Product Communities

Hi Team,I wanted to check on two items related to our upcoming A/B testing plans in Adobe Marketo Engage:1. A/B Testing ApproachCurrently, we use a single Email Program for our batch and blast campaigns. Each campaign is a monthly series with 4 to 6 emails, and we have separate Smart Campaigns set up to send each email. At the moment, these are standard sends without A/B testing. Going forward, we would like to run A/B tests for each email in the series (for example, 4 or 6 A/B tests within the same campaign). From what I understand, the out-of-the-box A/B testing functionality within an Email Program allows only one A/B test per program. If we want to run multiple A/B tests, one option might be to use Smart Campaigns and manage the test and winner selection manually. Please let me know if my understanding is correct?  I also wanted to check if there have been any recent updates or enhancements that allow multiple A/B tests within a single Email Program using the native functionality,

I’m looking for a non fusion solution. We are building a process where new requests will be submitted but we need to pull in certain fields from a previously related request. An example: A request was submitted to create a material and it gets mailed. Now we need to submit a new request but reference data from the old request so the partner teams have easy access to all of the data in that request. I did discover that I can create an internal lookup field so the requestor can key in the previous request and it provides a link to the request. I need to take this a step further and provide other key details so they don’t have to open the request and find the important information they need. Any suggestions? 

When I publish a document with an image on the front cover, it is supposed to set it to front and center, but instead it takes up the entire page:  The image itself is not that large. Guides seems to be expanding it. Here is the source: <?xml version="1.0" encoding="UTF-8"?><!DOCTYPE concept PUBLIC "-//Emerson//DTD DITA Concept//EN" "mmiconcept.dtd"><concept id="GUID-9e62db09-26b4-4e6e-9ec3-9fc68b449cda-enHYPHENus" outputclass="frontcover">  <title>G-SeriesPDSFront</title>  <conbody>    <image align="center" height="3in" href="GUID-0b554440-e72e-4261-bec9-8efa22a822d0-enHYPHENus.png" id="image_mpl_bdw_jz" outputclass="front_center" placement="break" width="4in">    </image>    <image href="GUID-b4720b3e-04db-4a4a-8ac8-1d6a401fe5bd.png" id="image_1" outputclass="brand_logo">    </image>    <section id="section_1" outputclass="pds-frontpage">      <title>Exceptional reliability and safety</title>      <ul id

Our site is authenticated — users must log in with ID and password. After login, Adobe Target (via Web SDK) sends personalization fetch and display calls to show personalized content.Setup:- Experience Targeting activity in Target- Audience = an Edge Segment in AEP, created to track qualifying activity within the last 24 hours- Profile shows status: realized for this segmentProblem:1. Open a fresh/incognito browser → new ECID is created2. This new ECID gets stitched to our already-qualified profile within seconds3. Despite that, the Target experience does not show — personalization:decisions comes back empty4. If we repeat the exact qualifying action again on this new ECID, the experience shows immediately after5. Same thing happens every time we use a new incognito sessionQA/Preview mode always shows the experience correctly (expected, since it bypasses real audience checks).Question:Is this normal for Edge Segmentation — does it only react to events it sees live on that specific ECID

Hi,If there is a streaming segment created with ONLY PORFILE ATTRIBUTE profile.firstname=’ABC’, and it has STREAMING INGESTION, and the schema and dataset are of class XDMPROFILE.Merge policy = ‘active-on-edge’ enabledNow how will this segment behave?First run? - what could be the evaluation methodNext runs? - what could be the evaluation method 

A new Adobe Universal Page Editor error is occurring for a code base that was previously working prior to 9/28.  The new error is “https://experience.adobe.com' has been blocked by CORS policy: Response to preflight request doesn't pass access control check: The value of the 'Access-Control-Allow-Origin' header in the response must not be the wildcard '*' when the request's credentials mode is 'include'.” The current OSGi configuration is as follows and does not establish anything from an origin perspective as a wildcard * ```{  "alloworigin": [    "https://localhost:3000"  ],  "allowedpaths": [    "/(content|conf)/xfinity.*",     "/graphql/execute.json/.*",     "/api/.*"  ],  "exposedheaders": [    "Authorization",    "Origin",    "Accept",    "X-Requested-With",    "Content-Type",    "Access-Control-Request-Method",    "Access-Control-Request-Headers",    "CSRF-Token"  ],  "supportedheaders": [    "Authorization",    "Origin",    "Accept",    "X-Requested-With",    "Content-Type",   

