Market Segmentation Analysis: Stop Targeting Everyone

Market segmentation analysis is the process of dividing a broad target market into distinct subgroups based on shared characteristics, behaviours, or needs, so you can allocate budget, messaging, and product focus where they will actually produce a return. Done well, it is one of the highest-leverage activities in marketing strategy. Done poorly, it produces slide decks full of personas that nobody uses.

Most segmentation work fails not because the data is wrong, but because the segments are never connected to commercial decisions. This article covers how to do the analysis in a way that changes what you spend, what you say, and who you prioritise.

Why Most Segmentation Work Produces Nothing Useful

I have sat in more segmentation presentations than I can count. The pattern is almost always the same: a research agency delivers four to six named personas, each with a stock photo, a life story, and a set of values that reads like a horoscope. The marketing team nods, the slides go into a shared drive, and six months later nobody can remember what the segments were called.

The problem is not the research. The problem is that the output was designed to be presented, not acted on. Segmentation analysis has commercial value only when it answers a specific question: which customers should we prioritise, and why? Everything else is decoration.

When I was running an agency and we grew the team from around 20 people to close to 100, one of the things that changed was how we approached client segmentation work. Early on, we built segments because clients expected them. Later, we built segments because they were the foundation of every media, messaging, and product decision. The difference in output quality was significant, and it came entirely from changing the brief at the start of the process.

If you want segmentation to produce something usable, start by defining the decision it needs to inform. Budget allocation across channels? Message hierarchy for a campaign? Prioritisation of product features? The decision shapes the analysis, not the other way around.

What Are the Core Segmentation Dimensions?

There are four standard dimensions used in market segmentation analysis. Each has different strengths depending on what you are trying to do.

Demographic segmentation divides markets by age, gender, income, education, household composition, and similar variables. It is the most commonly used approach because the data is easy to obtain and simple to communicate. It is also the least predictive of actual purchase behaviour in most categories. Knowing that your customer is a 35-to-44-year-old homeowner tells you almost nothing about what they will buy or why.

Psychographic segmentation groups people by attitudes, values, interests, and lifestyle. This is more useful for brand positioning and creative strategy than for media targeting, because psychographic data is harder to match to addressable audiences at scale. It is genuinely valuable in categories where emotional drivers dominate purchase decisions, but it requires primary research to do properly.

Behavioural segmentation is based on what people actually do: purchase history, product usage, search behaviour, engagement patterns, loyalty status, and response to previous marketing. In my experience, this is the most commercially useful dimension for performance marketing, because it reflects revealed preference rather than stated preference. What people say they will do and what they actually do are frequently different things.

Geographic segmentation divides markets by location, from country level down to postcode. It is essential for businesses with physical distribution constraints, regional pricing differences, or location-dependent demand patterns. It is also underused in digital marketing, where the assumption that geography does not matter often turns out to be wrong.

Most effective segmentation combines at least two of these dimensions. A segment defined only by demographics is a blunt instrument. A segment defined by demographics, purchase behaviour, and geographic concentration starts to become something you can actually target.

If you are building out a broader research and intelligence capability, the Market Research and Competitive Intelligence hub covers the tools and frameworks that support this kind of work, from audience analysis through to competitor monitoring.

How Do You Identify Which Segments Are Worth Pursuing?

Segment identification is the analytical part of the process. Segment prioritisation is the commercial part. They are different skills, and confusing them is where most segmentation projects go wrong.

To identify segments, you need data that captures variation in your market. That data can come from several sources: customer transaction records, CRM data, first-party survey research, third-party audience panels, or a combination. The goal is to find natural groupings in the data, people who behave similarly to each other and differently from other groups.

Statistical clustering methods such as k-means clustering or latent class analysis are commonly used for this. If you do not have a data science resource in-house, most CRM platforms and some research agencies can run this analysis on your customer file. The output is a set of candidate segments with profiles describing their characteristics.

Prioritisation requires a different set of criteria. The four I use most consistently are:

Size. Is the segment large enough to justify dedicated investment? A segment that represents 2% of the total addressable market may still be commercially significant if the unit economics are strong, but you need to know the number.

Profitability. What is the average revenue per customer, margin contribution, and lifetime value within this segment? Segment size and segment profitability are different numbers. I have seen businesses spend years chasing volume segments that were structurally unprofitable, while ignoring smaller segments with far better economics.

Reachability. Can you actually get in front of this segment through channels you can afford and measure? A segment that is well-defined in theory but has no viable acquisition path in practice is not a commercial opportunity. This is where a lot of persona work falls apart. The segment exists in the research, but there is no media product, channel, or tactic that can reach it efficiently.

Strategic fit. Does this segment align with where the business is going? Sometimes the most profitable segment today is not the right one to build around if the product roadmap, brand positioning, or competitive dynamics are moving in a different direction.

What Data Sources Should You Use for Segmentation Analysis?

The quality of your segmentation is constrained by the quality of your data. That sounds obvious, but it is worth being specific about what good data looks like for this purpose.

First-party transactional data is the most reliable starting point if you have it. Purchase history, product usage patterns, and customer service interactions all reflect actual behaviour. If you have a CRM with reasonable data hygiene, start there before going anywhere else.

