Adobe Target Engineering Webinar #4

Q1 : Can AI assistant create profile scripts ? Good question. Not at first, but we are looking at scenarios where an assistant would be able to create more than text or images. Some example include: small components like a pop-up, scripts or code blocks like a product card in a category page Q2 : Is the AI Assistant available for the standard edition as well as premium ? The AI Assistant will be available for both Standard and Premium provided that your company has signed the necessary agreements for using GenAI features. Q3 : A non-technical user may struggle to use a tool like Target effectively. Has the visual editor been improved to clearly indicate when a user is modifying an attribute that won’t affect classes or IDs ? For example: if I update the content of a CTA, the editor sometimes freezes the existing CSS classes, causing the loss of dynamic behaviors that depend on adding or removing classes at runtime. Has this issue been addressed ? The Visual Experience Composer (VEC) is a powerful tool, but it works by applying static DOM modifications. When an application relies on dynamic or reactive components, these static changes can be overwritten or conflict with the framework’s rendering cycle. In cases like yours, this can lead to inconsistent behavior or unexpected UI results. To ensure the personalization behaves correctly within a dynamic app, a custom code approach is often required so the changes are applied at the right time and in the right way. Adobe Consulting works with these scenarios regularly and can help design a solution that aligns with the application architecture while still supporting your personalization goals. Q4 : If there is flicker, is it better to use the pre-hiding code snippet or implement server-side architecture ? Both approaches would improve or eliminate the flicker. Server-side implementation would ensure there is no flicker since the content changes server-side. Client-side implementation with pre-hiding code snippet would ensure flicker is managed by pre-hiding the body or custom elements while personalization is loaded Q5 : Is there any way Adobe Target can track back to a visitor PII ? I have trouble with how to bucket Adobe Target into our cookie policy. Should it be a required cookie or an advertisement/performance cookie ? Does a visitor need to accept cookies for Adobe Target to work in global regions ? Acceptance of cookies should depend on your organizations consent policies and consent management implementation. Ideally Target should not use any PII in most of use cases. Q6 : Some visitors get Adobe Target experiences blocked by add blockers and/or sometimes Firefox blocks Target experiences based on their browser privacy settings. Is there a way to deliver experiences to 100% of visitors even those with an add blocker ? CNAME will help you to address this situation. Q7 : What is the equivalent event of web sdk for 'content rendering succeeded' event of at.js ? You can leverage 'Monitoring Hooks' to mimic the content rendering succeeded events. Namely the onContentRendering hook. https://experienceleague.adobe.com/en/docs/experience-platform/web-sdk/monitoring-hooks#onContentRendering Q8 : How can we define cookies when implementing some XT ? For example lightbox pop ups stuff where we always want the promo to be offer once only ? You can use profile script for such use case. Q9 : If AEM XF is updated and re-exported to Target, will it reflect in the variation or should we manually use newly imported XF again. ? Once we change the existing XF and publish to target it should reflect the changes in target and live activities offers. Q10 : When will drag and drop feature return to the UI ? The re-arrange feature has been redesigned and is currently available in the contextual menu ribbon displayed when you select an element. Using this menu, users can change the element position within the parent element, like the drag-and-drop method used in the previous UI version. Q11 : Can we allow "viewer" access to Target Activities that are created using the API ? View-only mode has been recently re-introduced in the Target UI for activities created within the UI. Expanding this mode to API sourced activates would be a good first step towards a streamlined workflow where activities are created everywhere and reviewed / approved for launch in Target. We would need to create a feature request and evaluate it against others. Q12 : Is the new UI preloading page for each device type ? The new VEC version offers users the option of selecting from a few popular screen types (Desktop, Mobile, etc) while also allowing users to chose custom screen widths while in VEC. If each device type has it's own view, once can switch editor size and then browse to the view to make it accessible in the editor. Q13 : Some features from the older VEC are missing from the new UI. Will there be any updates to address these gaps ? Most of the features available in the previous version of the VEC are also available in the new version, even though in a new form / location. We are looking forward to hearing your feedback related to missing features in the new UI. Q14 : How to move stuff around with the new UI ? The re-arrange feature is now available in the contextual blue menu displayed when selecting an element. The Properties panel also allows users to define or change the element position. Q15 : I would love to know more about how to use the Components panel in VEC As compared