STP marketing: the segment you give up decides your strategy

STP marketing connects three decisions: segmentation, targeting and positioning. Segmentation groups buyers with similar needs, selection criteria or buying situations. Targeting determines which groups the company will serve, and positioning defines what those buyers should believe about the company.

Segmentation can stop at a simple list. Customers are grouped by industry, size and region, and the list goes into a slide deck, then into a drawer. Meanwhile, the offer, the website and the sales team continue addressing everyone in the same way.

STP marketing becomes strategy when the company decides which segments it will stop prioritising. Budget and sales time are limited, and a message aimed at everyone is harder to remember.

This article draws on our published audits of four Romanian companies. Three are full segmentation cases, covering a maker of special vehicle bodies, a medical laboratory and an aluminium façade manufacturer, with sources alongside the figures. The fourth, a dental clinic, illustrates brand architecture.

At the end you will find a 45-minute exercise you can run on your own customers from the past year.

Our guide to brand positioning explains how a company becomes known for one thing. This article covers the preceding decision: who the company addresses and who it deliberately leaves out of its message.

What does STP stand for in marketing?

STP stands for segmentation, targeting and positioning, three decisions made in sequence. Segmentation identifies groups in the market, targeting selects the groups the company will serve, and positioning defines what those buyers should believe about the company.

Market segmentation itself is older than the acronym. Wendell R. Smith is generally credited with introducing it to the marketing literature, in an article that treated product differentiation and market segmentation as alternative strategies. Since then, textbooks have collected dozens of criteria for splitting a market, and in many companies segmentation has turned into an exercise in classification.

Daniel Yankelovich and David Meer described that drift in their article “Rediscovering Market Segmentation”, published in Harvard Business Review in 2006.

The psychographic profiling that passes for segmentation these days is, in their words, “a mostly wasteful diversion” from its original purpose, which was to discover customers whose behaviour can be changed or whose needs are not being met.

The purpose of segmentation is to identify buyers whose choices you can influence. A table with five customer types may describe the market well, but it does not tell you where the company has a chance of winning.

In B2B, the classic criteria become firmographics: industry, revenue, headcount and region. These details are readily available in company filings, so they are often the starting point.

Two companies in the same industry, with the same revenue, may need your product or service to solve different problems. Firmographics tell you what kind of businesses your potential customers are, but these details alone do not explain what each company wants to achieve by buying from you. A list built solely on firmographic data therefore does not tell you which companies to target with your offer or which benefits matter to them.

Why is targeting the step that matters most?

Targeting matters most because it is the only step that forces a choice. The model is simple to explain, and its value depends on whether the company actually makes that choice.

The steps can be explained separately, but they work together in practice. Without targeting, segmentation describes the market without setting a priority. Targeting without positioning gives the company a list of suitable customers, whom it then addresses with the same message as its competitors.

Targeting is easy to skip because it demands an uncomfortable decision. Michael Porter wrote about that decision in his article “What Is Strategy?”, published in Harvard Business Review in 1996 in one short line. “The essence of strategy is choosing what not to do.”

In the same article, Porter applies the idea directly to customers. Choosing which customer groups, product varieties and needs to serve is fundamental to strategy, he writes, as is deciding which other customers or needs the company will not serve.

Targeting is the point where segmentation turns into strategy. If, at the end of an STP exercise, the company still serves every segment it found, it has done good market research and has not yet made a single decision.

Positioning depends on those earlier choices. A company trying to position itself for five segments at once often ends up with a statement broad enough to cover them all. In our view, that statement could usually appear on a competitor’s website too.

Our article on buyer persona vs ICP explains how we build the ideal customer profile used in targeting. It starts with the chosen segment and identifies the people involved, down to the person who signs off on the purchase.

What are the four types of market segmentation?

The four classic types of market segmentation are geographic, demographic, psychographic and behavioural. In B2B, demographic segmentation becomes firmographic. Our audits add two criteria that often change the outcome: the buyer’s role in the decision and the buying situation.

Geographic segmentation groups buyers by country, region or city. In B2B, it matters most when delivery, servicing or a local relationship determines the choice, as with equipment that requires a fast on-site response.

Demographic and firmographic segmentation uses age, income or life stage for consumers, and industry, revenue and headcount for companies. These details are the easiest to obtain and, in our judgement, rarely determine the purchase on their own.

Psychographic segmentation groups people by values, attitudes and lifestyle. In B2B, it distinguishes companies that prioritise safety from those that prioritise the lowest price, even when they are the same size and operate in the same industry.

