@pujakumari216 I had some of these and added some extra context. Let me know if this helps
A/B/n Test: Compares two or more experiences (variations) to see which one performs better for a specific audience and objective.
- Example: An e-commerce website wants to test two different versions of its homepage to see which layout leads to higher conversion rates. One version features a relevant image, while the other version includes a relevant short video.
Auto-Allocate: Identifies a winner among two or more experiences, while automatically allocating traffic to the winner, increasing conversions as the test runs and learns.
- It's an A/B/n test except Target pushes more traffic to the highest performing experience, meaning you can see the benefits of the winning experience earlier than the end of the test period.
Multivariate Test: Tests multiple elements or components of a page simultaneously to find the optimal combination.
- Example: An eComm store wants to test multiple combinations of the "Buy" button on a product page to see which generates the most sales. They choose 2 different colors (Red & Black) and 3 button labels: "Buy Now," "Purchase: and "Add to Cart." An MVT test will combine these variations to find the most effective one (towards your goal). That's 6 different variations. So, 1 user may get a red button that says, "Buy Now" and another user may get a black button that says, "Add to Cart." (and so on).
Experience Targeting: Delivers tailored experiences to specific audiences based on predefined rules and criteria.
- Example: An insurance company targets visitors from different major cities with customized, regional content (i.e. Based on their location, users in a given city can be shown relevant imagery of landmarks within their city).
Recommendations: Uses machine learning to provide personalized product or content recommendations to your visitors, based on their preferences and interests. Recommendations uses built-in criteria or algorithms that you can choose from or create your own custom criteria.
- Example: A streaming service provides personalized recommendations for movies and TV shows based on each user's viewing history, preferences, and ratings. Also, an eComm site may say, "People who bought this, also liked this" and show you recommendations. It can be used across industries.
Auto-Target: Uses machine learning to serve the most tailored experience to each visitor based on the individual customer profile and the behavior of previous visitors with similar profiles. So, it's similar to an A/B/n test, in that the user is seeing different predefined experiences you created, however, it's not actually a test. It's sending users to the most relevant experience based on their changing profile & behaviors.
- Example: Going back to the A/B example, Auto-Target might show the home page that is most likely to convert, based on the specific user profile. Maybe desktop users prefer video, or people in the US, on phones like images.
Automated Personalization: Uses machine learning to personalize content and drive conversions by combining specific offers, and then matching different offer variations to visitors based on their individual customer profiles.
So, where Auto-target is at the experience level, this is even more personalized, with no predefined setup of an entire experience, is necessary. You can have multiple offers on a page & it will show the combo that works best for that user.
- Example: A financial services company wants to deliver very relevant — and very different — offers to a different user such as a new homeowner, a recent graduate, and someone approaching retirement. Based on the various attributes, their browsing history, previous purchases, etc. each individual is provided the most relevant offer.
Note: Recommendations requires classification of products or offerings and generally a number of other set-up work, which is why it’s a very different activity than the others.