Survey research adds the attitudinal and psychographic layer that transaction data cannot provide. A well-designed survey of your existing customer base, combined with a sample of non-customers in your category, can reveal the motivations and barriers that explain the behavioural patterns in your transactional data. The two sources are complementary, not interchangeable.

Third-party audience data, available through platforms like Semrush’s audience analysis tools or data management platforms, can extend your view beyond your existing customer base to the broader market. This is particularly useful when you are trying to understand segments you are not currently reaching, or when you want to size a market opportunity before committing to it.

Search data is an underused source for segmentation work. The language people use when searching, the questions they ask, and the products they compare reveal a great deal about their needs and decision-making process. I have used keyword research to identify distinct need states within a single category that were not visible in any other data source. Understanding how search intent fragments across different queries can be a useful proxy for how your market actually segments.

One caution: be careful about over-relying on any single data source. Each source has its own biases and gaps. Transaction data tells you about buyers, not non-buyers. Survey data reflects what people say, not what they do. Search data captures active consideration, not passive preference. The best segmentation work triangulates across multiple sources and is honest about where the evidence is thin.

How Do You Translate Segments into Marketing Decisions?

This is the step that most segmentation projects skip. The analysis produces segments. The strategy work turns those segments into decisions about where to spend, what to say, and how to measure success.

For each priority segment, you need to define four things: the acquisition strategy, the message, the channel mix, and the success metric.

The acquisition strategy answers how you will reach people in this segment who are not yet customers. This requires knowing where they spend time, what triggers their consideration of your category, and what the competitive dynamics look like in the channels you plan to use.

The message should reflect the specific need state or motivation that defines the segment, not a generic brand message applied uniformly. Early in my career at lastminute.com, I ran a paid search campaign for a music festival that generated six figures of revenue within roughly a day. The reason it worked was not sophisticated targeting. It was that the message was precisely matched to what people in that moment were looking for. Specificity converts. Broad brand messages rarely do in performance channels.

The channel mix should be driven by where the segment is reachable, not by what channels you are comfortable with. Different segments often require materially different channel strategies. A segment of high-intent, category-aware buyers may be best reached through paid search. A segment of latent demand, people who have the need but are not yet actively searching, may require social, display, or content-led approaches. Matching your conversion strategy to the intent level of your audience is one of the more reliable ways to improve campaign efficiency.

The success metric needs to be segment-specific. If you are targeting a high-value segment with a long purchase cycle, optimising for short-term conversion rate will give you a misleading picture of performance. Define what success looks like for each segment before you start spending, not after.

How Often Should You Revisit Your Segmentation?

Segmentation is not a one-time exercise. Markets change, customer behaviour evolves, and the competitive landscape shifts. A segmentation model that was accurate 18 months ago may no longer reflect the audience you are actually reaching.

The practical question is how frequently to run a full refresh versus monitoring for drift. A full segmentation refresh typically makes sense every 18 to 24 months, or when there has been a significant change in the market: a new competitor, a category disruption, a change in your product or pricing, or a material shift in customer acquisition patterns.

Between full refreshes, you should be monitoring for signals that your segments are drifting. Changes in the demographic or behavioural profile of new customers, shifts in which channels are driving acquisition, changes in customer lifetime value by cohort, these are all indicators that the underlying segmentation may need to be updated.

I judged the Effie Awards for several years, and one pattern I noticed in the winning entries was that the strongest campaigns were built on segmentation work that was genuinely current. The teams that won were not working from a three-year-old audience model. They had invested in understanding how their market had shifted and had adjusted their strategy accordingly. That is not glamorous work, but it is what separates campaigns that perform from campaigns that look good in a presentation.

What Are the Most Common Segmentation Mistakes?

A few patterns come up consistently across the segmentation projects I have been involved in or reviewed.

Segments that are too broad to be actionable. “Women aged 25 to 54 with an interest in health” is not a segment. It is a demographic slice that contains dozens of different need states, motivations, and purchase behaviours. Broad segments produce generic strategies, which produce average results.

Segments defined by aspiration rather than evidence. There is a common tendency to build segments around the customers you want rather than the customers you have or can realistically acquire. This produces strategies that are disconnected from commercial reality. Start with the evidence, then test whether the aspirational segment is actually reachable.

Ignoring the competitive dimension. A segment may be attractive in absolute terms but heavily contested by better-resourced competitors. Segmentation analysis should include an assessment of competitive intensity by segment. Choosing a segment where you have a structural advantage, whether through product, price, distribution, or brand, is more commercially rational than chasing the most attractive segment regardless of competitive dynamics.

Failing to connect segments to media buying. This is the most operationally damaging mistake. A segment that cannot be translated into a targetable audience in the channels you use is not a viable commercial segment. Before you commit to a segmentation strategy, verify that your priority segments can actually be reached through your media plan. Audience receptivity to specific channels and formats varies significantly across segments and should be factored into your channel strategy from the start.

Treating segmentation as a marketing function rather than a business function. The most valuable segmentation work I have seen was done at a business level, informing product development, pricing, distribution, and customer service decisions, not just marketing. When segmentation is confined to the marketing team, it tends to produce campaign-level outputs. When it is owned at a business level, it produces structural competitive advantages.

For more on building the research and intelligence capability that supports this kind of work, the Market Research and Competitive Intelligence hub covers the full range of tools, frameworks, and methodologies worth knowing.

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