/documentation/product/features/laboratory/lookalike-models/accuracy-vs-reach) ## Targeting Predictive Audiences ### Creating Predictive Audiences Marketing campaigns run with **Predictive Audiences** from a healthy **Lookalike Model** are very likely to result in higher conversion rates. Even better, Lytics makes it super easy to create Predictive Audiences with the click of a button. Really, it's that easy! **Ok, show me how to create a Predictive Audience!** Once a Lookalike Model is built and users are scored, follow these steps to make a Predictive Audience. 1. Find the model of interest and go to its **Summary** page. 2. Scroll to the **Model Usage** section. 3. Click **Create Predictive Audience**. 4. Now in the Audience Builder, you'll see the model predictions pre-populated as a user field named \`segment\_prediction\`. 5. You can adjust the prediction decision threshold or add rules to further refine the audience. ![model\_usage.png](https://images.contentstack.io/v3/assets/bltebc53cfaf0dd6403/blt41dfebed02ce28ed/686bf67219cab28ee2cecd4b/model_usage.png) Refer to the [Create Predictive Audiences](https://learn.lytics.com/documentation/product/features/laboratory/lookalike-models/model-builder#create-predictive-audiences) documentation. ### Selecting the right Source and Target Building effective Lookalike Models that power your Predictive Audiences is an iterative process.  Some experimentation is to be expected before you find the right audiences and configuration options for your use case, which ties into the idea behind the "Lytics Laboratory."  #### Selecting the Right Source and Target Audience One of the most important factors for Lookalike Model performance is your selection of the source and target audience.  ![Lytics\_Lookalike\_Models\_audience\_selection\_diagram.png](https://images.contentstack.io/v3/assets/bltebc53cfaf0dd6403/blt69987467fd5b4f68/686c0d29f1c7c242864ecc5f/Lytics_Lookalike_Models_audience_selection_diagram.png) Check out the documentation linked below for examples of how to select the right audiences for your use case. Refer to the [Selecting the right audiences](https://learn.lytics.com/documentation/product/features/laboratory/lookalike-models/interpreting-model-messages#selecting-the-right-source-and-target-audience) documentation. **Match the audience type to the example use case.** Adjacent Audience unknown audience → audience with email Divergent Audience unknown audience → audience with newsletter signup and single purchase Overlapping Audience unknown audience → audience with no purchases ### Iterating your Predictive Audiences Most Predictive Audiences are built by identifying users in the source audience who have a high model score, indicating they are likely to convert to the target audience.  This "high score" is called the Decision Threshold. #### Adjusting the Decision Threshold The decision threshold is a score from 0-1 * 0= very unlikely to convert * 1= very likely convert  * In most cases, the decision threshold is set to 0.5 To adjust the **reach** of the audience you are building, you might consider using a different decision threshold. * Lower decision threshold - **reach more users** in the source audience. * Higher decision threshold - **be more accurate but reach less users** in the source audience. Refer to the [Decision Threshold](https://learn.lytics.com/documentation/product/features/laboratory/lookalike-models/interpreting-model-messages#adjusting-the-decision-threshold) documentation **How do I pick the right threshold?** You may see diagnostic messages suggesting you create audiences using a different decision threshold. But keep in mind any relevant domain knowledge (which the model doesn't have) when choosing the decision threshold. **If you want to expand the reach of your Predictive Audience, how should you adjust the decision threshold?** A. Raise the decision threshold B. Keep it the same but pick a larger source audience C. Lower the decision threshold Answer: C #### My model is unhealthy. What can I do? See the [Improving Unhealthy Models](https://learn.lytics.com/documentation/product/features/laboratory/lookalike-models/interpreting-model-messages#improving-unhealthy-models) section for examples of why your model might not be performing well and suggested next steps. **What is the most common reason a model is considered unhealthy?** A. Auto Tune wasn't selected B. Source and target audiences were divergent or overlapping C. Decision threshold was too high D. Not enough signal in the underlying data Answer: B **If you are trying to predict which users are likely to make a purchase, which of the following data sources are required to build an accurate Lookalike Model?