A market is never one crowd. It's a room full of people who want slightly different things, will pay slightly different amounts, and reach for slightly different reasons. STP, segmentation, targeting, positioning, is the three-step habit for turning that messy room into a decision: who you serve, which of them you'll fight for, and what you want to own in their mind.

The quick version

  • Segment: split the market into groups whose needs, behaviour or value differ enough to matter.
  • Target: pick the segment (or few) you can serve better and more profitably than the alternatives, saying no to the rest is the point.
  • Position: decide the single place you want to occupy in that segment's mind, then make everything else line up behind it.
  • STP is the strategic front-end; the marketing mix is how you deliver on the choice. Skip STP and the mix has nothing to aim at.

The idea in depth

The intellectual root is older than most people assume. In 1956, Wendell R. Smith published "Product Differentiation and Market Segmentation as Alternative Marketing Strategies" in the Journal of Marketing, and drew the distinction that still anchors the field: you can try to bend demand toward a single offering (differentiation), or you can accept that demand is genuinely varied and build offerings that fit the variation (segmentation). Smith's move was to treat heterogeneity as a fact to be served, not a problem to be averaged away.

The sequence we now call STP, segment, then target, then position, was popularised by Philip Kotler through successive editions of Marketing Management from the late 1960s on, until it became the default first chapter of strategic marketing. Kotler's own framing is hard to improve on: he called the formula "the essence of strategic marketing." The key word is strategic. STP sits a level above the tactical "what price, what channel, what message" of the mix; it decides the questions the mix then answers.

So the move is: before you argue about ads, pricing or features, force the prior question, who is this for, and who is it deliberately not for? A useful test is to write the sentence "We are the obvious choice for ____ who need ____." If you can't fill both blanks without listing "everyone," you don't have a strategy yet; you have a hope.

Positioning is what you do to the mind, not the product

The third step carries the most weight and is the most misunderstood. The canonical statement comes from Al Ries and Jack Trout, whose 1981 book Positioning: The Battle for Your Mind argued that "positioning is not what you do to a product; it is what you do to the mind of the prospect." Their claim was that buyers cope with a noisy market by holding a short, ranked ladder of brands per category, and that your job is to own a word or a slot on that ladder, not to describe yourself accurately. The mind has limited shelf space; the brand that claims a clear position first is hard to dislodge.

"Positioning is not what you do to a product. Positioning is what you do to the mind of the prospect.", Ries & Trout, 1981

The practical upshot: stop trying to be liked for many things and pick one. The tool that forces it is the positioning statement, for [target segment], [brand] is the [frame of reference] that [single differentiating benefit], because [reason to believe]. The discipline isn't the sentence; it's everything you have to give up to make it true. If your positioning could equally describe two competitors, it isn't a position, it's a category description. A perceptual map, plotting rivals on the two attributes buyers actually trade off, is the fastest way to spot the empty, defensible corner.

flowchart TD
  M("The whole market: varied needs, varied value") --> S("Segment: group by need, behaviour or value")
  S --> T("Target: choose the segment you can win and serve profitably")
  T --> P("Position: own one clear place in that segment's mind")
  P --> Mix(["Marketing mix delivers on the position"])
  S -.- Drop(["Segments you choose NOT to serve"])
STP narrows from the whole market to a single defensible position; the mix then executes it. Leaders Loop

The honest limitation. STP is taught as settled, but two serious challenges deserve airtime. First, the theory-practice gap: Sally Dibb and Lyndon Simkin, tracking real segmentation projects, found that firms routinely fail to implement segmentation because of thin expertise, no clear process, and segments drawn around products or operations rather than customers (their "Implementation rules to bridge the theory/practice divide in market segmentation" catalogues the barriers). Elegant segments die in the org chart. Second, the evidence challenge from Byron Sharp and the Ehrenberg-Bass Institute: in How Brands Grow (2010), Sharp argues that brands mostly grow by reaching the whole category, building mental and physical availability, not by serving an ever-narrower niche, and that category buyers behave far more alike than segmentation lore suggests. He also reframes the goal from differentiation (a meaningful difference) to distinctiveness (being easy to recognise and recall). It's contested, Mark Ritson and others push back hard, but the caution stands: don't segment yourself into a niche too small to grow in, and don't confuse a tidy slide with a real behavioural difference.

