How to Segment First-Time Purchasers in Adobe Analytics?

Hi @mezzome ,

To segment first-time purchasers in Adobe Analytics, you can isolate users who made their first-ever purchase during a specific time period by using sequential segments and optionally visitor-level logic. Here's a step-by-step guide to do this using segments, which is the most straightforward method:

Go to Components > Segments

Click "Create New Segment"

Use the Following Settings:

• Because we want to find users whose first purchase ever happened in this time window.

• Then logic
• First condition (before your date range):
• Hit where Purchase Event exists
• With date range set to before your analysis period (e.g., before July 1)

• Hit where Purchase Event exists
• With date range within your analysis window (e.g., July 1 to July 31)

Example: 

• Does NOT have: Purchase Event before July 1, 2025

• Has: Purchase Event between July 1 and July 31, 2025

Hi @mezzome 

This is a bit of a tricky solution but is possible. 

If you want to isolate the first time purchasers create a visitor based segment where orders is equal to 1.

This will check it based on the time frame in the calendar in the workspace. 
You will notice that if you change the date range the values change.

If you want it only looking back a specific number of days, then you can add a rolling date parameter to the segment which will lock it to only look back to the specified time.
This will look like this:

Applying this segment to the reports will give all the first time orders in the look back window, even those that aren't in your reporting window (the calendar set timeframe).

To only show the data in your reporting window, you will need to use a date dimension to filter the data to the specific window of time you want to report on. Example of this is below:

Thanks,

Dan

Adobe Analytics has no built-in “first purchase date” dimension, so you use sequential logic to mimic that behavior.

You can use the Orders event or Revenue depending on what defines a “purchase” in your implementation.

Always use Visitor container — since you’re working across time.

Hi @mezzome ,

I don’t work for an eCommerce company, so I don’t have access to typical metrics like orders, etc. However, I was able to simulate your use case with a similar scenario:

Since I have access to raw data feeds, it was easy for me to experiment and validate the results directly from the source.

My Use Case:

Visitors who were active in the last 7 days but not in the previous 3 months.
As of today (8th August), this means I’m looking for users who visited the site between 1st–7th August, but not between 1st May–31st July.

Approach & Experiments:

Experiment 1:

I created a segment to identify users who were active in both the last 7 days and the previous 3 months.
This allowed me to subtract that group from the total visitors during the last 7 days to isolate the desired audience.

Findings: This method worked as expected when validated against the raw data.

(For your case: it would give you the visitors who order in both the date ranges, and you can subtract these users from the total user who purchased in the later date range)

Experiment 2:

I created a segment for users who were active in the last 7 days, and explicitly "EXCLUDED" those who were active in the last 3 months.

Important Note: This method only works accurately when the date range selected on the analysis panel includes both timeframes (i.e., the last 7 days and the prior 3 months).

Regarding Your Segment (Based on the Screenshot You Shared):

From what I can see, using the "THEN" (sequential segment) logic might not be suitable for this use case.

That’s because you’re trying to filter users based on two separate date ranges, rather than defining a sequence of actions over time. Sequential logic is typically useful when you’re tracking event order, like users who did X and then Y.

Special Note:

Be cautious when using rolling date ranges within segments—these are calculated relative to the current date, which can affect your segment logic depending on when you run the report. so for this case, you can use the static date ranges in your segments.