Markets Decide

Segmentation, Targeting, Positioning

Segmentation, targeting, and positioning divide a market into segments, choose which ones to serve, and define the place the offer should hold in those customers' minds.

The questionWhich customers should we build this offer for, which should we leave to others, and what should the ones we choose think of us compared with the alternatives?

STP. Also called target marketing or the STP model.

Process · Wendell R. Smith; Al Ries and Jack Trout; Philip Kotler, 1956 to 1981

1
Segment Split the market by how groups buy
2
Target Choose the groups to serve, and drop the rest
3
Position Set the place to hold against rivals

The three decisions in order, each one using the output of the step before.

When to use it

A client sells one offer to everyone in a category, and win rates and margins vary widely by customer type with nobody able to say why.
A launch or market entry needs a decision on whom to sell to first, and the sales team wants to chase every inbound lead.
Marketing owns a segmentation study from a research agency that no one uses to set budgets or sales coverage.
Two brands or product lines in the same portfolio are chasing the same customers with similar messages.

When to avoid it

The market is a few dozen named accounts. Planning each account does more than grouping them into segments.
Nobody yet understands why customers buy or switch. Run Jobs to Be Done interviews first, because segments built on demographics alone will miss the reason.
The company cannot vary its product, price, channel, or message by group. Segments it cannot serve differently change nothing it does.
The only question is how big the market is. A sizing model answers that without the targeting and positioning work.

Overview

STP is a three-step sequence for deciding whom to sell to. Segmentation splits a market into groups that respond differently to an offer. Targeting chooses the groups worth serving, given their size and margin and what the company can deliver. Positioning defines how the offer should be seen by those groups, compared with the alternatives they already know. The choice it supports is where to concentrate product, sales, and marketing effort, and where to stop spending.

Exclusion is the hard part. A segmentation that leaves every group in the target list changes no budget or sales plan. Segments also have to be findable: a needs-based cluster that sales cannot recognize in a lead or reach through a channel stays on paper.

The three decisions

Segmentation

Split the market into groups that respond differently to the offer. The usual bases are geographic, demographic (firmographic in B2B, such as industry and headcount), psychographic (attitudes, values, lifestyle), and behavioral (usage rate, occasion, benefit sought, loyalty). Behavioral and needs-based bases tend to predict buying better, and demographic ones are easier to find in data and reach with media, so working segmentations usually define on the first and describe with the second. A usable segment is measurable, substantial enough to be worth serving, accessible through channels the company can use, differentiable in how it responds, and actionable with the resources the company has.

Targeting

Score each segment on attractiveness, meaning size, growth, margin, and how crowded it is, and on fit with what the company can deliver through its capabilities, channels, and brand. Then pick a pattern. Serve one segment, serve several with offers tailored to each, or run one undifferentiated offer across the market. Write down the segments you will not serve, because that list is what changes sales and media plans.

Positioning

Define the place the offer should hold in the target customer's mind, relative to named alternatives. A working positioning statement names the target, the frame of reference (what the customer compares the offer with), the point of difference, and the evidence that makes the difference believable. Ries and Trout's argument was that a position is set against what competitors already hold in the customer's mind, so claiming a position a rival owns rarely works.

What it is not

A persona exercise. Personas describe an imagined person for design and messaging work. Segments are sized groups chosen so the company can allocate money and people.
A study you run once. Segments shift as the market and the product change, and a targeting choice made at launch can be wrong two years later.
A tagline. The position is the choice of target, frame, and difference. Advertising copy expresses it and can change while the position stays the same.

The steps

Fix the market and the decision

State the market boundary and the decision the work will inform, such as sales coverage, product range, or media spend. A segmentation with no decision attached tends to end as a description.
Segment on behavior and need

Start from evidence of how customers buy and what they value, using transaction data, interviews, or a survey. Group customers who respond alike, then describe each group with demographic or firmographic traits so sales and media can find them. Test each group against the five usability criteria.
Size and score the segments

For each segment, estimate size, growth, margin, and competitive intensity, and score the company's fit. Use margin and retention, not revenue alone. The output is a scored list with the evidence behind each score.
Choose targets and name the exclusions

Pick the segments to serve and the pattern for serving them. Write the excluded segments down, with what will stop, such as a channel, a discount, or a sales motion.
Write a position for each target

Name the target, the frame of reference, the point of difference, and the evidence. Test it with customers in the segment and with the sales team against the competitors they actually meet.
Hand it to the mix and track it

Turn each position into product, price, channel, and message decisions. Track win rate, share, and retention by segment, so you can see whether the choice is working.

In practice

Six weeks into a growth review for an insurance broker's small-business unit, the team finds an old segmentation by headcount and industry that predicts nothing. Retention and margin look the same in every cell. The team re-cuts two years of policy data by how clients buy. Owner-run firms that shop on price online renew at low rates and leave for direct insurers. Firms with a part-time finance manager want someone to review their cover at renewal, renew at the highest rate, and produce about 40 percent of commission. Multi-site firms with specialist risks need underwriting the unit cannot provide.

The unit targets the advice-seeking group and stops paying for price-comparison listings aimed at the first. The positioning is written for the sales team, not for an advertisement: for firms of 10 to 50 staff that want one adviser to handle cover, compared with buying direct online, the broker offers a named adviser who reviews cover every year, and the evidence is claims handled by its own team. The specialist-risk firms are referred to a partner broker. Adviser capacity, renewal outreach, and the website are changed to match.

What you leave with

A segment map with each segment's size, margin, retention, and the traits used to find it.
A scored list of segments and the reasons behind each score.
The target segments, and a written list of the segments the company will not pursue.
A positioning statement for each target, tested with customers and sales.
The changes to product, price, channel, and message that follow, and the segment-level measures to watch.

Failure points

Segments that do not predict buying

Groups cut by age, region, or company size respond to the offer in the same way. Targeting them changes nothing, and the team concludes that segmentation does not work here.

Segments nobody can find

Needs-based clusters from a survey cannot be identified in the CRM or reached through any channel. Sales cannot tell which segment a new lead belongs to, so the segmentation stays in the research deck.

Targeting every segment

No executive wants to give up revenue, so every segment is declared a target. The result is the original undifferentiated offer with more slides.

A position the offer cannot back

The point of difference is something the product does not deliver, or something a competitor already owns in customers' minds. Customers do not believe it, and the spend behind it is wasted.

Stopping at the deck

The work ends with a positioning statement, and prices, channels, sales targets, and budgets stay as they were. The choice has to show up in the marketing mix and the sales plan to mean anything.

The failures above are usually Framing: the wrong problem was chosen before this framework was applied.

Origin

Wendell R. Smith is usually credited with bringing market segmentation into the marketing literature, in "Product Differentiation and Market Segmentation as Alternative Marketing Strategies," Journal of Marketing, July 1956. He described segmentation as adjusting the offer to the differing demands of groups in the market, as opposed to differentiation, which tries to bend demand toward one product. Advertisers and brand managers had been grouping customers informally for decades before him.

Positioning came from advertising. Jack Trout wrote "Positioning is a game people play in today's me-too marketplace" for Industrial Marketing in June 1969, and he and Al Ries followed with "The Positioning Era Cometh," a three-part series in Advertising Age in 1972, and the book Positioning: The Battle for Your Mind in 1981. Philip Kotler's Marketing Management, first published in 1967, set segmentation, targeting, and positioning out as one sequence in its later editions, and that is where most practitioners learned the STP label. None of the terms is trademarked.