Most organisations describe their market with one word: "students", "families", "small businesses". Market segmentation goes further. It divides a market into groups of customers with similar needs or behaviour, who respond to marketing in a similar way. Each group is a segment.

There are four common ways in, called bases for segmenting:

• Demographic: age, life stage, income, household. Easy to find, but weak at explaining why people buy.
• Geographic: where people live or work. It matters for delivery areas, local services and anything people travel to.
• Behavioural: what people do, such as how often they buy, how they use the product, which channel they buy through, and whether they come back.
• Needs-based: the benefit people want, or the job they're trying to get done.

Business markets use firmographics: sector, size (staff or turnover), location, and how the business buys. You met the decision-making unit in Module 2; a segment of businesses where the owner decides alone needs different marketing from one where a buying team decides.

Start from insight, not labels. The strongest segments come from needs and behaviour, usually combined with one or two other bases to find and reach people. Go back to the insight statements you wrote in topic 2. Where customers want clearly different things, you may have found the edges of a segment. A label such as "Gen Z" lumps together people who want very different things.

A useful segment passes four tests. Its needs differ from the other segments. You can find and reach its members. It's big enough to matter (in words, if you have no figures). And you have evidence it exists, not just a hunch.

UK example: Gymshark. Gymshark says it began in 2012 in a garage in Birmingham. CNBC reported in May 2023 that, with no advertising budget, founder Ben Francis and co-founder Lewis Morgan gave free products to popular fitness YouTubers at a bodybuilding expo in 2013, and some wore the clothing on their channels. The segment was defined by behaviour and needs: people who trained seriously and watched fitness creators. Age alone would have missed the point, and so would "people who exercise".

Once you have segments, you have to choose. Targeting means choosing which segments to serve, and how many, based on how attractive each one is and how well your organisation can win it.

Attractiveness asks whether a segment is worth having:

• Size: are there enough of these customers to matter?
• Growth: is the group getting bigger or smaller?
• Reachability: can you find them, and are there places where they gather or look?
• Competition: who already serves them well, and could you offer something better?

Fit asks whether your organisation can win it. What is it already good at? What does it already own, such as a reputation, a location, skills or a product? A large, growing segment is no use if a competitor serves it better and you have nothing different to offer.

Describe each score in words and say what evidence it rests on: "growing, judging by the rise in local brewing classes" is more honest than a made-up number. If you do have figures, give their source.

There are three broad approaches. Focus on one segment, and do it well. Serve several segments with one offer. Or take a differentiated approach: a different offer or message for each segment you choose. Small organisations usually start with one or two, because every extra segment costs money and attention.

Say why not. A good targeting decision explains what you've chosen not to do. Writing one reason for each rejected segment stops the team drifting back to "everyone", and it shows a manager you weighed the alternatives.

AI activity. Ask an AI assistant to suggest segments for your organisation, describing what it sells and to whom. Then test each suggestion against your own evidence from topic 2 and your source log: keep it, change it or reject it, and write one reason each time. Watch for segments the AI invented from a label, and for any figures it gives without a source. Keep the AI's list and your verdicts; they show your judgement.

A persona is a short, evidence-based portrait of a typical customer in a target segment. It gives a team one person to picture when it writes an email, designs a page or chooses a channel. "Would this help Jordan?" is an easier question than "would this help the segment?"

A persona is only as good as its evidence. Build it from your research in Module 2: survey answers, interviews, reviews, order data, comments on social media. Every detail should earn its place by changing a decision. Where they look for advice matters; their favourite colour probably doesn't.

A useful persona covers:

• Name and a short situation: their circumstances as they relate to what you sell.
• Needs: what they want from the product, and what puts them off.
• What they read, watch and listen to: the places your organisation could meet them.
• Evidence: where each detail came from.

The traps. The first is invented detail. Teams add a job, a dog and a hobby to make the persona feel real, then start making decisions based on things nobody checked. The second is the cliché: assumptions based on age or gender, such as "older people don't shop online" or "women buy the gifts". These are stereotypes, not evidence, and they can lead you to ignore customers who don't fit.

Test each detail by asking "how do we know?". If the answer is "we don't", cut it, or mark it as an assumption to research. Then give the persona a plainly fictional name, and say it's illustrative. A persona stands for a group; it isn't one real customer, so never use a real person's name or details.

One persona per target segment is enough to start. If your organisation targets two segments, write the second persona later, once the first has proved useful.

The path to purchase is the steps a customer moves through from first noticing a need to buying and buying again, including where they look and who influences them. It builds on the decision process from Module 2, but adds the places your organisation can show up.

This topic uses five stages:

1. Trigger: something starts the need, such as running out, a life change or seeing a friend's purchase.
2. Search: they look for options, online or by asking people.
3. Compare: they weigh up a shortlist on what matters to them.
4. Buy: they choose, often trying something small first.
5. After purchase: they use it, judge it, and decide whether to buy again and tell others.

At each stage, list the touchpoints: any point where a customer meets the organisation or hears about it, online or offline. A search result, a review, a shop shelf, an email and a delivery box are all touchpoints. Then list the influences: friends, reviews, creators, staff, price comparison. Some touchpoints belong to you, some don't, and the ones you don't control often matter most.

The customer funnel is a simpler model of the same idea: awareness, consideration, purchase and loyalty. It narrows because fewer people reach each stage. Its value is that it suggests a different goal at each stage: get noticed, get considered, get chosen, get chosen again. You'll turn those goals into measures with baselines in Module 6. Real paths loop and skip, so treat the funnel as a planning aid, not a description of how people behave.

Map the path for your persona, not for "customers" in general. Read the worked example later in this topic before you start the task: it maps Larkfield's persona, Jordan, stage by stage. Then do the same for your organisation, marking any touchpoint you've assumed rather than seen in your evidence.