I'm trying to trigger a Journey entry using a unitary event condition based on an unsubscribe event coming from a landing page.Entry condition used:(isNotEmpty(@event{All_Consent_Update_Event._experience.customerJourneyManagement.messageInteraction.subscription.subscriptionListID})) and (in(@event{All_Consent_Update_Event._experience.customerJourneyManagement.messageInteraction.subscription.subscriptionChoice}, ["no"]))Context:When a profile unsubscribes via a landing page, the event is correctly generated and visible on the profile. Under messageInteraction.subscription, both subscriptionListID and subscriptionChoice are populated with the expected values (subscriptionChoice = "no"). However, when this same event/condition is used as the entry trigger for a Live Journey, the profile does not enter the journey.What I've already verified:The event is present on the profile timeline with the correct field values. The XDM path used in the condition matches the schema field path on the eve

Setup: one Org, five Launch properties serving five subdomains under one registrable domain (www, oem, my, … .example.com). First-party collection via CNAME smetrics.example.com. ECID v5.5.0, AppMeasurement 2.27.Goal: each subdomain should start with no Adobe identity cookie before its own consent is given, and ideally identity should not be shared across the subdomains.What I've confirmed:The server (CNAME) sets both AMCV and s_ecid with Domain=example.com (registrable domain). These bleed to every subdomain. The ECID cookieDomain config scopes only the client-side AMCV to the subdomain; it has no effect on the server-set cookies. Setting it to the literal subdomain, the Set-Cookie response still returns Domain=example.com. (Adobe docs also list cookieDomain/cookieDomainPeriods as retired.)My open question:If I provision a per-subdomain CNAME (e.g. smetrics.oem.example.com as trackingServerSecure on the OEM property), will the collection server set the cookies on .oem.example.com, or

Hi,I am seeing a data discrepancy in CJA when building a 12-month YoY chart inside a panel locked to "This Month" (September 2026).Identical months (e.g., July, August, Sep) show completely different numbers for the People metric depending on the table setup:Setup A: Generic Month of Year dimension in rows + custom Jan 1, 2026 - Dec 31, 2026 column override. (shows 9.9K people for Aug)Setup B: Rolling date ranges in rows (This Month, 1 Month Ago, 2 Months Ago) + custom Jan 1, 2026 - Dec 31, 2026 column override. (Shows 6.8K people for Aug)Why does the time-parting dimension (Setup A) data mismatch with that of the relative date components (Setup B)?

What if the native agentic skills within Marketo Engage could improve your team’s productivity and overall performance?There are a growing set of agentic skills in Marketo Engage, each one built to handle a specific marketing task that users invoke with a natural language interface. Early users are reporting significant time savings and new insights from their Marketo Engage data.View the on-demand recording to see the team who built them will walk you through how they work, live:Harman Bola (Sr. Product Marketing Manager) will cover the new AI capabilities and how early customers are already using them to save time John Dictson (Sr. Product Manager) will demo skills like import leads, validate programs, and surface insights, so you can see them in actionYou'll walk away knowing:What AI capabilities are available in Marketo Engage and how to use them tomorrow How customers are using agentic skills to cut down on repetitive tasks How they work, demoed liveView the on-demand recording: h

