Build credible proof before you have client results by showing your method, judgement and safeguards, then selling a small, bounded first engagement.
Short answer: Do not pretend to have a track record you lack. Build credibility from evidence you can produce now: verify who you are, show a real piece of your process, demonstrate judgement on a representative example, state safeguards, and offer a small paid first engagement with a clear output. You need one convincing proof item for the buyer’s largest concern, not a page full of unsupported claims.
Case studies are useful because they compress several doubts into one story. They suggest that a customer trusted you, you completed the work and something improved. Without them, you must answer those doubts separately.
That is possible. Buyers do not purchase history for its own sake. They look for evidence that you understand their situation, can perform the work and will not expose them to unreasonable risk. A new business can show all three honestly, although it cannot yet claim repeatable client outcomes.
Build the Precedent-Free Proof Ladder
Use the Precedent-Free Proof Ladder. Each rung answers a different buyer concern, and each depends less on assertion than the one before it.
- 1. Identity: Buyer concern: Are you a real, accountable operator?; Evidence you can show now: Named people, address or service area, direct contact details; What it must not imply: A larger team or premises you do not have
- 2. Relevance: Buyer concern: Do you understand my problem?; Evidence you can show now: Specific diagnosis, informed explanation, limits of fit; What it must not imply: Experience with clients you have not served
- 3. Method: Buyer concern: Can you perform the work?; Evidence you can show now: Steps, decision criteria, sample inspection or demonstration; What it must not imply: A proprietary method if it is routine practice
- 4. Safeguard: Buyer concern: What happens if something goes wrong?; Evidence you can show now: Scope, checkpoints, insurance where relevant, clear terms; What it must not imply: A guarantee you cannot fund or control
- 5. Commitment: Buyer concern: Can I test this without a large bet?; Evidence you can show now: Small paid first engagement with a defined output; What it must not imply: A disguised full project delivered free
My view is that giving away full projects to obtain case studies is usually a mistake. Free work proves that people accept free work. A small paid engagement gives stronger evidence of demand, establishes professional expectations and limits risk for both sides.
Establish identity without performing size
Use your real name, real operating location or service area, a monitored email address and a telephone number you answer within a stated period. Explain who performs the work. If it is you, say so. “Our team” reduces trust when the customer later discovers one person and a network of occasional subcontractors.
Publish relevant career history with dates, responsibilities and boundaries. “Ten years in manufacturing” is less useful than “managed weekly production planning for a 24-person joinery workshop from 2018 to 2024”. Do not disclose a former employer’s confidential information or suggest its endorsement.
Demonstrate relevance through diagnosis
You can show expertise before client results by explaining the problem with enough specificity that the buyer recognises their operating reality. Describe the inputs you would inspect, the order in which you would inspect them and the signals that would change your recommendation.
Good diagnosis includes disqualification. A quote-review adviser might decline a project where the labour records are absent because any margin finding would be invented. That boundary demonstrates judgement.
Make the method inspectable
Choose one representative part of your work and perform it using invented, public or your own information. Label illustrative data clearly. A photographer can plan and shoot a self-directed brief. A procurement adviser can compare publicly available supplier terms. An operations consultant can annotate a fictional workflow while explaining which real records would be required.
Show the reasoning, not only the final output. Include one limitation. A flawless sample with no trade-offs can look less credible than a clear account of what the evidence does not establish.
Do not publish material created for a former employer or client without explicit permission. Removing a logo may not remove confidential processes, personal data or ownership rights. If permission is unclear, create a fresh demonstration from first principles.
Put safeguards where claims used to be
New firms often compensate for missing proof with absolute promises. That increases the very risk the buyer is trying to evaluate. Replace “satisfaction guaranteed” with operating controls you can honour: written scope, named deliverables, agreed access, review points, change approval and a documented completion test.
State response and update standards as numbers. “Updates every Tuesday by 4 pm” is testable. “Excellent communication” is not. Explain dependencies too. If completion requires the customer’s data by a date, put that in the proposal rather than allowing an impossible deadline to become your apparent failure.
Insurance, licensing, consumer rights and professional obligations vary by country and sector. Check the rules that apply to your work, and obtain advice from a qualified local legal, insurance or regulatory professional for a specific decision. Never display cover or a credential you do not currently hold.
Worked example: BenchMark Quote Review
BenchMark is a new quote-checking service for small bespoke joinery workshops. It has no client case studies. Instead of claiming to “maximise profitability”, the owner publishes an illustrative review of a fictional fitted-kitchen quote and explains each assumption.
- Materials: £4,800
- Specialist subcontractor: £700
- Workshop labour, 190 hours at £24: £4,560
- Delivery: £280
- Total recorded cost: £10,340
The customer price is £12,500, so apparent contribution is £12,500 minus £10,340, or £2,160. BenchMark’s check then identifies 22 installation hours omitted from the labour allowance and £160 of disposal cost. At £24 an hour, omitted labour is 22 × £24 = £528. Total omissions are £688, making corrected cost £11,028 and contribution £1,472.
