How to segment your store's customers

Not all of your store's customers behave the same way. Some come in every week, others a few times a year. Some spend a lot per visit, others buy a little but often. Some have just made their first purchase, while some former regulars are starting to drift away.

Customer segmentation is the practice of grouping customers who share characteristics or behaviors so you can better understand your customer base and tailor your actions to each group.

The goal isn't to create as many segments as possible. It's to build groups meaningful enough to lead to a different decision or action.

Why segment your customers?

Taken as a whole, a customer base tells you very little you can act on. Say you have 5,000 identified customers. Knowing that their overall average basket is €42 doesn't tell you:

• which ones buy regularly;
• which ones are buying for the first time;
• which ones spend the most;
• which ones no longer come back;
• which product categories interest certain groups;
• which ones seem to be drifting away.

Segmentation turns an undifferentiated mass of customers into groups whose behavior you can analyze separately. You can then use those groups to tailor a message, an offer or a follow-up, or simply to refine your customer base analysis.

Which criteria should you use for customer segmentation?

There's no one-size-fits-all criterion. The right choice depends on what you're trying to understand or do. These are the most common ones.

Recency

How long has it been since the customer's last purchase? Recency separates recently active customers from those whose last purchase is getting further back. But it has to be read in light of your purchase cycle: a customer who hasn't come in for 30 days may be unusual for a store people visit every week, and perfectly normal in a sector where purchases are several months apart.

Purchase frequency

How many times does the customer buy over a given period? Frequency separates regular customers from occasional buyers. Again, context matters: six purchases a year can be a high frequency for one business and a very low one for another.

Amount spent (monetary value)

How much revenue has the customer generated over the period you're looking at? Always specify the period: a ranking based on the last three months won't necessarily match one based on the last two years.

Average basket (average transaction value)

How much does the customer spend per transaction, on average? Two customers can each generate €1,000 in annual revenue with completely different behaviors:

• 20 purchases of €50;
• 2 purchases of €500.

The annual total is the same, but the customer relationship is not. See also: how to increase your average basket.

Products or categories purchased

Purchase history also lets you build segments based on demonstrated interests. For example:

• regular buyers of a given category;
• customers who have bought a particular brand before;
• customers who have never bought from a given product line;
• buyers of a product that needs to be replaced periodically.

This kind of segmentation is especially useful to avoid sending an offer to customers it isn't relevant to.

Channel or location

If your data allows it, it can also be worth breaking down behavior by:

• brick-and-mortar store;
• online store;
• click & collect;
• location visited.

For a business with several locations, this dimension helps you spot, among other things, customers who shop at more than one store.

Combining several segmentation criteria

The most useful segments rarely rely on a single criterion. Take this one:

Customers with no purchase in 90 days

This segment can include a former regular just as easily as a customer who made a single purchase three months ago. They probably don't matter equally to the business or call for the same action. So you can refine it:

Customers who made at least 6 purchases over the last 12 months and haven't bought anything in the last 90 days.

Then, if needed:

Customers who made at least 6 purchases over the last 12 months, haven't bought anything in the last 90 days and spent more than €500 over the period.

Each additional criterion shrinks the group but sharpens the segment. Don't go too far the other way, though: a segment so narrow that it contains almost no one loses its practical value.

The RFM method: recency, frequency, monetary value

A classic segmentation method combines three dimensions:

• Recency: the date of the last purchase;
• Frequency: the number of purchases;
• Monetary value: the amount spent.

That's the RFM method. Each customer is usually scored on all three dimensions, which makes it possible to group profiles with similar behavior. For example, one customer might be:

• a very recent buyer;
• a very frequent buyer;
• a high spender.

Another might have spent a lot in the past but not bought anything recently. RFM helps you tell apart behaviors that revenue alone wouldn't reveal.

That said, you don't need RFM for every segmentation. For a campaign about a specific product category, the products a customer has bought can be far more relevant than a full RFM score. The method should always serve the need, not the other way around. See also: how to identify your best customers.

Which customer segments can you create?

There's no universal list of segments that works for every business. But a few groups make a good starting point:

Customer segments aren't set in stone

A segment isn't necessarily permanent. A new customer can become a regular. A regular customer can start to drift away. An inactive customer can come back. Behavior changes with every purchase.

So if you use segments to drive recurring actions, they need to be recalculated regularly from the latest data. This matters most for segments based on recency or frequency.

How to use your customer segments

Segmentation is only useful if it actually changes what you do.

That makes it a tool for managing your customer relationships, too.

Watch out for segments that are too broad

“Inactive customers,” “good customers” or “loyal customers” may sound specific enough. They aren't always. A segment should be defined by measurable criteria. For example:

Loyal customers

is ambiguous. By contrast:

Customers who made at least 10 purchases over the last 12 months

describes a rule you can verify. That doesn't mean ten purchases is the right definition of loyalty for every business: the threshold depends on yours. Good segmentation relies on explicit, easy-to-understand criteria.

Watch out for segments that are too small, too

Adding criteria improves precision, but it also shrinks the group. If you ask for:

customers who bought at least 12 times, spent more than €1,000, bought a specific category, visited a given store and haven't bought anything for exactly 60 days,

the resulting segment may be too narrow to be useful. You need to strike a balance between precise targeting and a group large enough to work with.

How do you know if your customer segmentation is working?

Measure what happens after each action: creating segments isn't an end in itself. Depending on your goal, you can look at:

• the return rate of the customers you followed up with;
• the revenue generated by the segment;
• the number of customers who made a new purchase;
• how many customers used an offer;
• the change in purchase frequency.

This also tells you whether the criteria you chose were actually relevant. A highly sophisticated segment is no better than a simple one if it doesn't deliver better results.

Customer segmentation with AI

Creating a segment in a traditional tool usually means setting up several filters. With an AI that can query your business data, you can express the logic directly in plain language. For example:

“Identify the customers who haven't bought anything in the last 90 days.”

Then:

“Keep only those who made at least 6 purchases over the previous 12 months.”

Then:

“Of those, keep the ones who spent more than €100 a month on average.”

The benefit is that you can build and refine the segment step by step without rewriting the whole query each time. The quality of the result still depends on the data available and on how precisely the criteria are defined.

How Gillia helps you segment your customers

Gillia, an AI agent for independent businesses, lets you query your customer data in plain language to build groups that match a specific need. For example:

“Which customers regularly buy this product category?”

You can then narrow it down:

“Keep only those who bought at least three times in the last six months.”

Then use this group as the basis for a customer loyalty action or a targeted message. The benefit is that you go step by step from intent to segment to action, with an approval step before any external action is taken. When you need to check the same segment regularly, the query can also become a recurring routine.

This is exactly where Gillia's role as an AI agent comes into its own.

Customer segmentation for multi-location businesses

With several stores, choose your level of analysis carefully. A customer may shop at more than one location. If customer data is consolidated at the brand level, you can analyze their behavior as a whole:

• total frequency;
• total spend;
• stores visited;
• categories purchased across the network.

It can also make sense to segment by location to answer a local question. The two approaches aren't mutually exclusive: the segment's scope just needs to match the decision you're trying to make (see multi-store management). To go further, explore Gillia's use cases for getting to know your customers: create, edit or target a customer group or prepare a targeted action for a selection of customers.

Keep your segmentation useful

The best segmentation isn't the most complex one or the one with the most criteria. It's the one that lets you clearly answer three questions:

• Who belongs to the group?
• Why is this group different from the others?
• Which decision or action will you adapt based on this distinction?

If you can't answer the third question, the segment is probably useless. Customer segmentation becomes truly valuable when it turns your data into groups that are understandable, measurable and actionable.