** A. Email data B. Demographic data C. Subscription data D. Purchase data Answer: D ## Next Steps ### More Resources To continue learning, we recommend you check out the following resources: * [Lookalike Models documentation](https://learn.lytics.com/documentation/product/features/laboratory/lookalike-models/overview) * [Leveraging Lookalike Models & Predictive Audiences](https://learn.lytics.com/use-cases/leverage-lookalike-models-and-predictive-audiences) * Learn about 6 different Use Cases across the customer lifecycle Just as you iterate Lookalike Models, the Lytics Team wants to keep iterating our training guides. #### Key takeaways - Connect **Lookalike Models** back to your stack configuration before moving to the next module. - Capture one concrete artifact (screenshot, Postman call, or code snippet) that proves the step works in your environment. - Re-read the delivery versus management boundary for anything you changed in the entry model. ### Lesson 05 — Lytics Query Language #### Video details #### At a glance - **Title:** Lytics Query Language - Learn LQL Basics - **Duration:** 14m 53s - **Media link:** https://cdn.jwplayer.com/previews/g3xlbE7o - **Publish date (unix):** 1751912463 #### Streaming renditions - application/vnd.apple.mpegurl - audio/mp4 · AAC Audio · 113475 kbps - video/mp4 · 180p · 200p · 154868 kbps - video/mp4 · 270p · 300p · 189302 kbps - video/mp4 · 360p · 400p · 217965 kbps - video/mp4 · 406p · 450p · 233756 kbps - video/mp4 · 540p · 600p · 296695 kbps #### Timed text tracks (delivery) - **thumbnails:** `https://cdn.jwplayer.com/strips/g3xlbE7o-120.vtt` #### Transcript Hello and welcome to the Lytx training session LQL 101. On today's agenda we will cover what LQL is, where you can find it in the UI. I'll show you the default LQL that comes out of the box in any account. We'll cover some basic syntax explanation so that people can read LQL at a beginner level. I'll also point out where our public documentation lives for anyone to view. In addition we'll cover why it's important to go back and read our public documentation to learn more about LQL. There are many ways to transform an incoming field and how you choose to transform that field may impact your experience in the audience builder. So what is LQL? This is the long answer which I won't read aloud but it's here and also available on our public documentation. The short answer for what LQL is is that LQL transforms our incoming data. LQL enriches user profiles by structuring incoming data with unique identifiers. The result of this is the ability to build educated audiences in Lytx. So in general this is the high-level process of how things are happening. An event will come in such as page click or logs in and it will go to our cloud storage. It'll also go through our LQL where that data is transformed in meaningful ways to enrich user profiles. So here's a basic skeleton outline of what a LQL file looks like. At the very top we have we typically put in the file name and some important descriptions about the file. An important note is that this will appear in the Lytx UI. Next we have our select statement which is telling us to select incoming fields which you'll find on the far left side as a Lytx slug which is just the normalized version of the incoming field names for Lytx to process those fields. Next we have the option to use conditional statements and then we have short descriptions. This is the user-friendly name that you'll find in the Lytx UI. So instead of having first name as camel case, first name as snake case, we'll just have in title case first name. Next is the long description. This is completely optional and it's only found in a very specific place in the UI. At the end of the LQL files we have our unique identifiers, our by fields. We separate these from the other fields just for easy identification. Essentially all the fields above the by fields are able to be stored onto the user profile via the unique identifiers. So first names will come in with email UID or account ID and any other field from the above will have to come in with either email UID or account ID. At the very bottom of the LQL file we have the statements from, into, by, and alias. From denotes which stream name that the incoming data will come from. So if it's click stream data or data from the web layer etc that'll be denoted here. Into user. User is just the table that we're storing the fields on. It's the most popular table. We also have a content table but we won't get into that today. By is the by fields that we're using to store those incoming fields on. Again email UID, account ID. So the incoming fields will be stored onto the user profile again via email UID or account ID. The final statement is alias and this is just the unique file name. Unique is important here. Every LQL file name must be unique. So this is our demo account. If you go to the data tab and then the sub tab queries, you'll find all the LQL for the account. Starting from Linux content all the way user web path fora. These eight files right here come out of the box. Next to the LQL file names we'll see which table. User, content, etc. Those fields are going to be stored upon. On the far right hand side this is where we see the titles of the LQL files. So this top section here where I mentioned we typically store the file name and or some descriptions and where it appears in the Linux UI. That appears right here. So let's click into an LQL file. We'll start with user web default. This is user web default data out of the box. So we'll store things like the anonymous ID, first name, last name, city, state, country, company, city, URLs, email list, email sent, email clicks, etc. It's coming to the default stream into the user table via these by fields. UIDs, email, or phone number. And the name of the LQL file is user web default. These are the names you see