How to Choose Your First Segments:Start With Five
How to choose your first segments: start with five, define each in one sentence, give each a different job, and add a sixth only when one is too broad.
To choose your first segments, start with five: new, engaged, cooling, dormant and customers. Define each in one sentence, give each one a different job, and add a sixth only when one of the five is plainly too broad to write a single message for.
How to choose your first segments: fewer than you think
Most solopreneurs do one of two things. They build no segments and mail everyone. Or they spend a weekend building thirty and never send to most of them.
Both are wrong. Both feel productive.
Here's the rule I'd hold you to. Segment as little as will change a decision. A segment that doesn't change what someone receives, start something automatically or make a number readable is a saved filter with a name. Delete it.
And you're alone. Every segment is a message you write differently, a definition you maintain and a number you interpret. Nobody else is going to do any of that. So the budget is small, and you should spend it where it counts.
Choose your first segments: the five that carry almost everything
Even with a thousand new leads arriving a month, a one-person business can run almost everything from five groups. Sorted by where the relationship stands, not by what the form happened to collect.
| Segment | One-sentence definition | What it needs |
|---|---|---|
| New | Joined, and I have not sent them anything yet. | A welcome. Not an offer. |
| Engaged | Opened or clicked recently. | Your strongest material and your direct asks. |
| Cooling | Last real engagement was a while ago. | Your single most useful piece of content, less often. |
| Dormant | Received a lot, engaged with nothing for a long time. | One honest win-back, then a rest. |
| Customers | Already bought. | Entirely different messages from everyone above. |
Read the third column again. That's the test when you choose your first segments. If two segments would get the same email, they are one segment.
Notice what this table doesn't contain. No country. No age. No job title. Those are easy to collect and they tell you very little about what someone is going to do next. Two people of the same age in the same city can be evaluating, never going to buy, and already a customer. The five above sort people by what is likely to happen, which is the only thing worth sorting on. If you want the full argument, fit versus behaviour makes it properly.
Free guide · 21 pagesStop writing to everyone the same way.The Segmentation Playbook: tags that stay clean, segments built on fit and behaviour.Get the free guide →Add one more: your best-fit target
Once the five exist, you may add one fit-based group for the kind of person your product serves best. The right role, the right situation, the right problem.
But only if you have enough contacts for it to mean something. A group of nine people teaches you nothing reliable. It just feels precise.
This sixth group is also what becomes your selling queue. Faced with the whole list, you write to the newest names or the ones you recognise. Faced with a group defined as high fit, a recent deliberate action and not yet contacted, you spend your one personal hour on the people most likely to buy this week. That's a better use of an evening than any amount of extra posting.
Check each contact against two questions
Before you trust the five, run every contact past two questions. Is this the kind of person my product genuinely serves? And have they done something lately?
Fit changes slowly. Behaviour fades. A click last week matters, a click last year barely does. Put the answers together and you get four boxes, and almost every decision you make about a contact is a decision about which box they are in.
- High fit, recent behaviour. Talk to them now, personally if you can.
- High fit, no recent behaviour. Nurture patiently with useful material. They are worth the wait.
- Low fit, recent behaviour. Serve them well and cheaply. Do not spend your own selling time chasing them.
- Low fit, no behaviour. Spend nothing.
That last box is the hard one emotionally. Every contact feels like a small asset. Most of them are a small cost. Spending nothing on them is not rude. It's arithmetic.
One more thing underneath it all: every contact should sit in exactly one lifecycle stage, defined by an event rather than a feeling. Subscriber, lead, qualified lead, opportunity, customer. If the stages are accurate, a handful of rules does the job. If they are missing or stale, no number of extra segments will rescue you.
Write the sentence before you build the rule
Here's a habit that saves you a lot of grief. Before you touch a filter, write the definition as one sentence.
"Leads tagged as interested in the advanced offer who visited the pricing page in the last two weeks."
If you can't say it in a sentence, you don't know what the segment is for. The rule will be an accident of whichever filters were nearest.
Then name it after the rule, not the campaign. "Advanced interest, recent pricing visit" will still make sense in six months. "October push" won't. You'll reuse it in March, assuming it means something it doesn't.
Two small things about building the rule itself:
- Prefer "all of" to "any of". "All of" narrows. "Any of" widens, usually far more than you expected, because one loose condition lets almost everyone in. When a segment comes back surprisingly large, a stray "any of" is the usual culprit.
- Choose dynamic or static on purpose. Ongoing groups like these five should be dynamic, so membership follows the rule. A one-off list of people on a call should be static. The difference is covered in dynamic versus static segments, and getting it wrong is how you mail people who bought last month.
When to split a segment, and when not to
The pressure to add segments is constant. Someone asks a good question and you want a group for it. Resist.
Split a segment only when you can say the following, out loud: "I cannot write one message that works for everybody in here." Not "it would be interesting to see them separately". Interesting is how you end up with fifty.
Here's how that plays out. Say you sell a productised audit and your engaged group has grown large. Within it, some people only ever open, and some have clicked through to your pricing. Those two want different things. One needs a single clear call to action near the top. The other needs a nudge to book. That's a real reason to split, because the message changes.
Compare that with splitting engaged people by city. Nothing you'd say would change. So don't.
What goes wrong with too many
Over-segmentation is a real failure and it's quiet. You get groups too small to learn anything reliable from. You get groups too numerous for you to remember what each one is for. And you get the worst outcome: an impressive-looking setup you stop trusting, so you go back to mailing everyone.
The other half of the warning is harder. A segment built on a dirty base is a precise wrong answer. If duplicates, stale records and mislabelled contacts sit underneath, the count looks confident and means nothing. So segmenting is not the first job. Cleaning the list is, or at least it runs alongside.
Free guide · 21 pagesStop writing to everyone the same way.The Segmentation Playbook: tags that stay clean, segments built on fit and behaviour.Get the free guide →Your Monday
- Choose your first segments by writing the five definitions, one sentence each.
- Build them as rules, not lists.
- For each, write down what is sent differently. If you can't, merge it into a neighbour.
- Check how many match against how many can actually receive. The gap is the first thing the base will tell you about itself.
- Stop. Send. Look at the results next week before you add anything.
That last step is the one people skip. A segment you never send to, sell from or measure by is just furniture.
Free assessmentHow trapped are you, really?Take the Corporate Suffocation Index, the free solopreneur readiness assessment.Take the assessment →Frequently asked questions
How many segments does a one-person business need?
Three or four if you have a few hundred contacts and one offer. Five is a comfortable ceiling for most, even with a lot of new leads. Add another only when you can show one of them is too broad to write a single message for.
Should I segment by country or industry?
Only if it changes what you say. Location and job title describe people but rarely predict what they will do next. Fit and recent behaviour do. If a field would not change your message, leave it as a field and do not build a segment from it.
What is the first segment to build?
Customers. They are your most valuable group and the one most often sent the wrong thing, such as acquisition offers for something they already bought. Separating them takes minutes and fixes an obvious mistake straight away.
Can I change my segments later?
Yes, and you should. Once you choose your first segments, retire any segment you cannot explain, and rename ones that describe a campaign rather than a rule. A short list you understand beats a long one you do not trust.
The full method, with the checklists, is in The Segmentation Playbook, a free 21-page playbook.
How trapped are you, really?
Take the Corporate Suffocation Index, the free solopreneur readiness assessment.