Moses is an Adobe Senior Solutions Consultant specializing in Adobe Target. He brings a depth of knowledge from his years of experience working closely with global brands on their optimization and personalization strategies from both a strategic and technical perspective. Additionally, Moses has extensive experience working with Adobe Target’s automation capabilities and loves talking with organizations on Target's Sensei features.
Curious about what an Adobe Target Community Q&A Coffee Break looks like? Be sure to check out our past 2020 Adobe Target Coffee Breaks, covering a WIDE variety of Adobe Target topics, and our latest thread from our 5/12/21 Adobe Target Coffee Break with @Shannon Hamilton
Topics help categorize Community content and increase your ability to discover relevant content.
@mosesmaxen is there a way to provide a heavier propensity score to certain data points of a personalization activity in Target? Also, is there any plan to allow experience preview URLs for AP activities that are using at.js 2.x?
@davidmailey wrote:@mosesmaxen is there a way to provide a heavier propensity score to certain data points of a personalization activity in Target? Also, is there any plan to allow experience preview URLs for AP activities that are using at.js 2.x?
Currently we have the ability to weight data points like a higher propensity score as part of a Recommendations activity. This can be leveraged as part of the Attribute Weighting feature. So for example, items with a margin higher than 55%, weight 15x in the algorithm to boost those items with margins higher than 55%. This can also be used to bury certain items.
Preview links for Automated Personalization activities are now supported with the latest at.js 2.5 version release. https://experienceleague.adobe.com/docs/target/using/release-notes/release-notes.html?lang=en
@mosesmaxen what new features are available from Sensei for Target?
@jillstolt wrote:@mosesmaxen what new features are available from Sensei for Target?
Here is a list of some of the recently made available features related to Target's Sensei capabilities.
Recommendations
Our new Recommended for You algorithm as part of Recommendations engine recommends items based off of each visitor’s browsing, viewing, and purchasing history. This provides you the ability to serve personalized recommendations to each individual user.
List-based filtering - as part of a recommendations criteria the ability to refine recommendation results based on values contained or not contained in a list.
Slot level control - This feature also added gives you the ability to control how many slots within a recommendations results are populated for a given criteria within a criteria sequence.
(Coming Soon) Cart Based Recs - the ability to recommended catalog items based on the current contents of the visitors cart.
Faster algorithm runs - The recommendations results for certain algorithm criteria will be executed more frequently.
A4T support for Auto-Allocate and Auto-Target
We now have A4T support for Auto-Allocate and Auto-Target activities for reporting analysis inside of Adobe Analytics. This also includes the ability to select an Analytics' metric to optimize towards as part of those activity types.
https://experienceleague.adobe.com/docs/target/using/integrate/a4t/a4t-at-aa.html?lang=en
Auto-Target and Automated Personalization
You now have the ability to select a control experience as part of an Auto-Target and Automated Personalization activity. This gives you the ability to test and compare the lift of the activity against the default or any particular experience.
Preview links are now also supported with at.js 2.x when using version 2.5 for Automated Personalization activities.
Hello Experience League Community, great to be with you all this morning to answers any of your Target testing, personalization, and product questions!
@mosesmaxen Does Target integrate with LiveChat to run tests within the LiveChat popup?
@jillstolt wrote:@mosesmaxen Does Target integrate with LiveChat to run tests within the LiveChat popup?
Although Target doesn't have a prebuilt integration with live chat solutions, Target can be used to test things like location and presentation of the live chat. One approach is to use the custom code editor within the Visual Experience Composer or the forms composer to do more advance testing if needed. Target also has a delivery API that can be used to invoke Target from any internet connected app so if LiveChat can call that API from inside of LiveChat the integration may be possible.
Thanks @mosesmaxen !
@mosesmaxen Is it recommended to remove Downlift Experiences in a running Auto-Target Activity once ML i.e. the green check is available and add new or other experience in order to optimize "on the go" the considered Experiences in Auto-Target?
@Nicolas_Swisscom wrote:@mosesmaxenIs it recommended to remove Downlift Experiences in a running Auto-Target Activity once ML i.e. the green check is available and add new or other experience in order to optimize "on the go" the considered Experiences in Auto-Target?
Great question! If any particular experience appears to be performing poorly should you remove it and replace it with another experience. There are a few of things to consider:
1. The key indicator of an Auto-Target or Automated Personalization activity is how well the overall Targeted serves are doing compared to the Control serves. If you are seeing strong performance there, you many not need to change anything.
2. The green check boxes indicate that models are built; however, models may still be maturing. So I wouldn't be too hasty in make that decision.
3. If there are a group of visitors to your site that simply aren't going to convert regardless of what Target shows them you may always have an expereince that appears to be under performing.That all said, it can be helpful to introduce new "challenge" experiences into an existing activity. One that performs poorly overall and when being Targeted (relative to the other experiences) can often be improved by finding a better message that might appeal to a segment of your visitors.
@mosesmaxen Thanks for your time Moses. In another session Offer-Engine was mentioned to be available in Target. Can you reveal more regarding funcionalities, features and dates?