in the audience builder and elsewhere in the UI. So to timestamp, sometimes stamp field, is going to be stored as time of last visit. Something I didn't mention earlier is that we have data types as well. We call them kinds, kind date, kind string, int, maps, many different data types we can use to transform your data from some subscription data, from some subscription status, to taking that subscription and mapping it to a subscription status. Now we'll take a look into some public documentation. If I search for Lix, go to our website, knowledge center, product documentation. I can search by LQL in here. And here's our LQL documentation for anyone to view. Scrolling through, we have some LQL examples, standard syntax descriptions, select from, into, where, by, alias, in case you were to forget. Here are all the functions available to transform your data from aggregate functions to logical functions, string functions, hash and encoding functions, and much much more. If you continue to scroll down, we'll eventually reach the data types we have, or kinds. Anything from an int, date, arrays, maps, many ways to take an incoming field to then transform it to enrich profiles. So now we'll take a look into the UI, and I'll show you how choosing specific functions or data types will change what you'll experience in the audience builder. So we're back at the queries tab, and before I go to the audience builder, we'll just click on this query here. So this is considered a custom LQL file based on email events. I'll open the new tab to the audience builder. Select create a new audience. In the custom rules tab here, we'll find all the fields from the LQL. Here I can choose a source. Custom email is what we just took a look at, and here are all the short descriptions from that LQL file. These. Going back to the LQL file, let's take a look at type date. So here we get this histogram where we can see on October 31st, 655 last email open dates came in, 587 on November 30th, and so on. We also have options to pick a relative date or a specific date. All of this is possible because the LQL is denoted type date. Now let's take a look at what happens with a string value in the audience builder, such as email address. So now we have the options to take a string and say I want to find a string that's exactly equal to something, contains some sort of substring, contains one string or another. If that string exists, equals one of, and there's also this be like button as well. So if that field simply exists, there's 10,000 users with it. In the case of email, we could say contains some email domain. In this case, approximately 3,000 users have an at demo in their email address. Now let's try to find a couple more different data types. I came back to the default web data LQL to find another data type for us to take a look at in the audience builder. Let's take a look at this one. We're mapping an email list to email sent. Email list is type string. Email sent, number of emails sent, is type int. I can find this field in the audience builder by searching for the short description. Now we have a different view. These are the different strings appearing in this map. Let's go to tech news. We can say we want tech news to be greater than two times, and this is our audience. We could say be less than 10, and now this is our audience. So we were able to select a string in that previous window and then select by the integers of how many email sents is stored on the user profile. There are many other ways to transform the audience builder based on the data type and functions that we're using along those incoming field names. Again, there is a lot to learn about LQL and this is meant to be an intro to just get you started. I highly recommend coming back to this page and reviewing all the different functions and data types and start thinking about different ways you can transform your incoming fields and enrich user profiles to meet your use cases. Thank you for watching. See you next time. Goodbye. #### Subtitles (WebVTT) ```webvtt WEBVTT 1 00:00:00.000 --> 00:00:08.400 Hello and welcome to the Lytx training session LQL 101. On today's agenda we 2 00:00:08.400 --> 00:00:14.520 will cover what LQL is, where you can find it in the UI. I'll show you the 3 00:00:14.520 --> 00:00:19.880 default LQL that comes out of the box in any account. We'll cover some basic 4 00:00:19.880 --> 00:00:25.680 syntax explanation so that people can read LQL at a 5 00:00:25.680 --> 00:00:31.000 beginner level. I'll also point out where our public documentation lives for 6 00:00:31.000 --> 00:00:38.240 anyone to view. In addition we'll cover why it's important to go back and read 7 00:00:38.240 --> 00:00:43.000 our public documentation to learn more about LQL. There are many ways to 8 00:00:43.000 --> 00:00:48.160 transform an incoming field and how you choose to transform that field may 9 00:00:48.160 --> 00:00:59.160 impact your experience in the audience builder. So what is LQL? This is the long 10 00:00:59.160 --> 00:01:04.560 answer which I won't read aloud but it's here and also available on our public 11 00:01:04.560 --> 00:01:14.080 documentation. The short answer for what LQL is is that LQL transforms our 12 00:01:14.080 --> 00:01:20.800 incoming data. LQL enriches user