BenchMark’s fixed first-stage fee is £425. The demonstration does not claim that every review will find £688. It proves that the owner understands the arithmetic, checks installation and disposal, and distinguishes an example from an outcome. A prospect can buy a review of one completed quote before deciding whether to use the service across future projects.
That paid first stage has a precise output: a reconciled cost list, stated assumptions and questions requiring the workshop owner’s answer. It does not include repricing the entire catalogue or implementing a job-costing system. The boundary keeps the first decision proportionate.
Ask for the smallest sensible commitment
Reduce scope, not standards. The first engagement should solve a complete small problem or produce a decision the buyer values. It should not be a teaser whose output is useless without a much larger purchase.
Price it from delivery time, direct cost and required contribution. Free may be rational when the activity is a brief sales diagnosis you provide to every qualified prospect, but call it sales work and cap it. Do not promise a free “audit” that requires hours of bespoke analysis and then pressure the buyer to continue.
At completion, ask whether you may document the starting problem, work and result. Agree what can be named. If the customer declines, retain only information you are permitted to use.
Related guides
Build your first proof set in seven days
On day one, write down the buyer’s three largest doubts and rank them by how likely each is to stop a purchase. On days two and three, create one representative demonstration that answers the top doubt with visible reasoning and clearly labelled example data.
By day four, publish or send a plain account of who performs the work, the exact first-stage scope, response standards, dependencies and limits. On day five, ask two experienced practitioners and three plausible buyers what the evidence proves and what it does not. Correct any accidental implication.
During days six and seven, offer the bounded paid engagement to five well-matched prospects. Record objections word for word. Add evidence only when it answers a repeated, material concern. Do not fill the page with badges because one buyer asked a question you can answer directly.
Frequently asked questions
Can I use work from my previous job as proof?
Only with permission and accurate attribution. Explain your personal role, the context and what you did not control. Do not imply that an employer’s result belongs entirely to you or that the organisation endorses your new business.
Contracts, confidentiality duties and intellectual-property rights vary, so review the relevant agreement and seek qualified local legal advice if ownership or disclosure is uncertain. When permission is unavailable, describe your responsibilities in general terms and create fresh sample work. Your judgement can be demonstrated without exposing someone else’s information.
Should I tell customers that the business is new?
Yes, when the fact is relevant, and never imply otherwise. You do not need to lead every conversation with an apology. State when the business began, then distinguish business age from your relevant personal experience.
A buyer may reasonably value the founder’s direct attention while also recognising the lack of organisational history. Explain the safeguards that address that risk. The exception is a procurement process that asks for trading history or prior contracts. Answer exactly as requested rather than substituting your career history for the company record.
Can friends or family give me testimonials?
Only for work they genuinely bought or experienced, with the relationship disclosed where it could affect how the endorsement is understood. A character reference is not evidence of a commercial result. Asking a friend to pose as a customer creates a false claim and can cause lasting damage once discovered.
Use named professional references for reliability or relevant prior responsibilities if they have consented. Keep them separate from customer testimonials. Advertising and endorsement rules vary by jurisdiction, so check local requirements before publishing any incentivised or connected endorsement.
Do qualifications make up for having no case studies?
They reduce one kind of uncertainty, not all of it. A relevant qualification may show that you met a knowledge or practice standard. It does not prove that you understand this buyer’s situation, communicate reliably or can deliver the promised scope. Pair the credential with a visible diagnosis, method and safeguard. In regulated work, qualifications or licences may be mandatory rather than optional proof. Requirements vary by country and sector, and a qualified local professional or regulator should confirm what applies to your specific service.
Should I offer a money-back guarantee?
Not unless the completion condition is objective, within your control and financially affordable. A broad satisfaction guarantee can create disputes because satisfaction is undefined and outcomes may depend on customer action. A tighter remedy is often more credible: correct a specified deliverable that fails the agreed acceptance check within a stated period. Put any guarantee and exclusions in clear terms. Consumer law can override your wording and varies by jurisdiction, so obtain qualified local advice before relying on a guarantee to limit obligations.
How much sample work should I show?
Show enough to answer the buyer’s largest delivery concern, usually one complete representative example plus a clear account of the process. Ten shallow samples do not beat one relevant demonstration with visible reasoning. Choose work close to the offer you want to sell, label fictional information and explain limitations.
Add a second example only when a different buyer segment or use case creates a genuinely different concern. Keep the set current. Old samples that no longer match your method can weaken credibility even when their presentation looks polished.
How quickly can I create a real case study?
As quickly as a suitable paid engagement reaches a result that can be measured honestly, but do not force the timing. Agree the starting measure and permission before work begins. Some outputs can be assessed at delivery; commercial outcomes may need months and may be affected by other changes.
Record both dates and avoid attributing the whole outcome to your work without evidence. If the customer will confirm the process but not share numbers, a process case study can still be useful. Say exactly what it demonstrates.
Comments (0)