@Nicolas_Swisscom wrote:@mosesmaxenThanks for your time Moses. In another session Offer-Engine was mentioned to be available in Target. Can you reveal more regarding funcionalities, features and dates?
Hey Nicolas! If you are referring to Adobe Target's integration with the offer decisioning capabilities, this is in active development after being announced at Adobe Summit. When we are closer to a beta or early access stage we will be sure to let you know!
@mosesmaxenIs there a way or will it be available in future to decide which audiences/variables are beeing considered to feed The Automated Segments Report for Auto-Target? Currently it shows us 100 Automated Segments of which many are not really relevant or based on very old or test segments/audiences...we would like to exclude/include segment/audiences if possible?
And can it show more than 100?
@Nicolas_Swisscom wrote:@mosesmaxenIs there a way or will it be available in future to decide which audiences/variables are beeing considered to feed The Automated Segments Report for Auto-Target? Currently it shows us 100 Automated Segments of which many are not really relevant or based on very old or test segments/audiences...we would like to exclude/include segment/audiences if possible?
And can it show more than 100?
While there are no specific items on the roadmap for that report we are reviewing the opportunities there and have ideas for the future.
@mosesmaxen When would we want to serve personalized recommendations to each individual user VS. using Auto-Target or AP (thinking of a homepage for example)?
What are the specific uses cases for each? Once I have a catalogue with hundreds or thousands of item it should be Recommendations?
@Nicolas_Swisscom wrote:@mosesmaxen When would we want to serve personalized recommendations to each individual user VS. using Auto-Target or AP (thinking of a homepage for example)?
What are the specific uses cases for each? Once I have a catalogue with hundreds or thousands of item it should be Recommendations?
Auto-Target and Automated Personalization are great for when you have a lower number of experiences or offers within a lower number of locations. Recommendations is the way to go when we are talking about hundreds / thousands of recommended items. Automated personalization can also scale heavily but would also require a larger amount of traffic.
Hi @mosesmaxen Thank you for taking the time today! This question was posted by Target Community member abid63370229:
How to interpret weights in criteria attribute weighting? (Recommendations): I'm using a Viewed This Viewed That criteria and finding that with no attribute weighting, a certain type of product is dominant and shows up about 65% of the time. I'm trying to use attribute weighting to "boost" the other product types. When I set those product types to 50% weights (and no weight for the dominant type), it does not seem to change the proportion of product types that I see in the resulting recommendations data. When I set the alternative product types to a 75% weight, they become too dominant, and my previously dominant type goes down too much (to about 3%). There's no option to set a weight in between 50% and 75%, so I'm wondering if there is something else I can do to achieve an intermediate result.
More broadly, I don't fully understand how to interpret the 0/25/50/100 attribute weights. If something is set to a 50% weight, does that mean that products with that attribute will be recommended 50% of the time? This doesn't seem to be the case since you can set the rules to sum to more than 100% (e.g. 3 product types with 75% weight each). Does 50% mean 2x as likely to recommend as another type?
Any help is much appreciated!
Views
Replies
Total Likes
@Amelia_Waliany wrote:Hi @mosesmaxen Thank you for taking the time today! This question was posted by Target Community member abid63370229:
How to interpret weights in criteria attribute weighting? (Recommendations): I'm using a Viewed This Viewed That criteria and finding that with no attribute weighting, a certain type of product is dominant and shows up about 65% of the time. I'm trying to use attribute weighting to "boost" the other product types. When I set those product types to 50% weights (and no weight for the dominant type), it does not seem to change the proportion of product types that I see in the resulting recommendations data. When I set the alternative product types to a 75% weight, they become too dominant, and my previously dominant type goes down too much (to about 3%). There's no option to set a weight in between 50% and 75%, so I'm wondering if there is something else I can do to achieve an intermediate result.
More broadly, I don't fully understand how to interpret the 0/25/50/100 attribute weights. If something is set to a 50% weight, does that mean that products with that attribute will be recommended 50% of the time? This doesn't seem to be the case since you can set the rules to sum to more than 100% (e.g. 3 product types with 75% weight each). Does 50% mean 2x as likely to recommend as another type?
Any help is much appreciated!
Thank you for your question. The value defined in the attribute weighting against an attribute determines the type of item that is more likely to display. The percentage does not directly correlate to the exact percentage of how often the item with that attribute will be displayed. It is more of a guide to boost or bury particular items. If you are looking for more control over a curated list of recommendations then the custom criteria with an uploaded curated list is a great approach as well. Thanks!
Views
Replies
Total Likes
@mosesmaxen A few customers have been asking when will there be support for response tokens and profile params in the new AEP Web SDK?
@MihneaD wrote:@mosesmaxenA few customers have been asking when will there be support for response tokens and profile params in the new AEP Web SDK?
Great question! There is now support for profile parameters when using the AEP Web SDK. We are also targeting the end of June for adding response tokens there as well. For keeping track of the latest AEP Web SDK release notes please see: https://experienceleague.adobe.com/docs/experience-platform/edge/release-notes.html
Views
Replies
Total Likes
Views
Likes
Replies
Views
Like
Replies
Views
Likes
Replies
Views
Likes
Replies
Views
Likes
Replies