profiles by structuring incoming data 13 00:01:20.800 --> 00:01:27.720 with unique identifiers. The result of this is the ability to build educated 14 00:01:27.720 --> 00:01:37.040 audiences in Lytx. So in general this is the high-level process of how things are 15 00:01:37.040 --> 00:01:44.920 happening. An event will come in such as page click or logs in and it will go to 16 00:01:44.920 --> 00:01:52.600 our cloud storage. It'll also go through our LQL where that data is transformed 17 00:01:52.600 --> 00:02:02.960 in meaningful ways to enrich user profiles. So here's a basic skeleton 18 00:02:02.960 --> 00:02:11.920 outline of what a LQL file looks like. At the very top we have we typically put 19 00:02:11.920 --> 00:02:16.800 in the file name and some important descriptions about the file. An important 20 00:02:16.800 --> 00:02:23.720 note is that this will appear in the Lytx UI. Next we have our select 21 00:02:23.720 --> 00:02:28.280 statement which is telling us to select incoming fields which you'll find on the 22 00:02:28.280 --> 00:02:35.960 far left side as a Lytx slug which is just the normalized version of the 23 00:02:35.960 --> 00:02:45.080 incoming field names for Lytx to process those fields. Next we have the option to 24 00:02:45.080 --> 00:02:51.320 use conditional statements and then we have short descriptions. This is the 25 00:02:51.320 --> 00:02:56.600 user-friendly name that you'll find in the Lytx UI. So instead of having first 26 00:02:56.720 --> 00:03:03.760 name as camel case, first name as snake case, we'll just have in title case first 27 00:03:03.760 --> 00:03:10.920 name. Next is the long description. This is completely optional and it's only 28 00:03:10.920 --> 00:03:17.160 found in a very specific place in the UI. At the end of the LQL files we have our 29 00:03:17.160 --> 00:03:22.160 unique identifiers, our by fields. We separate these from the other fields 30 00:03:22.160 --> 00:03:26.880 just for easy identification. Essentially all the fields above the 31 00:03:26.880 --> 00:03:32.640 by fields are able to be stored onto the user profile via the unique identifiers. 32 00:03:32.640 --> 00:03:39.440 So first names will come in with email UID or account ID and any other field 33 00:03:39.440 --> 00:03:48.320 from the above will have to come in with either email UID or account ID. At the 34 00:03:48.320 --> 00:03:55.920 very bottom of the LQL file we have the statements from, into, by, and alias. From 35 00:03:55.920 --> 00:04:03.040 denotes which stream name that the incoming data will come from. So if it's 36 00:04:03.040 --> 00:04:13.160 click stream data or data from the web layer etc that'll be denoted here. Into 37 00:04:13.280 --> 00:04:19.600 user. User is just the table that we're storing the fields on. It's the most 38 00:04:19.600 --> 00:04:24.680 popular table. We also have a content table but we won't get into that today. 39 00:04:24.680 --> 00:04:33.120 By is the by fields that we're using to store those incoming fields on. Again 40 00:04:33.120 --> 00:04:39.640 email UID, account ID. So the incoming fields will be stored onto the user 41 00:04:39.760 --> 00:04:47.960 profile again via email UID or account ID. The final statement is alias and this 42 00:04:47.960 --> 00:04:53.720 is just the unique file name. Unique is important here. Every LQL file name must 43 00:04:53.720 --> 00:05:05.880 be unique. So this is our demo account. If you go to the data tab and then the 44 00:05:05.880 --> 00:05:15.400 sub tab queries, you'll find all the LQL for the account. Starting from Linux 45 00:05:15.400 --> 00:05:21.480 content all the way user web path fora. These eight files right here come out of 46 00:05:21.480 --> 00:05:30.520 the box. Next to the LQL file names we'll see which table. User, content, etc. Those 47 00:05:30.560 --> 00:05:36.840 fields are going to be stored upon. On the far right hand side this is where we 48 00:05:36.840 --> 00:05:45.000 see the titles of the LQL files. So this top section here where I mentioned we 49 00:05:45.000 --> 00:05:50.280 typically store the file name and or some descriptions and where it appears 50 00:05:50.280 --> 00:05:59.240 in the Linux UI. That appears right here. So let's click into an LQL file. We'll 51 00:05:59.240 --> 00:06:07.120 start with user web default. This is user web default data out of the box. So 52 00:06:07.120 --> 00:06:16.280 we'll store things like the anonymous ID, first name, last name, city, state, country, 53 00:06:16.280 --> 00:06:28.960 company, city, URLs, email list, email sent, email clicks, etc. It's coming to the 54 00:06:28.960 --> 00:06:37.600 default stream into the user table via these by fields. UIDs, email, or phone 55 00:06:37.600 --> 00:06:48.560 number. And the name of the LQL file is user web default. These are the names 56 00:06:48.560 --> 00:06:54.640 you see in the audience builder and elsewhere in the UI. So to timestamp, 57 00:06:54.640 --> 00:07:00.320 sometimes stamp field, is going to be stored as time of last visit. Something I 58 00:07:00.320 --> 00:07:06.000 didn't mention earlier is that we have data types as well. We call them kinds, 59 00:07:06.000 --> 00:07:16.400 kind date, kind string, int, maps, many different data types we can use to 60 00:07:16.400 --> 00:07:24.040 transform