Behavioural segmentation groups buyers by purchase frequency, the benefits they seek and their loyalty. It is usually closest to the buying decision because it starts with how customers behave, rather than how they are described.

The two criteria we add start from the same observation. The same capability is bought differently in different situations. In the Romturingia audit, for example, ambulances are sold to ambulance services and hospitals, while armoured vehicles are sold to defence institutions and security operators, so each mission has a different buyer and a different decision flow.

Our principle of the customer profile as a pattern follows this approach. We look for customers with a recurring need and for places where decision-makers already gather, extending segmentation beyond firmographics to geographic and psychographic criteria.

A group of decision-makers in a trade association or chamber of commerce can therefore be more useful for reaching buyers than a list of industry codes. A useful segment has a clear definition, identifiable buyers and a way to reach them. Industry codes help with the first two, but do not provide that route to the buyer.

Why does segmentation require trade-offs?

Every additional segment divides the communication budget, complicates the message and makes the company harder to remember. In our judgement, these effects take time to appear, which makes them difficult to trace back to the decision that caused them.

Keeping the company in the buyer’s memory is the hardest part. In 2021, Peter Weinberg and Jon Lombardo of the LinkedIn B2B Institute presented the findings of their research with Professor John Dawes of the Ehrenberg-Bass Institute in their Marketing Week article on the 95:5 rule.

According to Dawes, only 5% of B2B buyers are in-market right now, while 95% will not buy for months or even years. The authors add that marketers cannot push a buyer into the market. The buyer already has what you sell and will not need a newer version any time soon.

For segmentation, this means a company has to remain in its chosen buyers’ memories for years, until they are ready to buy. As the number of segments grows, each one encounters the company less often. Many people may have heard of it, but too few remember it when they need it.

This is where the principle we call cognitive ownership comes in. The end goal is to own the mental space of the category, so that when the need appears, the market comes to you on its own. We have written at length about how that space is built in our article on cognitive ownership.

Cognitive ownership is built on a chosen segment, because a single attribute cannot answer five different needs. The useful question at the end of segmentation is not how many segments exist, but how many you can afford to lose.

This needs a qualification. Choosing not to prioritise a segment concerns the message, communication budget and sales team’s time. The company can still fulfil orders from other segments, but those orders no longer determine what the website says or where the communication budget goes.

There is another possible cost, which we state as a hypothesis. The segment you stop prioritising may include longstanding customers or a product line the team is attached to. If it brings reliable orders or forms part of the company’s history, the decision may be postponed year after year.

Can you give me an example of STP marketing?

In the audits below, segmentation starts with public market data and leads to a recommendation that includes a trade-off, either stated explicitly or built into the proposed actions. The cases cover different industries.

In each audit, we define the segment boundaries ourselves. The slides identify them as our judgement, to be checked against the company’s internal data.

The figures come from ANAF, Romania’s tax authority, which publishes the annual balance sheets of every company, and from INS, the national statistics institute. Anyone can check them without us.

Romturingia, a maker of special vehicle bodies in Câmpulung Muscel, builds refrigerated vehicles, ambulances, fire-service vehicles, armoured vehicles and conversions of production vehicles. In our Romturingia audit, we shifted segmentation from industry to the vehicle’s mission, linking the ideal customer profile to the mission, operating requirements and promise.

The slide title sums up the method in two sentences: “We do not segment only by industry. We rebuild the situation in which the decision is made.” In an ambulance, the value of space is measured by time saved and the crew’s access to the patient. The same manufacturing capability therefore supports a different promise, with different evidence, for each mission.

The figures show why the segment boundary matters. The companies competing on exactly Romturingia’s segments had combined revenue of 559.9 million lei in 2025, added up by us from ANAF balance sheets, and Romturingia held 4.1% of that served market.

In the same year, Deltamed, the largest player, held 80.6% of the segment, and the audit links that share to a public contract for 1,200 ambulances. With Deltamed removed from the calculation entirely, the remaining market comes to 108.4 million lei, and Romturingia’s share rises to 21.4%.

Nothing changed in the factory between those two numbers. The same company holds 4.1% or 21.4%, depending on where you draw the segment boundary. The first figure says you are a small player, the second says you hold a fifth of the market left without its dominant leader, so the budget decisions that follow from each of them differ.

The audit also recommended a trade-off. Half of the non-branded traffic went to the Dacia Duster pick-up conversion, which the audit identifies as the lowest-margin line. We proposed reducing the company’s dependence on that traffic.

In its place, the audit proposed one page per mission, from ambulances and fire-service vehicles to light armoured vehicles, cash-in-transit and dangerous-goods transport.