to the previous VEC editor, which allowed users to insert Target specific content (Offers, Recommendations, Fragments), the new version allows users to insert a few more standard HTML components like: Dividers, Text, Heading, Paragraphs. These add specific HTML elements in the page, for testing or personalization. Q16 : What changes or improvements is new UI bringing to Experience Fragments ? Experience and Content Fragments integration will work like in the old UI version with minor changes, except the visual ones. One example could be the removal of AEM deep-links in the Quick info section of these content types, as Target users do not usually have access to AEM. Q17 : Target UI does not support large screens We would need a bit more context on this feedback. You might want to submit a feature request either through a customer care ticket or on the Adobe Target Ideas section on Experience Leage. Q18 : When are trainings going to be made available on the new UI. ? We are working internally to plan and update existing videos related to the new UI. In interim, you might check the recordings from previous webinars available on Experience League and include demos of the new UI. Q19 : Will the new UI allow users to inject code both globally and at the single view (target_view) level ? If you're referring to having the ability to apply a modifications both on Page Load and on a Specific view, this is not planned any time soon. Alternative options could be creating same modification again and applying one to page load and another to the view(s). Q20 : Would love to learn more about strategic best practices related to Adobe Target Recommendations pertaining to catalog setup (particularly category ID) for user-based criteria. User-based criteria and category affinity work best when categories are comparable and not too fine-grained or too meta like Mens or Womens. Use the same taxonomy and category naming across all environments. Isolate category level on PDP. That means, have the normal entity.categoryId list, but also include entity.catL1, entity.caL2 etc. This allows for more precise filtering to match current item Do not recommend passing department level user.categoryId on high level pages because every visitor will always have that high level category as their affinity. Start it on L2 or L3 cat pages Q21 : How are server-side implementations handled if not using Launch or other tag manager ? Pure server-side implementation is supported through custom XDM calls with a propositionFetch event type. Impression events can either be sent to a proxy server integrated with Target after the client-side impression occurs or, if WebSDK is present on the page, the impression details can be sent on the page view event, along with the sendEvent call or an additional propositionDisplay call. In cases where WebSDK is not included and there is no server-side proxy impression call, a custom client-side XDM call could be made for the proposition scopeDetails. Reference (https://github.com/adobe/alloy-samples/tree/main/target/personalization-server-side) Q22 : Integration of Target with GA4 Integration of Target with Google Analytics requires extracting Response Tokens data from qualifying experiences in Target response. Both Web SDK (alloy.js) and at.js deliver Response Tokens with activity ID, activity name, experience ID, experience name, and other useful meta data that can be pushed to GA as well as any other third-party services. Q23 : Target on mobile app Please have a look into the following documentation, it guides you in implementing Adobe Experience Cloud in mobile apps tutorial: https://experienceleague.adobe.com/en/docs/platform-learn/implement-mobile-sdk/overview Q24 : How we can solve flickering issue in SPA ? Pre-hiding style or pre-hiding code snippet still remains the best way to address a flicker in SPAs. It works on initial page load to ensure default content remains hidden. Because page load event prefetches all potential experiences to display, navigating from one SPA page to another does not usually require additional pre-hiding as long as the new SPA page renders personalized content immediately after rendering default content. Q25 : Debug profile scripts Understanding how profile scripts run is important : Every active profile script runs on every Target call (mbox/delivery) server-side, before audience and activity evaluation. Order is not guaranteed and can change over time; do not rely on script A always running before script B. Looks like JS but runs on Target servers (Rhino), not in the browser. Browser-only objects/functions are unavailable: window, document, location, navigator, console, alert, etc. Disabling a script does not erase existing values; they remain in profiles until TTL expiry or you explicitly reset them. Debugging: Use Experience Platform Debugger + Target Trace to check beforeExecutionProfileSnapshot / afterExecutionProfileSnapshot values of a profile script. Test with clean (new user) and returning profile Be mindful of hitting instruction limits (complex regex / loops) Keep scripts small and deterministic Q26 : Which in your experience is the best way to handle Server Side implementations with IT slow paced times ? 1. To bridge the gap between IT and marketing users, a personalization-first approach can help the communication between IT and marketing users. - IT should be prepared to request potential alternative personalized content when content is rendered to the screen. The most successful implementations we have seen implement a "personalization api' that integrates with Target and other third-party personalization tools. - Marketers require an interface to be able to update text, images or component layouts. - The marketing interface and the implementation (web/mobile etc.) will communicate through data-based offers, for example JSON files that provide a recipe for the app to follow. If JSON is already building the content, consider using the same JSON schema in Target offers to minimize extra customization