Hi all,I'm trying to reference a Salesforce-synced custom object in an email via Velocity, but it never renders — because the object doesn't appear in the Email Script Editor's "Custom Objects" tree at all.Setup:Native Salesforce sync. Custom object Producer Lead Detail (Producer_Lead_Detail__c), related to Lead via a Lead__c lookup. Sync is enabled.What works:It works as filters/triggers (e.g., "Has Producer Lead Detail", "Added to Producer Lead Detail"). A smart list on "Has Producer Lead Detail = true" returns the right people, and the records/fields are populated on the person (First Name, Person Source, Lead Created Date, etc.).What doesn't:In the Email Script Editor, the object is not in the "Custom Objects" section of the tree — it only appears nested under Person, and dragging it inserts ${lead.Producer_Lead_Detail__c} (a scalar), not a custom-object list. No list variable resolves. A dump via a real send returns blank for every variant:$Producer_Lead_Detail__cList -> blan

Hi everyone,I'm implementing Web Push in Adobe Journey Optimizer with the Web SDK, and I'm trying to pass custom metadata to the service worker using the Custom data section of the push content editor. The key-value pairs don't appear in the payload the service worker receives, and I couldn't find official documentation describing how Custom data is handled for the Web platform. SETUP- Web SDK (alloy.js) 2.35.0 with pushNotifications configured (vapidPublicKey, appId, trackingDatasetId)- Official alloyServiceWorker.js served from the site root- Push channel configuration with the Web platform selected- Event-triggered journey with a Push action; Custom data added to the content and the journey republished WHAT I TESTEDTo inspect the raw payload before the Adobe service worker processes it, I added a push listener at the top of the service worker file:self.addEventListener("push", function (event) {  console.log("[DEBUG]", JSON.stringify(event.data.json(), null, 2));});The notification

Do you think excluding visits where:Page Views is less than or equal to 1 AND Time Spent per Visit - granular is less than or equal to 1 (second)makes sense? (The page views aren’t that much important to my question, Time spent is more)In my opinion, it kinda does. Because 1 second or less is extremely fast for normal human traffic. This started because I spotted obvious bot/crawler/uncommon traffic with those conditions on specific countries and domains. Which then left me wondering about the meaning of a 1 second (at most) visit…I believe it could be real people entering and immediately leaving. Either by clicking by mistake or other reason, which none seem ‘meaningful’ to me (at least at the moment).Another reason I can think of, is tracking set up problems, where the dimensions/metrics aren’t correctly populated. But confirming the tracking is properly set, and only thinking of Marketing/Traffic analysis, not on a developer/programmer analysis..Thank you in advance

All Gantt export to PDFs fail.  Fails for all my colleges.  I’ve tried all the Google and AI troubleshooting steps. Any ideas?  AI research is pointing to a possible server issue that Adobe needs to address and nothing I can do.

Hi Adobe Community,We are currently working on optimizing a responsive web platform that serves dynamic HLS video streams alongside personalized media overlays using Adobe Target / AEM dynamic components.While the core dynamic media delivery works as expected, we are looking to fine-tune frontend performance—specifically reducing Layout Instability (CLS) during video player initialization and optimizing hardware-accelerated GL buffer management on mobile viewports.I’d love to gather insights from developers and technical architects here:1. What are the best practices for wrapping custom HTML5 / HLS video player scripts within AEM components without blocking critical-path rendering?2. How do you handle smooth component teardown and memory cleanup when switching dynamic video streams on single-page layouts?3. Any recommended client-side caching or lazy-loading strategies for high-concurrency video portals?Looking forward to hearing how other teams manage media-heavy, high-performance web

Hello,We are investigating an issue in AEM as a Cloud Service where some customer requests are being rejected by the Adaptive Rate Limiter with the message:server load exhaustedThe logs show that the rejection is based on the calculated server load and the predicted additional load, rather than the maximum number of in-flight requests being reached.*INFO* [qtp1005821649-5497] com.adobe.granite.ratelimit.AdaptiveRateLimitFilter Rejected <url>, server load exhausted. Server load: 749009/750000, inflight: 68/200, prediction: 35381Any guidance on interpreting these Adaptive Rate Limiter log fields and recommended troubleshooting steps would be appreciated.Thanks,Prince

Has anyone implemented a holiday send freeze for an always-on nurture program in Marketo Engage? I am looking for a way to temporarily pause email send from a specific nurture journey built with trigger smart campaigns during holiday periods while keeping members in their current wait steps and preserving the existing journey logic.

