Validation is cheap evidence before expensive commitment. Before paying for a business plan, signing a lease or ordering stock, find out whether people have a problem worth solving and whether you can serve them on workable terms. A plan documents a validated idea; it does not create demand simply by describing it convincingly.
You do not need certainty. You need enough evidence to decide what to test next, what to change and what to stop. Work through the five validations below. Keep a simple evidence log: assumption, test, result, source and next decision. Record objections alongside encouraging responses.
1. Problem: listen before you pitch
Define one target customer and one situation in which the problem arises. “Small businesses” is too broad; a particular type of owner handling a particular recurring task gives you something to investigate.
- Arrange conversations with 10–15 people who fit that description. Treat this as a practical starting exercise, not a statistically representative sample. Include people outside your friends and family.
- Ask about the last time the problem happened: “What triggered it? What did you do? What did that cost in time or money? What have you already tried?” Listen for actual events.
- Ask who chooses a solution, who pays and what would trigger a change. In a business sale, the user and budget holder may be different people.
- Write down the customer's wording and current workaround. Group similar accounts, then note contradictions and differences between customer types.
A useful signal is a specific recent problem, an existing workaround and a consequence the person cares about. Someone showing you their current process or introducing you to the buyer gives you something concrete to follow up.
“Great idea” and “I would probably use that” are politeness until behaviour supports them. Avoid “Wouldn't this make life easier?” If conversations reveal an occasional annoyance with no urgency, narrow the problem or test another customer group before commissioning a plan.
2. Market: count reachable customers
Start from the customers you could actually serve. Use the TAM, SAM and SOM framework to distinguish total demand, the segment your offer can serve and a realistic obtainable portion.
- Set your geography, customer criteria and sales channel. Build a count from identifiable businesses, locations or households using traceable directories and official sources. Record dates and remove duplicates.
- Estimate purchases per customer and the price per purchase. Multiply eligible customers × plausible purchase frequency × price for the same period. Label assumptions; do not treat every eligible customer as a buyer.
- Limit the obtainable market by your outreach capacity, sales cycle, delivery capacity and competitors. Explain how you would reach the first customers rather than claiming an arbitrary market share.
- Find 3–5 real competitors. Record their published prices, source URLs, access dates, package contents, minimum commitments and extra fees. Where prices are not published, mark them unknown and request a quote; never fill the gap with a guess presented as fact.
Include the alternative of doing the task internally or doing nothing. Compare like with like: a subscription and a one-off service may solve different parts of the problem. Your output should be a short competitor sheet and a market calculation another person can retrace. If the reachable customer list cannot support your required sales, revise the offer or territory.
3. Money: build a one-page model
Use one period, such as a month, and one clear sales unit: an appointment, order or subscription. Start with price × realistic volume. Ground volume in available hours, seats, leads or production capacity rather than the income you hope to earn.
- Calculate revenue excluding sales tax collected for remittance. Show discounts, refunds and seasonal changes where relevant.
- List variable costs per sale: materials, packaging, shipping, payment fees, commissions, direct labour and expected waste or rework. Include your own delivery time at a reasonable replacement cost.
- List fixed operating costs: rent, utilities, insurance, software, administration, marketing and professional fees. Add a separate startup list for equipment, deposits, fit-out, permits and opening inventory.
- Calculate contribution per sale as price minus variable cost. Divide fixed operating costs by positive contribution per sale to estimate operating break-even volume. A zero or negative contribution cannot cover fixed costs through extra sales.
- Compare break-even volume with capacity and reachable demand. Then test lower sales, higher costs and slower customer payments.
Keep profit and cash separate. Inventory may need paying for before it sells; customers may pay after you deliver. Map when cash enters and leaves, including tax remittances, loan principal, interest and owner withdrawals, without double-counting items. Identify the funding gap before launch. If the model works only when you work unpaid indefinitely, it needs revision.
4. Willingness to pay: make a real offer
Interest in a problem does not establish demand for your solution at your price. Describe a specific offer: who it serves, what is included, the price, delivery timing and the customer's next action. Test it with people who can make the buying decision.
- Choose a suitable commitment: a pre-sale, a deposit for a pilot, or a letter of intent from a business buyer. Only take money when you can meet the stated terms and applicable requirements.
- Explain what exists today and what is still proposed. Set clear delivery, cancellation and refund terms. Keep customer funds available for refunds where required.
- If taking payment is premature, use a waitlist tied to the real offer and price. Ask people to select a package or request a booking, rather than merely subscribe to general updates.
- Record how many qualified people saw the offer, what they did and why others declined. Separate friends, free users and people without purchasing authority.
A paid order is stronger evidence than an email address, although refunds and repeat demand still matter. A letter of intent may be non-binding: record its conditions, budget approval and expected timing. A waitlist tests interest at stated terms, not completed sales. If people decline, ask whether the obstacle is price, timing, trust or the offer itself. Change one important element and test again.
5. You: check whether you can deliver
The unglamorous blockers belong near the beginning. Customer demand cannot make an unsuitable premises lawful or a supplier's minimum order affordable.
- List the licences, permits, qualifications and insurance your exact activity may require. Check the relevant municipal, provincial and federal authorities. Confirm eligibility, permitted activities, application steps and dependencies directly.
- For premises, check zoning, permitted use, occupancy, accessibility, utilities and any required inspections or alterations before making an unconditional commitment. Ask for written confirmation where available.
- Request supplier terms: minimum orders, lead times, deposits, payment timing, delivery charges, returns and continuity arrangements. Feed those terms back into the cash model.
- Compare your skills, available time and personal cash needs with the work required. Identify who covers gaps and what that costs. Check whether the launch schedule depends on approvals you cannot control.
Keep a blocker register with an owner, evidence needed and a decision date for each item. An unresolved essential approval is a reason to pause spending, even when interview feedback sounds encouraging.
Red flags: stop or fix these first
- Nobody will pre-commit. When qualified buyers repeatedly decline a clear, realistic offer, investigate the reason. More presentation polish does not resolve weak demand.
- The economics require impossible volume. Break-even sales above physical capacity or a defensible customer pipeline call for different pricing, costs or scope.
- The required licence does not exist. If the concept depends on permission unavailable for that activity or location, stop that version. A possible future rule change is not current authorisation.
For each red flag, name the assumption that failed and a testable change. If no workable change exists, walking away is a useful validation outcome. Do not pay for a full plan to justify money already spent.
When validation says yes
A provisional yes means the problem is concrete, buyers have shown commitment, the numbers are plausible and no essential blocker remains unexplained. Keep the evidence log: it becomes an input to deeper analysis.
Before a substantial commitment, a feasibility study can test the concept more fully. Depending on agreed scope, it connects market research, operating capacity, startup costs, cash requirements, break-even and sensitivity analysis into a proceed, revise or stop assessment. It adds structured investigation of gaps and alternatives beyond your initial DIY tests.
Once the chosen concept is supported, business plan writing services can turn the evidence into an operating strategy, financial projections and a funding case for the intended reader. Neither document removes uncertainty; both should expose assumptions that still need checking.
Try TheBizPlanAI — the free AI business plan generator
Create a structured draft plan with an executive summary, market analysis and financial projections.
Try TheBizPlanAI