Most positioning advice stops at the inspiring part. Find your zone of genius, serve the people you love. Then Monday comes, you open LinkedIn search, and there is nothing to type. That is the moment you discover your niche was a feeling, not a target.
Niche selection for consultants fails for one boring reason: the niche cannot be turned into a list. If you cannot produce 300 named humans with filters that exist on LinkedIn, you do not have a niche. You have a mood board.
Here is the test I use, and why "consultants" — or "SMEs", or "ambitious founders" — never passes it.
Why "consultants" is not a target
Say your offer is lead generation for independent consultants. Try to build that list.
LinkedIn gives you job title. "Consultant" as a title covers a McKinsey analyst, a freelance SAP specialist, an in-house change manager at a bank, a part-time nutrition coach and someone who put "Consultant" there while between jobs. One search, five different economies. You cannot write one message that lands on all of them, because they do not share a problem, a budget, or a week.
That is the real failure. A target is not a label, it is a shared situation. The fastest diagnostic: write the first line of your outreach and see if it stays true for every profile in the search. "Most independent consultants struggle with irregular pipeline" is true for roughly everybody and useful to nobody. "You are running delivery solo while two old clients make up most of your revenue" is only true for a slice — which is exactly the point.
The second failure is that broad targets make you invisible in search. People do not look for a consultant. They look for someone who did this exact thing, for someone exactly like them, last year. Vague targeting also wrecks your profile copy, because a headline written for everyone converts no one.
The four filters a niche must survive
Before you commit, run the candidate niche through four questions. Any no and you narrow or move.
1. Can I build the list? Not "does the market exist" — can I, this morning, produce names? That means the niche maps onto filters that LinkedIn actually has: job title, industry, headcount, geography, membership of a group, attendance at an event, a technology mentioned in the profile. "Companies that just raised money" is buildable. "Companies with poor internal communication" is not, until you find a visible proxy for it.
2. Is there a trigger? A reason to contact them this month rather than any month. New role in the last 90 days. A job posting for the function you replace. A funding round. A merger. A regulation deadline. Without a trigger, your outreach is permanent background noise.
3. Do they already buy this? Not "should they". Do they have a line in the budget, or an equivalent they currently pay a person or an agency for? Selling into a category that does not yet exist in your buyer's head costs you two years of education you cannot fund.
4. Can one message fit them all? If you need five variants, you have five niches. Pick one.
Most candidate niches die at question one or two. That is useful information, not a setback.
From a category to a situation
Narrowing is rarely about going deeper into an industry. It is about adding a situation to the category.
Start with what you actually sold in the last 18 months. Not what you can do — what someone paid for. List the last ten paid engagements. Write down, for each: what the company was, who signed, what had just happened to make them call, and what they were afraid of.
Now look for repetition. You will usually find a pattern in the moment, not the sector. Something like: a head of operations, 50 to 200 staff, who inherited a process built by someone who has left. Or: a founder at the point where the first sales hire is failing. These are targetable because each has an observable signal on LinkedIn.
Then write the niche in one sentence with four parts: role, company shape, moment, pain. For example: "Operations directors in 50-200 person manufacturing firms, in the six months after an ERP migration, who still run the critical reports in spreadsheets."
It sounds absurdly narrow. That is the correct reaction. Count the people who fit in your country. If it is above 400, you can prospect it for a year. Smaller than 150 and you widen one dimension — usually geography or company size, never the pain.
Testing niche selection with 60 messages, not a workshop
You do not choose a niche by thinking. You choose it by sending.
Take two candidate niches. Build 60 names for each — not 500, you are testing the message, not scaling. Write one first message per niche, built on the specific moment, and send both batches over two or three weeks. Keep everything else identical: same profile, same time of day, same follow-up rhythm.
Then compare three numbers, in this order of importance:
- Reply rate, including polite refusals. A clear "not us, we already have an agency" is a sign your message was understood.
- The quality of the replies. Do people argue with the premise, or do they recognise it? "How did you know we're doing that" is the signal you want.
- Calls booked. This comes last because the sample is tiny and one friendly contact distorts it.
Two things usually happen. Either one niche is obviously alive and the other is silence, in which case you have your answer cheaply. Or both are flat, which means the pain you named is not the pain they feel — rewrite the sentence, not the list.
One honest note on volume. Doing this by hand costs maybe three hours a week for a few weeks. There are tools that automate it; BIZLINKED is one of them — its Chrome extension runs in your own browser and LinkedIn session and automatically sends invitations and AI-written messages and follow-ups, inside its own internal limits (at most 100 invitations a week, 150 once more than 40% are accepted, business hours, random pauses). Be clear-eyed: LinkedIn's User Agreement does not allow third-party software or extensions that automate activity on the platform, and accounts can be restricted at any volume. No internal limit removes that risk, and using any such tool means you accept it. For a 60-message test, manual is perfectly viable.
While the test runs, watch who looks at you. Profile views from the niche, with no message from you, tell you the positioning is readable. Treating profile visits as a signal costs nothing and often tells you more than your reply rate during the first fortnight.
What changes once the niche is narrow
The obvious gain is that your list gets easy to build. The quieter gains matter more.
Your follow-ups get writable. When you know the moment, the second and third message have somewhere to go — a specific consequence, a specific example. Broad targeting is why most people's follow-up sequences collapse into "just bumping this up".
Your content stops being generic. You are writing to 400 people, not to LinkedIn. Posts get sharper because you know what they argue about internally.
Your network becomes relevant instead of large. Five hundred connections who all share one situation outperform five thousand random ones, which is the whole argument in why network size matters less than relevance.
And referrals start working, because people can repeat your sentence. Nobody refers "a great consultant". They refer "the guy who fixes reporting after an ERP migration".
What to do Monday morning
Open a document. List your last ten paid engagements with the four columns: company shape, buyer role, triggering event, fear. Find the repetition.
Write one sentence: role, company shape, moment, pain. Run it through the four filters. If you cannot build the list on LinkedIn search in ten minutes, the sentence is still too abstract — add a visible signal and try again.
Then build 60 names, write one message, and send it this week. You will know more by Friday than you would from a month of positioning exercises. Narrowing is not a commitment for life. It is a commitment for one quarter, and you can widen whenever the list runs dry.