your data from some subscription data, from some subscription 61 00:07:24.040 --> 00:07:34.800 status, to taking that subscription and mapping it to a subscription status. Now 62 00:07:34.800 --> 00:07:43.920 we'll take a look into some public documentation. If I search for Lix, go to 63 00:07:43.920 --> 00:07:55.720 our website, knowledge center, product documentation. I can search by LQL in 64 00:07:55.720 --> 00:08:11.240 here. And here's our LQL documentation for anyone to view. Scrolling through, we 65 00:08:11.240 --> 00:08:20.560 have some LQL examples, standard syntax descriptions, select from, into, where, by, 66 00:08:20.560 --> 00:08:27.880 alias, in case you were to forget. Here are all the functions available to 67 00:08:27.880 --> 00:08:34.120 transform your data from aggregate functions to logical functions, string 68 00:08:34.200 --> 00:08:44.680 functions, hash and encoding functions, and much much more. If you continue to 69 00:08:44.680 --> 00:08:52.640 scroll down, we'll eventually reach the data types we have, or kinds. Anything 70 00:08:52.640 --> 00:09:02.080 from an int, date, arrays, maps, many ways to take an incoming field to then 71 00:09:02.080 --> 00:09:09.360 transform it to enrich profiles. So now we'll take a look into the UI, and I'll 72 00:09:09.360 --> 00:09:17.720 show you how choosing specific functions or data types will change what you'll 73 00:09:17.720 --> 00:09:24.400 experience in the audience builder. So we're back at the queries tab, and before 74 00:09:24.400 --> 00:09:29.840 I go to the audience builder, we'll just click on this query here. So this is 75 00:09:29.840 --> 00:09:39.200 considered a custom LQL file based on email events. I'll open the 76 00:09:39.200 --> 00:09:43.040 new tab to the audience builder. 77 00:09:50.680 --> 00:09:59.160 Select create a new audience. In the custom rules tab here, we'll find all the 78 00:09:59.160 --> 00:10:08.320 fields from the LQL. Here I can choose a source. Custom email is what we just took 79 00:10:08.320 --> 00:10:15.960 a look at, and here are all the short descriptions from that LQL file. 80 00:10:16.200 --> 00:10:19.200 These. 81 00:10:20.040 --> 00:10:32.360 Going back to the LQL file, let's take a look at type date. 82 00:10:36.680 --> 00:10:45.280 So here we get this histogram where we can see on October 31st, 655 last email 83 00:10:45.320 --> 00:10:57.440 open dates came in, 587 on November 30th, and so on. We also have options to pick a 84 00:10:57.440 --> 00:11:01.480 relative date or a specific date. 85 00:11:05.920 --> 00:11:14.160 All of this is possible because the LQL is denoted type date. Now let's take a 86 00:11:14.160 --> 00:11:24.840 look at what happens with a string value in the audience builder, such as 87 00:11:24.840 --> 00:11:33.800 email address. So now we have the options to take a string and say I want to find 88 00:11:33.800 --> 00:11:38.320 a string that's exactly equal to something, contains some sort of 89 00:11:38.320 --> 00:11:46.520 substring, contains one string or another. If that string exists, equals one 90 00:11:46.520 --> 00:11:51.680 of, and there's also this be like button as well. 91 00:11:55.920 --> 00:12:05.520 So if that field simply exists, there's 10,000 users with it. In the case of 92 00:12:05.520 --> 00:12:15.160 email, we could say contains some email domain. In this case, approximately 3,000 93 00:12:15.160 --> 00:12:25.880 users have an at demo in their email address. Now let's try to find a couple 94 00:12:25.880 --> 00:12:32.280 more different data types. I came back to the default web data LQL to find 95 00:12:32.280 --> 00:12:39.160 another data type for us to take a look at in the audience builder. Let's take a 96 00:12:39.160 --> 00:12:49.440 look at this one. We're mapping an email list to email sent. Email list is type 97 00:12:49.440 --> 00:12:56.760 string. Email sent, number of emails sent, is type int. 98 00:13:02.280 --> 00:13:13.880 I can find this field in the audience builder by searching for the short 99 00:13:13.880 --> 00:13:16.480 description. 100 00:13:23.480 --> 00:13:29.080 Now we have a different view. These are the different strings appearing in this 101 00:13:29.080 --> 00:13:41.360 map. Let's go to tech news. We can say we want tech news to be greater than two 102 00:13:41.360 --> 00:13:51.640 times, and this is our audience. We could say be less than 10, and now this is our 103 00:13:51.640 --> 00:14:01.200 audience. So we were able to select a string in that previous window and then 104 00:14:01.200 --> 00:14:09.280 select by the integers of how many email sents is stored on the user profile. There 105 00:14:09.280 --> 00:14:13.960 are many other ways to transform the audience builder based on the data type 106 00:14:14.080 --> 00:14:23.320 and functions that we're using along those incoming field names. Again, there 107 00:14:23.320 --> 00:14:28.240 is a lot to learn about LQL and this is meant to be an intro to just get you 108 00:14:28.240 --> 00:14:33.480 started. I highly recommend coming back to this page and reviewing all the 109 00:14:33.480 --> 00:14:38.400 different functions and data types and start thinking about different ways you 110 00:14:38.400 --> 00:14:46.520 can transform your incoming fields and enrich user profiles to meet