I’m setting up Recommendations for the first time using this as reference: How Do I Implement Recommendations Activities? | Adobe Target but also many other pages of instructions.I understand that to build the list in the Catalog, I can use Feeds or the Recommendations API. There is a third option, Delivery API, that I haven’t explored. I tested using the Recommendations API and it worked. I was able to pass an entity.I’m not clear about this step (from the above page): 3. Pass behavioral information and contextIn Launch (aka Data Collection) I set up a Web SDK Variable and Rule that sets various entity attributes such as entity.id, entity.name, entity.categoryID, and so on. I can see this data on the interact payload and assume it’s being passed to Target. More inf here: Send parameters - Migrate Target from at.js 2.x to Web SDK | Adobe TargetWhat is the purpose of passing Target params? Is it to inform Target about visitor behaviour? Is this a different thing than adding entities to

I am investigating an Adobe Analytics implementation issue and would appreciate guidance on understanding the root cause.Current SetupWe are using:Adobe Analytics Extension (AppMeasurement) Custom Tracking Server configured in the Adobe Analytics Extension AEM as a Cloud Service Author and Publish environmentsThe Adobe Analytics Extension is configured with: Tracking Server:smetrics.company.comSSL Tracking Server:smetrics.company.comBased on Adobe documentation for trackingServerSecure, my understanding is that Analytics hits generated by this implementation should be sent through the configured tracking server rather than Adobe's default collection domains.Observation #1While testing an AEM Author URL similar to:https://author-pxxxxx-exxxxxx.adobeaemcloud.com/content/site/page.htmlI inspected Network requests and found Analytics calls being sent to:amcglobal.sc.omtrdc.netwith report suite information resembling:amc.global.prodinstead of using the configured tracking server:smetrics.co

Hi All, I wanted to know if anyone has attempted to make a Proof HQ api call in order to remove/close out an active approval process?   Looking for a way to close out outstanding proof approvals on projects that should be closed.

I got the case like this: Would have multiple identities for the logged user in my AEP implementation - however its easily done using IdentityMap. I would have multiple business entities However in AT there can be only one Identity - Target Third Party ID Namespace (so Target Profile can recognize only one Identity from IdentityMap How the user recognition works? AEP connects with AT on Edge so I assume its based on the IdentityMap not Target Third Party ID Namespace? Did you have the case like that? 

Hi ​@Community I am working on activating qualified audiences to an SFTP destination in Adobe Experience Platform.We are seeing an issue where some profiles that have already exited the segment are still being sent in the SFTP export and misses the realized ones.Current setupAudience qualification / segment evaluation happens at 8:45 PM IST Export for same Audience is configured at: After segment evaluation to the SFTP server 10:30 PM IST 11:30 PM IST What we are observingThe 1st and 2nd exports look the same In these exports, 1–5 profiles out of around 100 sometimes appear even though they have already exited the segment The 3rd export at 11:30 PM IST contains the correct and additional data all are in realized state. ImpactBecause of this issue, the same profile  (1-5) is being sent again on the next day. My questionI expected the export job to run only after segment evaluation is completed. In the AEP UI, it looks like the evaluation finishes first and then the export starts.Could

Hello, Can anyone help me with the intended use of being able to map segment membership in the destination out of the box mapping tool? segmentMembership.ups I am attempting to map a master audience to a destination and use ECN segment membership from other audiences to derive an attribute. I do not want to map the audiences I need segment membership from to the destination.  Can anyone tell me How can I use segmentmembership.ups in destination mapping? Solutions for mapping segment membership from other audiences into a calculated field where the audience is not published to the destinationThank you!

I use a Contentful as a CMS for web. CMS requests Edge Decisioning API and then render the offer. I need to debug what identities are sent in the request do you know if that request land is some specific dataset? 

Have you used that API?My Prod and Dev are not 1:1 so I am expecting problem. I assume as a first step do the POC: Prod Offer Decisioning → Dev Decisioning for one specific Decision ScopeDo you recommend creating all the profile attributes, audiences etc on Dev and then map prod → dev?Any advices?