Four missions, covering ambulances, fire service, armoured vehicles and cash-in-transit, accounted for 1,390 monthly searches with keyword difficulty between 0 and 1 in the Ahrefs data from 1 September 2026 cited in the audit. The company lacked pages that would give it visibility for those searches.

Recoltare Acasă, the service through which the Pathologic laboratory sends a home kit for cervical cancer screening, seems at first glance to address “women who get tested”. In our Recoltare Acasă audit we redefined the market as the women who do not get tested.

According to the latest European Health Interview Survey, from 2019, only 38.9% of women in Romania reported a cervical screening test within the survey’s reference period, and 61.1% did not report one. That put Romania last in the group of countries compared in the audit.

Applying the 61.1% share to the INS population registered by domicile on 1 January 2026 gives roughly 3.74 million women aged 25 to 64. Using the smaller resident population gives a lower result, so the figure should be read as an upper limit for the segment’s size.

The target segment was narrower and more precise. Gynaecology practices offer a route to these women through existing patient relationships, without relying on advertising. The audit counted 2,061 specialist obstetrician-gynaecologists, from INS data for 2024, and proposed a pilot with ten practices in Cluj county.

Choosing the segment also meant dropping an argument. The slide with the proposed actions was titled “Four moves, one of which is giving something up”. The audit proposed dropping convenience as the main selling point when addressing the practice.

For the doctor, what matters is that the patient stays with them, while her convenience remains a good argument only in communication aimed at the woman herself.

BIG ALUMINIUM, a manufacturer of aluminium façades and joinery with 30 years in business and 1,300 projects, addressed everyone at once on its website. In our BIG ALUMINIUM audit, we segmented buyers by three roles in the decision: the architect, the general contractor and the developer.

The architect wants the design intent respected, the contractor needs predictable site work, and the investor wants the building to retain its value over time. When the message addresses none of these concerns, price becomes the easiest criterion for all three to compare.

The audit proposed replacing the homepage title, “Innovation in construction”, a promise any construction company could make. It recommended using the company’s existing statement about lasting quality from its About page, supported by evidence.

The single message was replaced, in the same recommendation, by one page for each of the three decision-makers.

The right-hand column shows a common pattern across the three cases. Each trade-off makes room for more relevant evidence. Romturingia gains pages for its missions, Recoltare Acasă gains a selling point that matters to the doctor, and BIG ALUMINIUM gains messages for three decision-makers.

How do you segment by role in the decision, not only by industry?

Segmenting by decision-making role distinguishes buyers within a customer company by what they need and what they risk if the choice goes wrong. In B2B, several people are involved in the same purchase, and each can block it for a different reason.

Gartner published on 7 May 2025 a press release on conflict in B2B buying groups, based on a survey of 632 B2B buyers. The survey ran in August and September 2024. According to the study, 74% of B2B buyer teams show “unhealthy conflict” during the buying decision.

Buying groups in 2025 range from five to 16 people across as many as four functions, according to the same Gartner release. Groups that reach consensus are 2.5 times more likely to report a high-quality deal.

The third figure is counterintuitive. The same Gartner release shows that, in 2025, individually tailored content had a 59% negative impact on buying group consensus, while content relevant to the group as a whole raised consensus by 20%.

Gartner explains the effect through confirmation bias, which reinforces individual perspectives and makes stakeholders less likely to embrace a unified direction.

Our conclusion goes beyond the release. Segmentation by role should help the group reach a shared decision. The group receives a coherent message about the common problem and the company’s positioning, with evidence relevant to each role’s responsibilities, from technical calculations to costs over time. Adapting the evidence to each role should not mean personalised messages that pull decision-makers apart.

The two audits above follow this approach. At BIG ALUMINIUM, we proposed a single positioning, summed up in the title “One system. Three decision-makers. One responsibility.” At Romturingia, the same central positioning was supported by different evidence for the operational user, the technical validator and the economic decision-maker.

A second principle from our methodology applies when segments have conflicting expectations of the same brand. The umbrella brand supports separate product brands, each with its own defining attribute, so that the promise to one segment does not weaken the promise to another.

Romturingia did not need separate brands because its missions do not conflict. The audit proposed five entry pages for the same brand, built around one association: a vehicle made for its mission. The test is whether the promise made to one segment puts another segment off.

At Crystal Dental, a dental clinic, the answer was yes. Its premium aesthetics brand, White, asked patients to pay a premium price, while the main brand encouraged the same patients to expect promotions and discounted subscriptions.