2. For Marketers: Integrating Target offers with AEM or another CMS can be helpful to offer marketers the ability to change offer metadata with a point and click form interface. These offers can be created and updated with AEM Experience or Content Fragments to integrate directly in Target. In other CMS environments, a similar approach can be leveraged where the content is editable via an interface and then offers will be published to the Target offer library via the Target Admin API. 3. For IT/dev: Define a global location/decisionScope schema for components in the web/mobile app, the location is an address to the element or page that should be personalized. Commonly personalized or tested elements can use the same location already in the application. New decisionScopes and locations may require an app update, therefore when defining a location schema during development, consider developing the application with locations defined throughout the app globally and not for a single use case. Leverage a component Id that will tie the location of content to the data in the CMS. Q27 : How can we personalize in a cookieless world ? Adobe closely monitors the cookieless future but today cookies are here to stay. Some browsers differ in ways how they persist cookies. When using Web SDK, the best practice includes setting an FPID server-side as the first party cookie, which seeds the same ECID, extending the lifetime of ECID. In addition, CNAME may also help with ad-blockers. Q28 : at.js vs SDK Below documentation link outlines the differences between AT.JS and WebSDK across multiple scenarios with the code examples which includes alternative of getOffer and applyOffer: https://experienceleague.adobe.com/en/docs/experience-platform/web-sdk/personalization/adobe-target/web-sdk-atjs-comparison#overview And below documentation provides a reference to help you evaluate areas of your implementation you may need to focus on during the migration process: https://experienceleague.adobe.com/en/docs/platform-learn/migrate-target-to-websdk/detailed-comparison Q29 : What are best practices in combining AEP Edge Audiences + Target Local Audience Capabilities for better personalization. Use AEP Edge API / AEP Web SDK when using combined Audiences. If all of the attributes used for the Audience rule is being already sent to one solution (AEP or Target), then it is better to setup the Audience in one location that creating a combined Audience which will make it easy to maintain it as well as debug qualification issues. Q30 : What are some best practices on how to use Experience Fragments ? Keep them small, self-contained experiences, avoid nested XFs Export XFs from AEM and manage the lifecycle from AEM, avoid manually creating/editing AEM-sourced XF offers in Target. Strip extraneous HTML and clientlibs from exported XFs so offers contain just the fragment markup (no <html>, <head>, <body>, duplicate JS/CSS), using the .nocloudconfigs.atoffer.html overlays. Use VEC for HTML XFs, Form-based Composer for HTML or JSON XF offers. Use smart naming conventions to identify XFs in Target UI. (Ex: include "env" names to differentiate AEM source environment) Don't use huge page-sized XFs for small tweaks. Regularly sync-up AEM XFs with Target to remove unused XF's from Target Q31 : Why would two pages from different experiences appear in the same experience ? For example, why is Page A registered under Experience B along with Page B ? It is possible that a single ECID can see two different experiences if the end user has multiple devices, and the Activity qualification occurs on the non-authenticated pages. For example, suppose an end user has seen Exp B on Device 1. If the same user uses Device 2, Target will initially display a random experience during qualification (e.g., Exp A). Subsequently, when the end user authenticates on Device 2, Target recognizes the end user being placed to Exp B on Device 1 and switches Device 2 to also display Exp B. This results in Analytics showing the end user viewing both Exp A and Exp B. Q32 : We have recently implemented server-side integration in our Next.js project, and since then, we've observed a decrease in tracking coverage on Google Analytics Please open a support ticket as we will need to investigate to understand what could be happening here. If you don't know how to open a ticket please contact your account team. Q33 : Will setting an activity LIVE with QA audience and allocation of 100% to variation skew traffic distribution in the actual live activity, once QA audience has been removed and traffic split has been reset ? If you use a QA audience with different traffic allocations than what will be set in production for QA purposes, it is recommended to copy and launch a new activity once ready to go live to be on the safe side. Activity QA URLs are a better alternative so that you do not need to copy and launch activities after QA-ing. Q34 : A4T and Web adk set up .. Enabling/Replacing Adobe