Product Communities

Summary: The office hours covered Marketo AI and Marketo Engage MCP Server updates, customer Q&A, and roadmap direction. Key topics included upcoming UI and roadmap changes, callable agents, campaign and program-building capabilities, differences between Marketo Engage MCP Server and Marketo AI, permissioning and API quota considerations, reporting limitations, data-movement constraints, and future connector expansion.AI generate topics with timestamps4:08 — Roadmap and product updates: John outlined upcoming roadmap visibility, UI changes, and near-term feature development, including callable agents, campaign and program building, and standardized data capabilities expected in the coming months.8:21 — Marketo Engage MCP Server vs. Marketo AI: The team clarified that Marketo Engage MCP Server is an external, prompt-driven interface where an LLM calls Marketo APIs, while Marketo AI operates inside the Marketo UI using built-in agent skills tied to user permissions.15:41 — Permission

Summary: The office hours covered Marketo AI access, governance, and product capabilities alongside the generally available Marketo Engage MCP Server. The discussion included setup requirements, API usage, reporting limitations, list import functionality, program validation, and planned enhancements to improve usability, analytics, and import workflows.AI Generated topics with time stamps3:13 — Office hours overview: The session focused on Marketo AI and the generally available Marketo Engage MCP Server, including Q&A and product walkthroughs.6:43 — Marketo AI access and permissions: Marketo AI is available through the My Marketo interface and is governed by admin permissions, generative AI terms, and existing workspace and asset permission structures.12:25 — Program QA and validation: The team demonstrated program validation using organizational rules and optional test plans for automated QA checks, though the walkthrough was partially limited by environment performance issues.16:

We noticed our original referrer data starting to get stripped which moved what would have been Natural Search traffic into Direct throughout the summer, but especially in early August. Are there any ideas on if this is related to the Safari 26 browser update tied to an iOS release, and are there any work-arounds to get that traffic back in under Natural Search in Adobe Analytics?

Previous, I had been able to generate an HTML5 output using the default DITA-OT template that Guides comes with. The result had been a folder of HTML, SVG, and PNG files.In the current version of Guides (5.1), the HTML5 output no longer generates anything except the primary index file with links to nothing. I have tried to generate an HTML5 output using both the Map Console and the Map Dashboard and both result in the same thing: two css files and one index.html file with links to nothing.What am I doing wrong? The default DITO-OT file in CRDXE hasn’t been changed, moved, or touched.

Hi all,I'm currently having trouble integrating traffic data from Adobe into our marketing platform (Datorama) using the 2.0 API.With the legacy 1.4 API, we were able to apply multiple segments within the Extended Properties section to filter the data. For example, excluding business-related visits, upgrade journeys, and other traffic types. However, in the 2.0 API connector, I appear to be limited to a single segment and cannot add additional segment filters.Can anyone confirm whether this is a restriction of the 2.0 API, or if there is an alternative approach to apply multiple filters/segments within a single stream?Any guidance or workarounds would be greatly appreciated.Thanks,H

Hi folks,For some of the pages, I am able to see Visits is greater than pageviews. Ideally this should not be the case, right? as per my understanding. but at overall level pageviews is greater than visitsThis started happening when i changed expiry of evar from visit level to hit(i breakdown page name with day and noticed this)Some of the notes :1. I am visiting website → accepting cookies → pageview fires first → and then engagements as i make so it seems no action is happening before pageview2. I am sending page name and URL with all my engagements (scroll and link clicks) as i want to know what happened on which page. Also can you please let me know if this is correct or do i need to change something in my implementation.

I made the following observation:In the Designer (FormDesigner.exe) for interactive forms, the preview window remained blank.Although the AcroRd32.exe process was launched in the background and the temporary PDF was created in %TEMP%, it was not displayed.A quick analysis revealed that the Windows Update KB5121003 had been installed the day before.Once it had been uninstalled, the problem was resolved.Is this behaviour known?

 If you've ever opened your browser's network tab while debugging an AdobeAnalytics implementation, you've seen it: a long, ugly request to somethinglike /b/ss/examplecompanyprod/1/H29-abc123 or /ee/v1/interact, packedwith query params or a JSON body nobody memorizes on purpose. Tools likeOmnibug or Adobe's ownExperience Platform Debugger will decode that for you -- nicely, even --so you can eyeball it and go "yep, that looks right."That's the thing, though: eyeballing it is the whole workflow. There's nostep where a machine checks the hit and fails a build if it's wrong. Youship a change, the tag either still works or it doesn't, and you find outwhen an analyst asks why last week's numbers look weird.This post is about what's actually inside these two hit shapes, and why Iended up writing a parser (and then a few more tools) instead of justliving with the debugger-and-eyeball loop.Two generationsAdobe Analytics tracking comes in two flavors depending on what'simplemented:AppMeasuremen

Hi,what module do you use in Fusion to get a decision when it’s made on a new Unified Review & Approval document? I cannot find it in a regular Workfront Watch Events module nor in the dedicated Unified Review & Approval module. Thanks!

If scheduled batch data ingestion fails is there a way to schedule an email notification?

Context: We run an AEM Edge Delivery Services site (repoless, Configuration Service; content authored in AEM). Sitemaps are generated by helix-sitemap. We push the sitemap config to content/sitemap.yaml via the Admin API and generate per-language sitemaps. Multilingual: es, en, pt, fr.What we've confirmed: Per-language <urlset> generation works via a languages map with source: /{lang}/query-index.json. <lastmod> is emitted when we add a lastmod format to the config. We could not find any way to emit <priority> or <changefreq> per URL, and we don't see them in the config schema. helix-sitemap doesn't appear to generate a <sitemapindex> for multilingual — only the per-language <urlset> files. Questions: Supported per-URL attributes: What sitemap attributes can helix-sitemap emit per URL (<loc>, <lastmod>, <changefreq>, <priority>, hreflang alternates)? Is there a definitive list? priority / changefreq: Is there ANY suppo

Hi , I am currently learning EDS and wanted to get hands on repoless. Iread that it needs admin key and api access to configure the repoless.Is there any way that we can do some hands-on on our own without using organisational keys?

