☕[AT Community Q&A Coffee Break] WEDNESDAY 4/12/23, 8am PT: AI Features, Automated Personalization & Auto-Target with Matthew Bailie & Georgia Frailey of the Adobe Target Consulting Team ☕

[Amelia_Waliany]

Join us for the next Adobe Target Community Q&A Coffee Break 

taking place Wednesday, April 12th, 2023 @ 8am PT - 9am PT 

REGISTER HERE

We'll be joined by Matthew Bailie (aka @MBailie) and Georgia Frailey (aka @Georgia_CA), both Senior Solutions Consultants for Adobe Target, and @simochen of the Adobe Technical Support Engineering team, who will all be signed in here to the Adobe Target Community to chat directly with you on this thread about your Adobe Target questions related to:

 Target's AI Features, specifically Automated Personalization and Auto-Target

PARTICIPATING IS EASY:

• Sign in to the Community, Reply to this thread with your questions related to the Adobe Target topics covered 
• Stand-by during the hour as the experts weigh in ~ this way you can clarify your question or ask for more details directly 
• THAT'S IT!  *(think of this as the Adobe Target Community equivalent of an AMA, (“Ask Me Anything”), and bring your best speed-typing game) 

INSTRUCTIONS 

• Click the blue “Reply” button at the bottom right corner of this post
• Begin your Question with Georgia's and Matthew's Community Usernames: @Georgia_CA and @MBailie
• 💡Something already coming to mind?💡 Feel free to post your topic-related questions in advance, to ensure our experts get a chance to weigh in on your question while they're signed in for the hour ⌛

REGISTER HERE

As a Senior Business Consultant for Adobe Target, Matthew leads online optimization go-
to-market strategies for Fortune 500 companies across retail, financial, hospitality, and healthcare business verticals. For Fortune 100 companies, he creates industry-leading personalization frameworks and optimization roadmaps toward improving customer journeys across traditional marketing channels ~ websites, mobile app, and email. 

Georgia has over a decade of experience as an Adobe Senior Business Consultant, with trusted strategic partner clients in the Fortune 100 that spans multiple industries, specializing in opportunity-identification and personalization, both in digital space and omnichannel. 

Curious about what an Adobe Target Community Q&A Coffee Break looks like? Be sure to check out the thread from our latest 2/1/23 Adobe Target Coffee Break  with Adobe Target Product Marketing Manager @bkostak, with extra support from @shelbygoff  of the Adobe Technical Support Engineering team. 

[GavinAttard]

Thanks Bailie, looking forward to seeing the doc. 

On 2, removal of segments or data points? The issue is control of what influences the machine's creation of a segment. Going through client care for this seems impractical as the activity really is BAU... 

[Ryan_Roberts_]

Hi @GavinAttard,

You can use the model and blocklist API's we recently released to remove certain attributes from consideration by the model for a single AP/Auto-Target activity or for all AP/Auto-Target activities. This may be more convenient than a ticket to client care. Documentation can be found here: https://developer.adobe.com/target/administer/models-api/

[GavinAttard]

wasn't aware of this. Looks interesting and a potential path forward. woudl be good to see a UI for this in AT

[Georgia_CA]

Good idea Gavin

[AjaSullivan]

Can you help explain the difference between Auto-Target and Recommendations? My company has used Auto-Target to display marketing offers on our homepage and wonder how using Recommendations for those offers would differ.

[bailie]

@AjaSullivan Auto-Target uses customer attributes to generate look-alike models to help predict who should see what content. These models incorporate both customer behavior (pages visited, what are they clicking on) along with their attributes (1st-3rd party data). Auto-Target is more about creating a personalized experience. Recommendations is a much lighter in that only the algorithm you select (e.g. post popular, most viewed) is used to determined what content to show to the customer. Also, with Recommendations you don't get ML insight reports. There are different use cases for Auto-Target and Recommendations.

