Choose a specialist or generalist position by testing demand, access, proof and delivery economics, rather than relying on personal preference.
Short answer: Position a new service business as a specialist when you can identify at least 50 reachable prospective buyers with the same costly problem, and you can reuse most of your proof and delivery process across them. Stay generalist when demand is genuinely sparse, work is driven mainly by trusted referrals, or no single problem can support your required sales. You may still deliver adjacent work, but market one clear doorway for the next 90 days.
Many founders choose “generalist” because it appears to preserve opportunity. In practice, a long list of capabilities transfers the sorting work to the buyer. Someone seeking help with a specific payroll failure does not want to interpret whether “business support and strategic solutions” includes their problem.
Specialisation is not automatically superior. A niche with twelve possible buyers, no affordable route to reach them and one purchase every five years is not a market. The decision should come from commercial evidence, not an identity you feel obliged to keep forever.
Use the Narrowing Wedge Matrix
The Narrowing Wedge Matrix crosses two forms of evidence: market concentration and operating transfer. Market concentration requires enough named buyers plus a usable route to them. Operating transfer requires proof that travels and a delivery process that substantially repeats.
- Proven: Operating transfer proven: Use a specialist doorway now; Operating transfer unproven: Run a 90-day specialist test while capping bespoke delivery
- Unproven or too thin: Operating transfer proven: Keep a broader buyer doorway while packaging the repeatable method; Operating transfer unproven: Stay generalist and gather evidence before narrowing
Do not call market concentration proven from a large industry count. The reachable pool must support your required customer count, with at least one route you can use now. Do not call operating transfer proven unless the same evidence answers most buyer doubts and at least half the delivery steps repeat. The working threshold of 50 reachable prospects can be useful for modest service engagements, but a £100,000 engagement may support a smaller pool than a £300 service.
These thresholds are working guides, not universal market facts. A £100,000 engagement can support a smaller buyer pool than a £300 service. Replace the prospect count with the number required by your own price, conversion rate and purchase frequency.
Start with problem density, not an industry label
“I work with hospitality” sounds specialised, but hotels, pubs, contract caterers and independent cafés can have different buyers and operating problems. “I reduce food waste in independent hotel kitchens” is narrower because it joins a buyer to a problem.
List prospects by name. Search trade directories, local listings, professional groups, supplier networks and your existing contacts. Remove businesses that are too small to pay, too large to appoint you, outside your delivery area or unlikely to face the problem. A niche is not validated by the number of search results. It is validated by a sufficient number of qualified buyers you can actually approach.
Then estimate the customer count you need. If you require £72,000 annual revenue and the average client buys £6,000 a year, you need 12 active clients before allowing for churn or failed sales. A pool of 50 may be plausible. A pool of 15 is dangerously tight unless most are already accessible and purchase predictably.
Check whether you can reach the buyer repeatedly
A specialist position becomes valuable when it concentrates your selling effort. You can attend the same events, answer related questions, build introductions through the same advisers and use language buyers recognise. If each prospect lives in a different professional world, the label may be narrow without creating efficiency.
Name two realistic routes to the decision-maker. “Social media” is not a route. “The monthly regional subcontractor breakfast attended by commercial directors” is. So is “introductions from three quantity surveyors who encounter failed bids”. Record how many relevant people each route can expose you to over 90 days and what it costs in time or money.
Do not choose a speciality whose buyers you dislike serving or cannot credibly understand. Access opens the conversation, but insight sustains it. You need enough exposure to recognise the customer’s constraints, vocabulary and buying process without pretending to have experience you lack.
Measure how far your proof travels
Specialists gain an advantage because one piece of evidence can answer the next buyer’s concern. A process map for dental-practice onboarding is more persuasive to another dental practice than a collection of unrelated projects, even when the underlying operational skill is similar.
Audit your available proof under four headings: problem knowledge, process, outcome and risk control. You do not need all four before choosing a position, but you need a credible route to obtain them. If confidentiality prevents named results, you can still explain how you diagnose the issue, what is measured and how you limit disruption.
Generalists sometimes have strong transferable proof because referrals supply the context. A trusted accountant may introduce an operations adviser with a precise explanation of why that person fits. The referral performs part of the positioning work. Without that introduction, the same adviser needs a sharper public message.
Test delivery repeatability and margin
The real financial benefit of specialisation is not permission to add a fashionable premium. It is the ability to diagnose faster, scope more accurately, reuse knowledge and reduce avoidable rework. Price can rise only when the resulting offer is worth more to the buyer or safer to purchase.
Track hours by delivery stage across five projects. If discovery, research and production are almost entirely different each time, the proposed niche has not created repeatability. If at least half of the steps, questions or quality checks recur, you can build better estimates and protect margin.
Worked example: TenderNorth Bid Services
TenderNorth is a solo bid-writing business in Newcastle. Its owner is considering a general offer for any small company versus specialising in framework bids for construction subcontractors.
The owner has 100 delivery hours a month. Monthly overhead and required owner pay total £4,800.
