It is often very valuable to base the personalization of customer marketing on the particular products or type of products, that customers have purchased in the past. Targeting cross-sell/up-sell offers, incentives, and recommendations based on historical purchase patterns has proven itself as a powerful way to increase customer spending and loyalty.

You can use this approach in Optimove by defining Target Groups using the Purchase History selection category. The selection criteria in this category consist of three elements:

Selection Element	Examples
Type of history	Customers who purchased (with activity) Customers who didn’t purchase (with no activity)
Product, product category, or product attribute	Department = Menswear Brand = Levis Product = Levis 501 Stonewash Color = Blue
Timeframe	On 10/12/2014 During the past 60 days This week/month/quarter/year Last calendar week/month/quarter/year In this date range

Additional Timeframes

When using "any campaign", the only available timeframe is "during the previous" X days.

When using the additional activity options ("while a member of a target group" or "One or more actions"), the available timeframes are:

During the previous / exactly (number of days) ago - customers who were targeted over the past X number of days or exactly X number of days ago.
This week (to date) – customers who were targeted during the week (Mon-Sunday). For example, if today is Wednesday, this timeframe will include Monday’s and Tuesday’s customers.
This month (to date) - customers who were targeted during this month.
This quarter (to date) - customers who were targeted during the current quarter.
This year (to date) - customers who were targeted during this year.
Last calendar week – customers who were targeted last week. For example, if today is Tuesday, August 29th, "last week" would be Monday the 16th – Sunday 22nd.
Last calendar month - customers who were targeted last month.
Last calendar quarter - customers who were targeted during the previous quarter.
Last calendar year - customers who were targeted during last year.
On this date - customers who were targeted on a specific date. -In this date range - customers who were targeted on a specific date range.

What's the difference between using two separate Purchase History criteria and using "And" within one single Purchase History criteria?

Let's use two examples to demonstrate the two different targeting options.

Option 1:

In the example below, we used two separate Purchase History criteria in one target group:

Here we are targeting customers who spent a total amount of $100 and purchased Accessories (not necessarily only on accessories). Meaning, there are customers who spent $100 on accessories, those that spent $30 on accessories and $5 on other product categories, those that spent $50 on accessories and $75 on other categories, and so forth. The customers' split would look like this:

Option 2:

Using "And" within one single Purchase History criteria (using the "Add Attribute" option) allows you to create conditions that hold two or more attributes. In the example below, we target customers who spent a total of $100 only on Accessories.

The customers' split would look like this:

The difference between the two options is, that in Option 1 we target a wider variety of customers - those who purchased for $100 and also purchased accessories, whereas in Option 2 we target customers that bought accessories for $100.

In Option 2 ("And" within one criterion) we expect fewer customers than Option 1 (two separate criteria) since Option 2's customers are included in Option 1 but option 1's customers aren't necessarily included in option 2.

Selecting customers who purchased

To select customers who purchased follow the steps below:

Go to Purchase History in the Selection Criteria navigation bar.
Select customers ”with activity.”
Use Attributes, Conditions, and Value fields to select the product, product category, or product attribute you want to use to segment your customers. For example, Category equals Bags.
Define the timeline for which the conditions should be met. For example, during the previous 14 days.
You should now have similar selection criteria to the one below. All that is left is to click “Add.”

Selecting customers who did not purchase

To select customers who did not purchase follow the steps below:

Go to Purchase History in the Selection Criteria navigation bar.
Select customers ”with no activity.”
Use Attributes, Conditions, and Value fields to select the product, product category, or product attribute you want to use to segment your customers. For example, Category equals Bags.
Define the timeline for which the conditions should be met. For example, during the previous 14 days.
You should now have similar selection criteria to the one below. All that is left is to click “Add.”

Example use cases

Back-to-School Campaign

An online apparel retailer wants to send a back-to-school incentive offer to all customers who spent more than $50 on clothing or accessory items, in the Kid's department, during the past six months:

Early Adopter Bonus Campaign

An online electronics retailer wants to entice "Apple Fans" with a fantastic offer for an expensive iPhone accessory. He decides to do so by targeting customers who purchased the iPhone 11 on launch day:

Follow-up Coupon Campaign

On October 1, a sporting goods retailer advertised that customers purchasing three or more North Face products between that date and October 31 will receive coupons for free The North Face items during the month of November. Now, it is time to send out the coupons to all qualifying customers:

Cross-sell Campaign

An online retailer wants to increase sales of party dresses leading up to New Year’s Eve. They chose to target all customers who bought women’s clothing in the past year (not shown below), but who did not purchase any items from the Party Dresses department during this period:

What’s the difference between using multiple attributes in one condition vs. separate conditions?

1. Multiple attributes in ONE condition

You’re asking the system to find customers who match ALL the attribute values at the same time, in a single transaction.

Example:

"Customers who placed a bet of over $100 AND placed it on mobile."

This means the same bet needs to meet all 2 conditions.
The result is a smaller, more specific group — they must meet all two in one activity.

purchase history example 1.png

2. Same attributes in SEPARATE conditions

Here, the system looks for customers who matched any of the conditions, across different transactions.

Example:

"Customers who placed a bet over $100
OR placed any bet from mobile"

This means the customer could have met each of these in different bets, not all in one.
The result is a broader group — they just need to match one or more of the conditions, even if it’s in separate actions.

purchase history ex 2.png

When to use each?

Use ONE condition with multiple attributes when you want customers who did all those things in the same activity.
Use SEPARATE conditions when you're looking for customers who did any of those things, even if across different activities or times.

Purchase History Update Frequency

Purchase History updates daily. Read more about the update frequency here.