Base Segmentation: Stop Treating Your Customer Base as One Audience

Base segmentation is the process of dividing your existing customer base into distinct groups based on shared characteristics, behaviours, or value, so you can allocate resources, tailor messaging, and prioritise growth efforts where they will have the most commercial impact. It is not the same as acquisition segmentation. It starts with people who already buy from you.

Done properly, it tells you which customers are worth growing, which are at risk of churning, and which are quietly costing you more to serve than they return. Most businesses skip this work entirely, or do a version of it once and never revisit it. That is where the commercial leakage starts.

Why Most Businesses Get This Wrong From the Start

When I was running the agency, we had a moment about three years in where we thought we understood our client base reasonably well. We had a sense of who our best clients were, which accounts were growing, and where the team’s time was going. Then we actually segmented it properly, by revenue, margin, growth trajectory, and service complexity, and the picture was completely different from the one we had in our heads.

Two clients who felt like anchor accounts were actually below-margin when you factored in the hours going in. One client we had mentally categorised as a mid-tier account was our most profitable relationship by a significant distance. We had been making resourcing and pricing decisions based on a narrative rather than data. That exercise changed how we ran the business.

This is not an unusual situation. Most businesses operate with an intuitive sense of their customer base rather than a structured one. The problem is that intuition tends to favour the loudest customers, the most recent wins, and the relationships that feel warm, not the ones that are actually driving commercial performance.

Base segmentation forces you to replace that intuition with something more reliable. And the first step is deciding what you are actually segmenting by.

What Are the Right Segmentation Variables for an Existing Base?

There is no universal answer here, but there are some variables that consistently produce more useful segments than others when you are working with an existing customer base rather than a prospect pool.

Value-based variables are usually the most commercially relevant starting point. This means looking at revenue per customer, margin per customer (where you can calculate it), purchase frequency, average order value, and lifetime value to date. These variables tell you what each segment is actually worth to the business, which is the foundation for any resourcing or investment decision.

Behavioural variables add a second layer of insight. How often does a customer buy? What do they buy? How do they engage with your communications, your service team, or your product? Customers who buy frequently but spend little per transaction have a different growth profile than customers who buy rarely but at high value. Both might sit in the same revenue band on a simple analysis and require completely different retention strategies.

Engagement and tenure variables matter more than most businesses acknowledge. A customer who has been with you for five years and whose spend has been flat is a different commercial risk than a customer who joined eighteen months ago and has grown spend consistently. Tenure combined with trajectory is a far better predictor of future value than either variable alone.

Demographic or firmographic variables, things like company size, sector, or geography in a B2B context, are useful for understanding which types of customers tend to cluster in which value segments. But they should be used to explain your segments, not to define them. Starting with demographics produces segments that feel neat but often have no commercial logic underneath them.

If you are thinking about this in the context of broader go-to-market planning, it is worth reading through the Go-To-Market and Growth Strategy hub for context on how segmentation connects to positioning, channel strategy, and commercial planning.

How Do You Actually Build the Segments?

The mechanics depend on the size and complexity of your customer base, but the underlying logic is consistent regardless of whether you are working with fifty clients or five hundred thousand customers.

Start with a clean data pull. You need transaction history, engagement data where available, tenure, and any service or support data that might indicate cost-to-serve. If your data is fragmented across systems, this is the stage that takes the longest, and it is worth taking seriously. Segmentation built on incomplete data produces segments that will mislead you.

Once you have the data, run a basic value distribution. Plot your customers by revenue or lifetime value and look at the shape of the curve. In most businesses, the distribution is highly skewed. A relatively small proportion of the customer base accounts for a disproportionate share of total revenue. Understanding exactly where that inflection point sits is one of the most useful things you can do with this analysis.

From there, you can apply a tiering framework. A simple three-tier model, high value, mid value, and low value, is often more actionable than a more complex segmentation in the first pass. The goal is not academic precision. It is a clear enough picture to make different decisions for different groups.

Add a behavioural overlay on top of the value tiers. Within your high-value segment, which customers are growing, stable, or declining in spend? Within your mid-value segment, which customers have the profile to move up? This is where segmentation starts producing genuinely useful strategic questions rather than just descriptive statistics.

One thing I have seen trip up a lot of marketing teams is the temptation to over-engineer this. I have sat in workshops where teams have spent weeks building twelve-variable segmentation models before they have done anything with a three-variable one. Start with something you can act on. Complexity is not the same as rigour.

What Should Each Segment Actually Tell You?

A segment is only useful if it produces a different decision than you would have made without it. This sounds obvious, but it is the test most segmentation exercises fail.

Your highest-value segment should tell you where to invest in retention and relationship depth. These are customers you cannot afford to lose and may be able to grow. The question for this segment is not how to market to them. It is how to make the relationship structurally harder to exit and how to identify expansion opportunities before a competitor does.

Your mid-value segment is usually where the most interesting growth opportunities sit. These customers have demonstrated enough commercial intent to buy from you, but have not reached their potential with you. The question here is what is stopping them from spending more, and whether the answer is a product gap, a relationship gap, or a communication gap. Those three causes require different responses.

Your lower-value segment requires a different kind of honesty. Some of these customers will grow. Some will never be commercially meaningful. Some are actively costing you more to serve than they return. The question is not how to market to all of them identically. It is which ones have growth potential worth investing in, which ones you can serve efficiently at low cost, and which ones you should be comfortable losing if they leave.