[Georgia_CA]

Also @AjaSullivan, think of Recommendations as things such as "people who viewed this, viewed that" scenario which involves a large library of content such as products or articles.

[Tushar_Kar]

We have a requirement on our website where we need to update a section from target after a link click on a dynamic section. What is Adobe recommendation for the way to do it.

[simochen]

Hi @Tushar_Kar, from my understanding of the question, you're looking to update a section of an activity after clicking on a link selector. You can use the Modifications panel in the VEC while editing the activity to make several possible changes, mainly through CSS Selector, Mbox, and Custom Code modifications. These changes include inserting offers or changes to the actual element itself.

You can learn more through the following documentation: https://experienceleague.adobe.com/docs/target/using/experiences/vec/modifications/vec-code-editor.h...

[Ryan_Roberts_]

Hi @Tushar_Kar,

Sounds like you are talking about needing to get a new decision or new content from a Target activity after a click event rather than a page load or SPA view? I'd say that are multiple ways to handle that and which is best probably depends on the specific use case. 

1. You could call a getOffers() when the click happens to retrieve new content from Target and then use the applyOffers() method to render it after the dynamic section has been updated. There are implementation level changes you would need to make. This allows Target to decide what specific activity and experience should be shown at the moment of the visitor's click.
2. You could use the prefetch and triggerView() function. When the dynamic section is clicked you call the triggerview() with a custom name for that click event. Then Target will render the content already loaded by the pageload request to Target but stored in memory until the triggerview() is called. Then in the VEC you will have the option to make modifications on the different "views" of that given page. More info here: https://developer.adobe.com/target/implement/client-side/atjs/atjs-functions/adobe-target-triggervie...

[Tushar_Kar]

Thanks for the response

[Optimizing]

@MBailie & @Georgia_CA hello! 

My question is around the latest release and now being able to use A4T with Auto Target activities. https://experienceleague.adobe.com/docs/target/using/release-notes/release-notes.html?lang=en Is there any different setup needed to use it or is it simply choosing Analytics as the reporting source vs. Target when you're setting up your test? Anything else to be aware of from a reporting standpoint? 

Thanks!

[Georgia_CA]

Hi @Optimizing! Yes, you can absolutely use A4T as the reporting source for Auto-Target. There are modifications that you must apply so that attribution is correct. Please see this link for thorough instructions and context: https://experienceleague.adobe.com/docs/target-learn/tutorials/integrations/set-up-a4t-reports-in-an...

Hope that helps!

[lonestarspock]

What are the SLAs for response times with Adobe Target APIs?(offer, profile update, profile retreival) 

[bailie]

Hi @lonestarspock, I recommend following up with Client Care on specific SLAs based on your use case.

[simochen]

To add to this, Adobe has micro services architecture, so there are downstream services which we invoke, but all downstream services combined fall under SLA of 500ms. That means Target APIs in terms of being received and sent by the server should fall under that 500ms. 

On the other hand, specific APIs such as "single profile update API" can take different times (up to 24 hours) for changes to actually be reflected in Target. Additionally, the 500ms SLA is for the 99.5th percentile. In practice, the p99.5 latency is typically <250ms and the average is much less than that, but we only provide guarantee at the 500ms SLA.

[JoeF2]

Is it possible to use the ML learned data from old campaigns for new campaigns with the same target audience but with different content?

[Georgia_CA]

Hey @JoeF2, thanks for your question - short answer is no you cannot.

[JoeF2]

So the learnings from ML are on a per campaign basis unable to be reused, it sounds like...

[Georgia_CA]

Correct!

[wozinga]

Hi Joe,

You may want to look at the grouping capability, as long as the offers have a similar "theme", it may do some of what you want to accomplish, but not by creating a new campaign. 

https://experienceleague.adobe.com/docs/target/using/activities/automated-personalization/offer-repo...

For example, if you have an offer than is aimed at "price shoppers" you could add a new offer on the same topic, and remove the original and the new offer effectively takes the place of the other and "inherits" it's targeting.