A worked example

Take a regional gym chain, a composite, and the figures below are illustrative, stuck competing on price against budget franchises on one side and premium boutiques on the other. The instinct is "appeal to everyone who might exercise." That's not a target; it's the absence of one.

Segment. Instead of demographics alone, the team segments by need and behaviour: price-driven beginners who want cheap and simple; results-driven enthusiasts who want equipment and classes; and a third, under-served group, people in their 50s and 60s returning to exercise after a health scare, who find both the budget barn and the lycra boutique intimidating. Target. The chain can't out-cheap the franchises or out-glamour the boutiques, but no local rival is built around that third group. So it targets the returners, a segment it can serve better and that's willing to pay for reassurance. Position. Not "the cheapest" or "the hardest workout," but "the gym that makes coming back feel safe": staffed quiet hours, instructors trained for deconditioned bodies, a slower on-ramp. The illustrative payoff: lower price sensitivity, higher retention (this group churns less when it feels looked after), and word-of-mouth inside a tight community, none of it requiring the chain to win the price war it was losing. That's STP on one page: a deliberate no to two crowded segments bought a defensible yes in a third.

flowchart LR
  subgraph Before["Compete for everyone"]
    A("Match budget franchise on price") --> B(["Margins bleed, no clear identity"])
    A2("Add a few classes") --> B
  end
  subgraph After["STP move"]
    C("Segment: returners after a health scare") --> D("Target: the group no local rival owns")
    D --> E("Position: 'the gym that makes coming back feel safe'")
    E --> F(["Higher retention, lower price sensitivity"])
  end
Illustrative: chasing the whole market dilutes; choosing one under-served segment and owning a position concentrates value. Leaders Loop

Frequently asked questions

Isn't segmentation just demographics?

No, and treating it that way is the classic mistake. Age, income and location are easy to measure but often weak predictors of what people actually buy and why. The more useful bases are behavioural (how they use the category, how loyal they are) and needs-based (the job they're hiring you for). Demographics can describe a segment after you've found it; they're a poor way to find one.

How many segments should we target?

Fewer than you want to. The whole value of targeting is the discipline of saying no, which concentrates your limited money and attention where you can genuinely win. A focused challenger usually beats a diffuse generalist on its chosen ground. As you grow you can add segments deliberately, but bolting on a second target before you own the first tends to blur both.

What's the difference between positioning and a tagline?

A tagline is words you publish; positioning is the place you occupy in someone's head whether you publish anything or not. You can have a polished tagline and a muddy position, which is the worst case, confident noise. Get the position right and the messaging, pricing and product decisions tend to fall out of it. Get the tagline right first and you've decorated a vacancy.

Does STP still work in the age of mass digital reach?

It's exactly the live debate. Byron Sharp's research suggests big brands grow by reaching the whole category, not by narrowing, so for an established mass brand, "sophisticated mass marketing" may beat tight niche targeting. But for a challenger with a fraction of the budget, focus is still how you get a foothold. The honest answer: match the strategy to your position. Niche to enter; broaden, carefully, to grow.

What's the single most useful thing to take from STP?

The forced no. The exercise that earns its keep is writing down the segments you are deliberately not serving and why. Most strategies fail not from a bad target but from refusing to choose one, and a market that's "everyone" is a market you've conceded to whoever did choose.

Related in the Toolkit

STP is the front-end of the marketing system: it decides the choices the rest of the toolkit executes. It hands directly to the marketing mix, and a position only holds if it's grounded in a real customer need rather than a demographic guess.

Where to go next