your use 111 00:14:46.520 --> 00:14:54.400 cases. Thank you for watching. See you next time. Goodbye. ``` ```transcript [00:00] Hello and welcome to the Lytx training session LQL 101. On today's agenda we [00:08] will cover what LQL is, where you can find it in the UI. I'll show you the [00:14] default LQL that comes out of the box in any account. We'll cover some basic [00:19] syntax explanation so that people can read LQL at a [00:25] beginner level. I'll also point out where our public documentation lives for [00:31] anyone to view. In addition we'll cover why it's important to go back and read [00:38] our public documentation to learn more about LQL. There are many ways to [00:43] transform an incoming field and how you choose to transform that field may [00:48] impact your experience in the audience builder. So what is LQL? This is the long [00:59] answer which I won't read aloud but it's here and also available on our public [01:04] documentation. The short answer for what LQL is is that LQL transforms our [01:14] incoming data. LQL enriches user profiles by structuring incoming data [01:20] with unique identifiers. The result of this is the ability to build educated [01:27] audiences in Lytx. So in general this is the high-level process of how things are [01:37] happening. An event will come in such as page click or logs in and it will go to [01:44] our cloud storage. It'll also go through our LQL where that data is transformed [01:52] in meaningful ways to enrich user profiles. So here's a basic skeleton [02:02] outline of what a LQL file looks like. At the very top we have we typically put [02:11] in the file name and some important descriptions about the file. An important [02:16] note is that this will appear in the Lytx UI. Next we have our select [02:23] statement which is telling us to select incoming fields which you'll find on the [02:28] far left side as a Lytx slug which is just the normalized version of the [02:35] incoming field names for Lytx to process those fields. Next we have the option to [02:45] use conditional statements and then we have short descriptions. This is the [02:51] user-friendly name that you'll find in the Lytx UI. So instead of having first [02:56] name as camel case, first name as snake case, we'll just have in title case first [03:03] name. Next is the long description. This is completely optional and it's only [03:10] found in a very specific place in the UI. At the end of the LQL files we have our [03:17] unique identifiers, our by fields. We separate these from the other fields [03:22] just for easy identification. Essentially all the fields above the [03:26] by fields are able to be stored onto the user profile via the unique identifiers. [03:32] So first names will come in with email UID or account ID and any other field [03:39] from the above will have to come in with either email UID or account ID. At the [03:48] very bottom of the LQL file we have the statements from, into, by, and alias. From [03:55] denotes which stream name that the incoming data will come from. So if it's [04:03] click stream data or data from the web layer etc that'll be denoted here. Into [04:13] user. User is just the table that we're storing the fields on. It's the most [04:19] popular table. We also have a content table but we won't get into that today. [04:24] By is the by fields that we're using to store those incoming fields on. Again [04:33] email UID, account ID. So the incoming fields will be stored onto the user [04:39] profile again via email UID or account ID. The final statement is alias and this [04:47] is just the unique file name. Unique is important here. Every LQL file name must [04:53] be unique. So this is our demo account. If you go to the data tab and then the [05:05] sub tab queries, you'll find all the LQL for the account. Starting from Linux [05:15] content all the way user web path fora. These eight files right here come out of [05:21] the box. Next to the LQL file names we'll see which table. User, content, etc. Those [05:30] fields are going to be stored upon. On the far right hand side this is where we [05:36] see the titles of the LQL files. So this top section here where I mentioned we [05:45] typically store the file name and or some descriptions and where it appears [05:50] in the Linux UI. That appears right here. So let's click into an LQL file. We'll [05:59] start with user web default. This is user web default data out of the box. So [06:07] we'll store things like the anonymous ID, first name, last name, city, state, country, [06:16] company, city, URLs, email list, email sent, email clicks, etc. It's coming to the [06:28] default stream into the user table via these by fields. UIDs, email, or phone [06:37] number. And the name of the LQL file is user web default. These are the names [06:48] you see in the audience builder and elsewhere in the UI. So to timestamp, [06:54] sometimes stamp field, is going to be stored as time of last visit. Something I [07:00] didn't mention earlier is that we have data types as well. We call them kinds, [07:06] kind date, kind string, int, maps, many different data types we can use to [07:16] transform your data from some subscription data, from some subscription ``` #### Lesson text Developer training on basic LQL syntax and how Lytics transforms incoming data into actionable user profiles. ## Intro to LQL ### Learn LQL basics **Note:** On January 10, 2023, we upgraded our UI with a new, refreshed interface. All of the underlying functionality is the same, but you will notice that things look a little different from this Academy guide. The most notable change is that the navigation menu has moved from the top of the app to the left side. We appreciate your patience as we work on updating our Academy. ### What will I learn? * What is LQL?  * Where can I view LQL files in the Lytics UI? * What is the basic syntax? * What are the different ways to transform incoming data fields? Watch the "Lytics Query Language - Learn LQL Basics" training video below (15 mins) for an overview of **Lytics Query Language** (LQL). We'll cover the basics of how LQL transforms data, the difference between default and custom LQL, and how key identifiers merge user profiles. **Keep in Mind** - The end result of LQL is the ability to build educated user profiles and audiences in Lytics. ### Knowledge Check Before you dive into writing your own LQL files, check your understanding of the key concepts covered. **Copies of raw events are stored prior to LQL processing.** A. True B. False Answer: A **Which name on the LQL file is displayed and searchable in the Lytics Audience Builder?** A. SELECT Field B. Short Description C. BY Field D. Long Description Answer: B **Each LQL file must have a unique name set as the alias.** A. True B. False Answer: A **How are data types (i.e. string, integer, map, etc.) referenced in LQL files?** A. SORT B. TYPE C. KIND Answer: C **KINDs will impact the data visualizations in the Audience Builder.** A. True B. False Answer: A **How many out-of-the-box LQL files are there?** A. 4 B. 3 C. 8 D. 0 Answer: C ### More Resources ### Documentation * [Lytics Query Language](https://learn.lytics.com/documentation/product/features/data-onboarding-and-management/lytics-query-language) - more detailed explanation and syntax examples * [LQL & Data Import Basics](https://learn.lytics.com/documentation/developer/academy/lql-and-data-import-basics) - introductory API documentation #### Key takeaways - Connect **Lytics Query Language** back to your stack configuration before moving to the next module. - Capture one concrete artifact (screenshot, Postman call, or code snippet) that proves the step works in your environment. - Re-read the delivery versus management boundary for anything you changed in the entry model. ## Resources & references | Page | Companion Markdown | | --- | --- | | /courses/beyond-the-basics/lytics-data-flow | /academy/md/courses/beyond-the-basics/lytics-data-flow.md | | /courses/beyond-the-basics/lytics-javascript-tag | /academy/md/courses/beyond-the-basics/lytics-javascript-tag.md | | /courses/beyond-the-basics/jobs-and-authorizations | /academy/md/courses/beyond-the-basics/jobs-and-authorizations.md | | /courses/beyond-the-basics/lookalike-models | /academy/md/courses/beyond-the-basics/lookalike-models.md | | /courses/beyond-the-basics/lytics-query-language | /academy/md/courses/beyond-the-basics/lytics-query-language.md | ## Supplement for indexing ### Content summary Dive into more advanced topics to optimize your use of Lytics and unlock new activations. Dive into more advanced topics to optimize your use of Lytics and unlock new activations. ### Retrieval tags - lytics - beyond-the-basics - Data - Flow - JavaScript - Tag - Jobs - and - Authorizations - Lookalike - Models - Query - Language - beyond-the-basics course ### Indexing notes Chunk at each "### Lesson NN — Title" heading; copy lesson_id and topics from the preceding HTML comment into chunk metadata for RAG filters. Course slug: beyond-the-basics. Union of lesson topic tokens: Lytics, Data, Flow, JavaScript, Tag, Jobs, and, Authorizations, Lookalike, Models, Query, Language. Do not embed or retrieve LMS-only quiz items or mastery exam answer keys from this export. ### Asset references | Label | URL | | --- | --- | | Video thumbnail: Lytics Data Flow | `https://cdn.jwplayer.com/v2/media/FQmwlMCw/poster.jpg?width=720` | | Video thumbnail: Lytics JavaScript Tag | `https://cdn.jwplayer.com/v2/media/d4p068ys/poster.jpg?width=720` | | Video thumbnail: Jobs and Authorizations | `https://cdn.jwplayer.com/v2/media/igYBULhs/poster.jpg?width=720` | | Jobs - Authorizations Nav.png | `https://images.contentstack.io/v3/assets/bltebc53cfaf0dd6403/bltd91798a2a2383c49/686b9df7b6f4b47e96082834/Jobs_-_Authorizations_Nav.png` | | mailchimp-job-type.png | `https://images.contentstack.io/v3/assets/bltebc53cfaf0dd6403/bltf4d8083e08623956/686b9e903687732f96eec3b9/mailchimp-job-type.png` | | aws-job-types.png | `https://images.contentstack.io/v3/assets/bltebc53cfaf0dd6403/bltc4e076777850d748/686b9e914b4fe965bf5c4f12/aws-job-types.png` | | jobs-wizard.png | `https://images.contentstack.io/v3/assets/bltebc53cfaf0dd6403/blt51edc03d0040b5f7/686b9fb54b4fe98df75c4f23/jobs-wizard.png` | | lytics-job-summary-example.png | `https://images.contentstack.io/v3/assets/bltebc53cfaf0dd6403/bltb0631065fb8f196c/686ba0fff1c7c2d1154ec68f/lytics-job-summary-example.png` | | Lytics\_authorizations\_dashboard.png | `https://images.contentstack.io/v3/assets/bltebc53cfaf0dd6403/bltf35e98764b7c09c4/686ba21828e271aa3aa3b482/Lytics_authorizations_dashboard.png` | | auth-creation-in-job-wizard.png | `https://images.contentstack.io/v3/assets/bltebc53cfaf0dd6403/blt6d58d1161a4d4755/686ba21802334227e192e5ee/auth-creation-in-job-wizard.png` | | authorizations-summary-example.png | `https://images.contentstack.io/v3/assets/bltebc53cfaf0dd6403/blt2d34fee5ed4eaa19/686ba3ef321163ea84005610/authorizations-summary-example.png` | | Video thumbnail: Lookalike Models | `https://cdn.jwplayer.com/v2/media/oXPDbF1w/poster.jpg?width=720` | | Click Laboratory > UI Models.png | `https://images.contentstack.io/v3/assets/bltebc53cfaf0dd6403/bltb24a9f4465cfcc2c/686bd5514e325563d15687ae/Click_Laboratory_UI_Models.png` | | click-create-new-model.png | `https://images.contentstack.io/v3/assets/bltebc53cfaf0dd6403/blt49a93af1fddafc74/686be0df47548f406f569af6/click-create-new-model.png` | | lytics-lookalike-model-summary.png | `https://images.contentstack.io/v3/assets/bltebc53cfaf0dd6403/blt9aae40f62f0e263e/686be6318f61ad7bdddc66d6/lytics-lookalike-model-summary.png` | | accuracy-reach.png | `https://images.contentstack.io/v3/assets/bltebc53cfaf0dd6403/blt62f4dd47b1e7fbb7/686bec9c77a15576584c156d/accuracy-reach.png` | | accuracy-reach-graph.png | `https://images.contentstack.io/v3/assets/bltebc53cfaf0dd6403/blt71c82eed35bbc7b9/686bec9d167482a55e1b16f9/accuracy-reach-graph.png` | | model-summary.png | `https://images.contentstack.io/v3/assets/bltebc53cfaf0dd6403/blt33a305dda3cad71c/686bec9d588d4685e5838125/model-summary.png` | | model\_usage.png | `https://images.contentstack.io/v3/assets/bltebc53cfaf0dd6403/blt41dfebed02ce28ed/686bf67219cab28ee2cecd4b/model_usage.png` | | Lytics\_Lookalike\_Models\_audience\_selection\_diagram.png | `https://images.contentstack.io/v3/assets/bltebc53cfaf0dd6403/blt69987467fd5b4f68/686c0d29f1c7c242864ecc5f/Lytics_Lookalike_Models_audience_selection_diagram.png` | | Video thumbnail: Lytics Query Language | `https://cdn.jwplayer.com/v2/media/g3xlbE7o/poster.jpg?width=720` | ### External links | Label | URL | | --- | --- | | Contentstack Academy home | `https://www.contentstack.com/academy/` | | Training instance setup | `https://www.contentstack.com/academy/training-instance` | | Academy playground (GitHub) | `https://github.com/contentstack/contentstack-academy-playground` | | Contentstack documentation | `https://www.contentstack.com/docs/` | | How does Lytics work? | `https://learn.lytics.com/documentation/product/features/getting-started/how-does-lytics-work` | | Onboarding Web Data | `https://learn.lytics.com/documentation/product/features/data-onboarding-and-management/onboarding-web-data` | | Integrated Marketing Tools | `https://learn.lytics.com/documentation/product/features/data-onboarding-and-management/integrated-marketing-tools` | | Data Streams | `https://learn.lytics.com/documentation/product/features/data-onboarding-and-management/data-streams` | | Data Schema | `https://learn.lytics.com/documentation/product/features/data-onboarding-and-management/schema-audit` | | User Fields | `https://learn.lytics.com/documentation/product/features/data-onboarding-and-management/user-fields` | | JavaScript Snippet | `https://app.lytics.com/connect?view=v3` | | Installation & Configuration | `https://learn.lytics.com/documentation/product/features/lytics-javascript-tag/using-version-3/installation-configuration` | | Version 3 documentation | `https://learn.lytics.com/documentation/product/features/lytics-javascript-tag/version-3-improvements` | | Working with Tag Managers | `https://learn.lytics.com/documentation/product/features/lytics-javascript-tag/working-with-tag-managers` | | Lytics JavaScript Tag introduction | `https://learn.lytics.com/documentation/product/features/lytics-javascript-tag/introduction` | | Collecting Data with V3 Tag | `https://learn.lytics.com/documentation/product/features/lytics-javascript-tag/using-version-3/collecting-data` | | Troubleshooting: Verifying Data is sent to Lytics | `https://learn.lytics.com/documentation/product/features/lytics-javascript-tag/troubleshooting#verify-data-is-being-sent-to-lytics` | | Installing Lytics Image Pixel | `https://learn.lytics.com/documentation/product/features/data-onboarding-and-management/onboarding-web-data#lytics-image-pixel` | | Jobs - Authorizations Nav.png | `https://images.contentstack.io/v3/assets/bltebc53cfaf0dd6403/bltd91798a2a2383c49/686b9df7b6f4b47e96082834/Jobs_-_Authorizations_Nav.png` | | mailchimp-job-type.png | `https://images.contentstack.io/v3/assets/bltebc53cfaf0dd6403/bltf4d8083e08623956/686b9e903687732f96eec3b9/mailchimp-job-type.png` | | aws-job-types.png | `https://images.contentstack.io/v3/assets/bltebc53cfaf0dd6403/bltc4e076777850d748/686b9e914b4fe965bf5c4f12/aws-job-types.png` | | jobs-wizard.png | `https://images.contentstack.io/v3/assets/bltebc53cfaf0dd6403/blt51edc03d0040b5f7/686b9fb54b4fe98df75c4f23/jobs-wizard.png` | | Jobs Dashboard | `https://learn.lytics.com/documentation/product/features/data-onboarding-and-management/jobs/jobs-dashboard` | | lytics-job-summary-example.png | `https://images.contentstack.io/v3/assets/bltebc53cfaf0dd6403/bltb0631065fb8f196c/686ba0fff1c7c2d1154ec68f/lytics-job-summary-example.png` | | Job Summary | `https://learn.lytics.com/documentation/product/features/data-onboarding-and-management/jobs/job-summary` |