That is why our Crystal Dental audit proposed an architecture in which the umbrella brand lends credibility to the product brands, each associated with a distinct patient problem.

Where does poor segmentation show up?

Poor segmentation affects all four layers of our methodology, from L1, performance marketing, to L4, positioning and category, with the AI Brain at L3. The symptoms usually appear in campaign costs at the bottom of the model, while the cause lies at the top: the company has not chosen a segment.

Read the figure from the bottom up. The red layer at the top shows where the problem starts.

L1 · Performance marketing. A campaign targeted by industry and company size pays for clicks from everyone who fits on paper. Without a chosen segment, the ad’s message stays generic, and the budget buys the attention of the 95% who will not buy soon, without leaving them anything to remember.

L2 · Revenue and sales process. A salesperson who does not know which segment the company has chosen gives every request for proposal the same priority. In our judgement, the team may end up spending most of its time with the customers who negotiate hardest, without a compelling selling point beyond price.

L3 · AI Brain, orchestration. An AI assistant asked about suppliers for a specific mission looks for pages that answer exactly that question. At Romturingia, the audit showed that Deltamed led the category with 21 reviews, because it had pages AI could read and cite, even though Romturingia had 145 reviews.

L4 · Positioning and category. Choosing a central attribute depends on a clear need within the segment. A company that has not chosen its segment ends up looking for an attribute that will please everyone. The result is the statement every competitor uses, described in our article on differentiation strategy and the competitor test.

How do you do an STP analysis?

You can conduct an initial STP analysis in about 45 minutes, using your customer list from the last 12 months and two or three public financial filings from competitors. The result is a shortlist of chosen segments and, just as importantly, a written list of what the company will stop prioritising.

In our judgement, the red link is the easiest to skip, because the first three steps feel like enough.

1. Group customers by buying situation. Take your customers from the last 12 months and group them by the reason they bought, and only then by industry or size. A good group can be described in one sentence about the customer’s problem.
2. Note who decides in each group. Write down the roles involved in the decision, what each person risks if the choice goes wrong, and who signs off. If you do not know, review the last five proposals you won and the last five you lost.
3. Calculate your share within an explicitly defined segment. Add up the revenue of your actual competitors from public filings and calculate your share. The Romturingia case shows why you must document who is included and who is excluded: removing a single dominant player from the calculation moved the share from 4.1% to 21.4%.
4. List what you will stop prioritising. For the segments you do not choose, note which campaigns you will stop funding, which pages are no longer a priority and which generic statements you will remove. If this list is empty, you have not made a choice yet.
5. Check the positioning for the chosen segment. Write one sentence about what those buyers should believe about the company and run it through the competitor test. If a competitor could use it word for word, go back to step two, where the supporting evidence is usually missing.

Step four may meet resistance because the segment you stop prioritising brings in real revenue. The decision concerns new investment in that segment; it does not require rejecting existing orders. The company can continue serving it, but no longer lets it determine the message and budget.

If you want an outside view of your company, you can request a strategic diagnostic, a few qualifying questions followed by a conversation about your own numbers.

For STP marketing to lead to a decision, someone in leadership must write down which segments the company will prioritise and what it will give up. The team can then build its positioning around those buyers and concentrate its budget on reaching them.

Frequently asked questions

What is market segmentation?

Market segmentation groups buyers with similar needs, selection criteria or buying situations. Its purpose is to help the company decide which groups to serve. A segment is useful only if its buyers share a selection criterion the company can meet better than its competitors.

What does STP stand for in marketing?

STP stands for segmentation, targeting and positioning. Segmentation describes the groups in a market, targeting picks the groups the company will serve, and positioning sets what those groups should believe about the company. The decisive step is targeting, because that is where the company also chooses which groups it gives up.

What are the four types of market segmentation?

The four classic types are geographic, demographic, psychographic and behavioural. In B2B, demographic segmentation becomes firmographic, meaning industry, revenue and headcount. In our audits we add the buyer’s role in the decision and the buying situation, which usually change the outcome.

How do you do an STP analysis?

Group your customers from the last 12 months by buying situation and note who decides in each group. Then calculate your share of each segment from public filings, list what you will stop prioritising, and check a positioning statement with the competitor test. The exercise takes about 45 minutes.

What is a good example of B2B market segmentation?

One example is Romturingia, a Romanian maker of special vehicle bodies, where our audit moved segmentation from product to the vehicle’s mission. With Deltamed, the largest player, removed from the calculation entirely, the company’s share of the remaining market rises from 4.1% to 21.4%, based on 2025 balance sheets.