That last category is the one most businesses struggle with. There is a psychological resistance to acknowledging that some customers are not worth pursuing. But if you are managing a finite team and a finite budget, the decision to not invest in a segment is just as important as the decision to invest in one. Go-to-market execution has become harder for most teams, and part of the reason is that too many businesses try to serve every segment with equal intensity rather than making deliberate choices about where to focus.

How Does Base Segmentation Connect to Retention Strategy?

Retention strategy without segmentation is essentially a blunt instrument. You end up applying the same effort and the same messaging to customers who have completely different risk profiles and completely different commercial trajectories.

When I was at the agency and we finally had a clear picture of our client base, one of the first things we did was build a differentiated contact strategy. Our top-tier clients got proactive strategic reviews, not just quarterly check-ins on campaign performance. Our mid-tier clients got a structured growth conversation every six months, with a clear agenda focused on where we could add more value. Our lower-tier clients got excellent delivery and efficient communication, but we were honest with ourselves that the relationship investment had to be proportionate to the commercial return.

The result was not just better retention in the top tier. It was a clearer internal understanding of what we were trying to do with each client relationship, which made resourcing decisions considerably less political. When everyone can see the commercial logic behind how clients are tiered, it is easier to have honest conversations about where time should go.

Segmentation also improves your ability to identify churn risk early. A high-value customer whose engagement has dropped, whose purchase frequency has slowed, or whose service interactions have become more transactional is sending signals. If you are not looking at your base through a segmented lens, those signals get lost in aggregate data. Growth-focused teams tend to be much better at reading these early indicators because they have structured their customer data to surface them.

What About Segmentation for B2B Versus B2C?

The underlying logic is the same. The variables and the decision outputs are different.

In B2B, base segmentation typically involves a smaller number of accounts with higher individual value, which means the segmentation can be more granular and the actions more tailored. You might segment by account revenue, by product adoption depth, by stakeholder engagement level, or by contract renewal risk. The output is usually an account prioritisation framework that guides how your sales and customer success teams allocate their time.

In B2C, you are typically working with a much larger number of customers at lower individual value, which means the segmentation needs to be efficient enough to apply at scale. Behavioural segmentation, based on purchase frequency, recency, and category breadth, tends to be more useful here than value segmentation alone. The RFM model (recency, frequency, monetary value) is a well-established framework for this and remains one of the more practical tools available for consumer-facing businesses.

BCG’s work on go-to-market strategy highlights how customer segmentation shapes not just marketing tactics but the entire commercial model, including pricing, channel design, and service investment. That framing is worth keeping in mind. Segmentation is not a marketing department exercise. It is a business design input.

How Often Should You Revisit Your Segmentation?

More often than most businesses do, and with more rigour than most businesses apply when they do revisit it.

Customer behaviour changes. A segment that was stable and high-value two years ago may have shifted structurally, because the market changed, because a competitor entered, because your product evolved, or simply because the customers in that segment have moved through a lifecycle stage. Treating segmentation as a one-time exercise produces an increasingly inaccurate picture of your base over time.

A reasonable cadence for most businesses is a full segmentation review annually, with a lighter-touch check on key metrics quarterly. The annual review should look at whether the segment definitions still hold, whether the distribution has shifted, and whether the strategic decisions attached to each segment still make sense. The quarterly check should flag any significant movement, particularly in the high-value tier, where individual account changes can have material commercial impact.

One thing worth building into the process is a mechanism for capturing why customers move between segments. If a cluster of mid-value customers has dropped to low-value over the past twelve months, that is a signal worth investigating. It might point to a product issue, a service issue, a competitive threat, or a pricing problem. The segmentation data tells you that something has changed. The investigation tells you what to do about it.

Scaling any commercial operation requires the ability to make faster and better-informed decisions about where to focus. Regularly refreshed segmentation is one of the most reliable ways to keep those decisions grounded in what is actually happening in your customer base rather than what you assume is happening.

Turning Segmentation Into Action

The test of any segmentation is whether it changes what you do. Not what you plan to do, what you actually do differently as a result of having it.

That means attaching a clear strategic intent to each segment before you finish the analysis. For each segment, you should be able to answer: what is the commercial objective for this group, what is the primary risk, what is the primary opportunity, and what does the team need to do differently as a result?

If you cannot answer those questions for each segment, the segmentation is not finished yet. You have done the analysis but not the thinking.

I have judged enough marketing effectiveness work, including time with the Effie Awards, to know that the campaigns and programmes that produce measurable commercial results almost always start from a precise understanding of who they are trying to move and what they are trying to get them to do. Base segmentation is the foundation of that precision. Without it, you are making broad bets and hoping the right customers respond.

With it, you can make specific bets on specific groups with specific objectives, and measure whether those bets paid off. That is a fundamentally different way of running a marketing programme, and it is considerably more defensible when someone asks you to justify the budget.

If you want to think about how segmentation fits into a broader commercial growth framework, the Go-To-Market and Growth Strategy hub covers the connected disciplines, from ICP development and positioning through to channel strategy and performance measurement, in a way that keeps the commercial logic central throughout.

About the Author

Keith Lacy is a marketing strategist and former agency CEO with 20+ years of experience across agency leadership, performance marketing, and commercial strategy. He writes The Marketing Juice to cut through the noise and share what works.

Frequently Asked Questions