Analytics reporting source for target activities Covered in presentation, please see recording. Q35 : Best practices in moving from at.js to web SDK. Covered in presentation, please see recording. Q36 : Do we have any live session for Recommendation set-sup how Website and Target get connected to each other. The recommendation topic has been tackled during our webinar 3, please see the link below. https://experienceleaguecommunities.adobe.com/t5/adobe-target-discussions/adobe-target-engineering-webinar-3-september-2-2025/m-p/777958#M13048 If you need more information please reach out to your account team or submit a ticket. Q37 : For a B2B company how to improve recommendations to users ? - Do not hesitate to A/B our algorithms to find the best one for you. - You can use this capability to fluidify the navigation of your users with great results - If you want to go further in the personalization do not hesitate to use firmographic data or to use IP address of your biggest customers in order to build for them a personalized landing page. If you want an expert on to help you on this activity please contact your Named CSM, TAM or Account Executive, he will direct you to the right team. Q38 : Will there be any changes to Target API in the near future ? No. We do not plan to update the Target Admin APIs in near future. Are you looking for any specific updates or new API features? If yes, please share your thoughts and ideas with your Named CSM, TAM or by submitting a ticket. Q39 : It will be great if you can do the QA without the audience, for example: when you made an experiment with multiple audiences, you must test the different variants. This would be a feature request. You can submit one either through the Experience League - Adobe Target Ideas section or by submitting a customer care ticket. Q40 : How do I receive release notifications ? To receive update notifications, please navigate to status.adobe.com and subscribe to Experience Cloud updates. Q41 : Uses cases for use AI assistant The AI Assistant for product knowledge focuses on helping users navigate and understand Adobe Target concepts. For more information read the following page: https://experienceleague.adobe.com/en/docs/target/using/introduction/assistant-ai/ai-assistant-product-knowledge Q42 : Has auto-allocation in multivariate testing been optimized ? Multi-variate testing (MVT) activities distribute traffic evenly between different experiences similar to manual A/B testing. Auto-Allocate activities are automatically distributing traffic to the winning experience but lack the ability to automatically compose experiences from a defined set of locations and offers. Q43 : Will there be the possibility to leverage AI for dynamic variant delivery (engine autonomously managing both audience allocation and content optimization) ? Auto-Target and Automated Personalization activities are automatically matching predefined experiences or offers to visitor profiles using a Random Forest ensemble. While the offer to audience matching can be automated already, in Adobe Target, the content must be pre-created by users. As a first step for content, we have started to integrate GenAI features for content creation, allowing users to generate content (and later image) variations which could be used in testing or personalization activities. Q44 : How can we integrate Offer Decisioning's evaluated offer (personalized or fallback) back into Target and Adobe Analytics ? Via A4T or otherwise ? You can use offers from Adobe Journey Optimizer into Target A/B Testing and Experience Targeting activities in Adobe Target. The integration supports only Target reporting and does not provide offer level reporting. For more information on Offer Decisioning integration, visit the feature page on Experience League: https://experienceleague.adobe.com/en/docs/target/using/integrate/ajo/offer-decision Q45 : When is the profile update feature across edges going to be live ? On an inbound channel like web or mobile, profiles are most of the times needed on a single Edge, as the visitor location does not usually change. When the visitor switches edges, Target considers it a new visitor and creates a new profile on the specific edge. Target does not have plans to replicate profiles across multiple edge servers. However, when using Adobe Read-time CDP Edge Profiles, these can be made available to multiple edges. Q46 : Allow for custom metrics as a primary metric whilst using auto allocate There aren't any near term plans to support custom metrics (from Adobe Analytics) as a primary goal for Auto-Allocate or Auto-Target activities.. To avoid this restriction, one option is to use a proxy metric which has a strong correlation with your calculated metric. Q47 : How we create a more insight full and presentable target reporting views ? You can expand the Adobe Target reporting views by taking advantage of the Analytics or Customer Journey Analytics integrations with Target. This allow customers to run deeper analysis on their activities and create more complex reports. Q48 : What's ahead for Target ? We are exploring a future where deeper integrations of Target with other Adobe tools will make real-time, cross channel personalization easier to achieve. Q49 : For more information about sdk please review the following documentation: https://experienceleague.adobe.com/en/docs/target-dev/developer/client-side/aep/aep-web-sdk-overview if you need any support please submit a ticket or reach out to your account team.