Hi, Not showing the create site from template option in the AEM trial version for EDS setup. Can anyone help me on this

Adobe's been promoting AEM Sites Optimizer as a GenAI-first layer on top of AEM as a Cloud Service, aimed at closing the gap between "the site is live" and "the site actually performs well." After checking it against Adobe's current docs, here's a practical breakdown of what it does and where teams should apply healthy skepticism.1. It's built around four pillars, not threeSites Optimizer identifies and diagnoses issues across:Technical SEO Core Web Vitals Accessibility Content engagementThat's a broader scope than most teams assume accessibility issues get flagged alongside performance and SEO problems, not treated as a separate audit process.2. It's not just "suggestions" , it follows an Identify → Suggest → Fix workflow, with one-click applyThis is the part that's easy to undersell: Sites Optimizer doesn't just surface a dashboard of recommendations for humans to manually implement later. It continuously collects and analyzes performance data, prioritizes opportunities by projected

Fellow AEM folks, The pressure to go "Headless" or "Hybrid" is real. Every client and every project seems to be asking for it. But is it always the right answer? The Good: For those who have made the leap, what was the biggest actual benefit you achieved? (e.g., developer freedom, faster site speed) The Bad: What was the most unexpected downside or complexity you faced? (e.g., content preview, author experience, cost) The Verdict: Is Headless AEM a must-have for future-proofing, or is it an overkill solution for most common use cases? When should a traditional AEM setup still be the preferred choice? Let's get past the jargon and talk about what really works (and what doesn't).

Does it bother anyone else how the versions are labeled in AEM? I have no idea which one is which. Instead of v1, v2, v3, etc they keep appearing as New Version 1.0, New Version 1.1, New Version 1.0.0. What on earth do any of those mean? Cant these be listed as Original, Version 2, Version 3, etc. and then show the date. they were uploaded and by whom? That is how the average human brain would read these things. Is that a setting I can change or is this the standard operating procedure of the system?  

Hi everyone,I don’t know if you can help me with this question abount Whatsapp Native Webhook: What we tested:We submitted a real Meta delivery callback payload (wamid + recipient_id + status read) directly to the AJO native WhatsApp webhook URL.The endpoint responded HTTP 200 – Webhook data processed successfully, but no event appeared in AJO Email Tracking Experience Event Dataset. Relevant docummentation (whatsapp-configuration) mention this points:"A profile must have at least one message sent from Journey Optimizer before incoming messages are captured in this dataset."Our interpretation: AJO requires having originated the message natively to register the wamid internally — without that record, it cannot correlate the Meta callback to a profile, so it accepts the payload silently (HTTP 200) but persists nothing. Questions:Is the AJO WhatsApp webhook designed exclusively for messages sent through the native AJO WhatsApp channel?   Is HTTP 200 expected even when AJO cannot correlat

Hi! a client is insisting that the whatsapp webhook URL is allowed to send any information to CDP, not only Inbound categories to tracking. They want to send whatsapp delivey logs using that webhook. I think that’s not possible, that webhook URL is only allowed for tracking logs, is that correct? thanks.

We are investigating an issue related to Unspecified values and are trying to determine whether we're encountering a configuration issue, a data collection issue, or a platform limitation.In the example below, some conversions are successfully attributed to the menu-details dimension using Link Position, while 68 other conversions appear as Unspecified and cannot be tied back to a dimension.Why would some conversions be attributable to a dimension such as Link Position or Link Name, while others remain Unspecified? Is it reasonable to expect that most click-tracking conversions should be definable and reportable against a link dimension, or are there scenarios where Unspecified is the expected behavior?More specifically, can Target users define click-tracking conversion metrics using link-level dimensions such as Link Position and/or Link Name, and are there any known limitations that would prevent these dimensions from being available for certain conversions? 

Learn more about Adobe Marketo Optimizer.The Lighthouse Program gives a small group of Marketo Engage users the chance to try Adobe Marketo Optimizer for free during a focused testing period. Adobe Marketo Optimizer helps you engage customers across every channel and continuously improve results with AI at the core. With AMO, you can:Optimize with AI - let AI-powered decisioning and insights surface what's working, recommend next best actions, and continuously improve campaign performance  Engage across channels - reach customers with connected, coordinated experiences across multiple channels Build journeys visually - design, adjust, and orchestrate customer journeys on an interactive journey canvas, no guesswork requiredWhat you'll get access toDuring the program, you'll be able to test person journeys, emails, and forms, plus a wealth of agentic skills that help you investigate your Marketo data and optimize your workflows.As a Lighthouse participant, you'll: Get early, hands-on acc

When we open a record in full-screen mode or in a new tab, the Record Type “Features” is displayed at the top, as highlighted in red in Screenshot 2. However, when we open the record directly in the minimized view, as shown in Screenshot 1, the Record Type “Feature” is not displayed at the top. Could we also display the Record Type in the minimized view? This record is a connected record. 

I have been testing with the AI Assistant and I have been able to have it fill out a Request Form from  an excel file. I am finding that it will not fill out the Typeahead fields, assuming the problem is that it needs something manually entered to get the choices to populate. Wanted to see if anyone else has tested this and found a workaround? 

Is the new Unified Approval with Frame.io capable of interactive proofs? We use Workfront Proof for review and approval of email coding before deployment. Is that possible from Workfront using the Frame.io viewer?