- General SME proposal: Price per project: £900; Direct cost: £90; Owner hours: 18; Contribution: £810; Projects possible: 5
- Construction framework bid: Price per project: £1,650; Direct cost: £100; Owner hours: 22; Contribution: £1,550; Projects possible: 4
Five general projects produce 5 × £810 = £4,050 contribution and use 90 hours. The business remains £750 below its £4,800 requirement. Four specialist projects produce 4 × £1,550 = £6,200 and use 88 hours, leaving £1,400 after the target contribution is met.
The specialist offer is not better merely because it costs more. TenderNorth’s previous construction work shows that its compliance checklist and evidence-gathering sequence can be reused, while the general projects require unrelated research. The owner also identifies 86 suitable subcontractors through two regional networks. Both axes are proven, placing the offer in the specialist-now quadrant.
Keep the doorway narrow and the delivery boundary sensible
Positioning describes the easiest reason to enter your business, not every piece of work you are legally allowed or practically able to perform. A specialist in onboarding for accountancy firms may accept a related workflow project from an existing client. It does not follow that the homepage should list twelve services.
Set an acceptance rule for adjacent work. Take it when the buyer fits your target group, the problem is close enough that existing knowledge transfers, and the engagement will not displace a better-fit project. Otherwise, refer it or decline it. Hidden variety can support cash flow. Public variety usually weakens recognition.
Practitioners disagree about whether specialisation should begin with a sector or a problem. Sector focus makes buyer access and language easier. Problem focus can produce a larger market and deeper method. My view is to use whichever produces greater proof reuse, then add the second dimension only when the buyer pool remains large enough.
Related guides
Run a 90-day positioning test
In the first week, place three possible specialist wedges and your current generalist position in the matrix. For each, name the buyer, urgent problem, reachable prospect count, access route, reusable proof and repeated delivery steps. Reject any option whose sales requirement exceeds the realistic pool.
During weeks two to six, use the highest-scoring position in 20 direct conversations, proposals or qualified introductions. Record response, sales stage, objection, quoted price, expected contribution and delivery fit. Keep the offer constant enough to learn from it.
At day 45, continue only if the specialist message creates more qualified conversations or clearer buying decisions than the general version. At day 90, compare total contribution and pipeline value, not likes or enquiries. Keep the narrow doorway for another quarter if it wins. If it fails, move one level wider by relaxing either the buyer type or the problem, but not both at once.
Frequently asked questions
Will specialising make me lose work?
Yes, it may reduce unsuitable enquiries, and that is part of the point. The useful comparison is not total enquiries but qualified opportunities and contribution from work you can deliver well. You can still accept adjacent projects under a written rule without advertising every capability.
The exception is a business in a very small local market where each service has low purchase frequency. There, a broader offer may be necessary, but organise it around a coherent customer group so referrals and trust still transfer between services.
How narrow is too narrow for a niche?
A niche is too narrow when your reachable buyer pool cannot support the customer count your economics require. Divide required annual revenue by realistic annual value per customer, adjust for failed sales and customer losses, then compare the result with named prospects you can reach.
Fifty prospects is a useful starting threshold for many small service businesses, not a rule. High-value, recurring work can survive with fewer. Low-value, one-off work needs far more. Also widen if one buyer controls your route to market or a regulatory change could remove most demand.
Can I offer more than one service and still be a specialist?
Yes. Specialisation concerns the customer and problem you are known for, not an arbitrary limit on service lines. Several services can strengthen the position when they form a logical sequence for the same buyer, such as audit, implementation and ongoing review.
Present them under one clear result rather than as an unrelated menu. If the services require different buyers, proof, sales channels and delivery skills, you are effectively operating several businesses. Keep them separate in your messaging or choose one to lead until demand justifies the additional complexity.
What if most of my work comes through referrals?
You can remain broader when trusted referrers consistently explain your relevance. Measure this rather than assuming it. Review the last 20 introductions: note who referred you, what problem they described and whether the work converted. If referrals cluster around one problem, the market may already see you as a specialist.
If they are genuinely varied and profitable, a generalist model can work. The risk is dependence on a small number of introducers. Build direct recognition around the strongest recurring problem before a referral source dries up.
Can I change my speciality later?
Yes. Treat the first position as a 90-day commercial hypothesis, not a permanent identity. Preserve the evidence behind the decision: prospect counts, interviews, conversion, contribution and delivery time. If results weaken, change one boundary at a time so you know what caused the difference.
Existing customers do not need to be abandoned when the public position changes. The exception is where your name, regulated credentials or major capital purchases bind you tightly to a sector. In those cases, test demand before making the visible or irreversible commitment.
What is the difference between a niche and positioning?
A niche defines the part of the market you choose to serve. Positioning defines why that market should choose you instead of an alternative. “Independent care homes” is a niche. “Staff-retention systems for independent care homes with fewer than 80 beds” adds a problem and begins to create a position, but still needs a credible advantage.
You can occupy a narrow niche with weak positioning if competitors make the same promise. Conversely, a distinctive method can position a broad firm, although communicating its relevance usually costs more because examples and buyer concerns vary.
Comments (0)