When a Client Asks for a Free Proof of Concept (What to Say Instead of Yes)

The Short Answer

Do not build a custom proof of concept for free. Give the client evidence instead, and if they need evidence specific to their own business, charge a small fixed fee for it. That is the whole answer, and everything below is how to say it without losing a deal you actually want.

Free custom work fails on both sides of the trade. You spend real hours on something you cannot reuse, and the client learns that your time is the part of the arrangement that is negotiable. Every conversation that follows, about scope, about timelines, about the invoice, starts from the position that your effort is available for the asking.

It also fails as sales. A free proof of concept has no owner on the client's side. Nobody scheduled it, nobody defended a budget for it, nobody has to explain what came of it. Work nobody paid for gets reviewed by nobody, and the deal it was supposed to unlock quietly does not close.

Why This Request Got More Common

Buyers now assume that building a working prototype is quick and nearly free, because they have watched software get demonstrated into existence in a few minutes. That assumption is not entirely wrong, and pretending otherwise makes you sound defensive.

What the assumption misses is where the cost actually sits. The generating is fast. The understanding of their process, the access to their systems, the edge cases their staff work around every day, the meeting to find out what the spreadsheet columns really mean: none of that got faster. A proof of concept that reflects their business is mostly discovery, and discovery is the expensive part of every engagement whether or not anybody is paying for it.

So the honest framing is not that a prototype is hard. It is that a prototype about them requires learning about them, and that learning is the work. Say it in those words and the request usually converts on its own, because it stops sounding like a refusal and starts sounding like an explanation the buyer has not heard before.

The Three Requests Hiding Inside It

Can you show us something first is one sentence covering three completely different problems. Answering the wrong one is how builders end up donating a fortnight and still not closing.

Ask one question before you answer: if the proof of concept works, what happens next. The reply sorts the three cases immediately. A buyer who says we would move to the full build in the new quarter has a process. A buyer who says we would take it to the team and see is telling you there is no budget and no decision maker in the room, which is a qualification failure rather than a proof problem. That conversation belongs earlier, in [qualifying a lead before the call](/blog/qualify-a-lead-before-the-call).

The Free Paid Line

The Profit Room's rule for this is the Free Paid Line, and it is a single question: who wrote the specification. Anything you built for your own reasons and can show to any prospect is free. Anything built to this client's requirements, on their data, inside their systems, is paid. Hours are irrelevant. Ownership of the spec is everything.

The line holds up in conversation because it is principled rather than grudging. You are not refusing to show them anything. You are explaining that the moment work becomes about their business it becomes an engagement, and engagements have a fee. Buyers who run their own businesses recognise that logic instantly, because it is how they price their own work.

The Reply That Keeps the Deal Alive

Answer in four moves: agree with the instinct, name the real cost, offer the free option you already have, then put the paid option on the table with a number. Never lead with the refusal.

1. Agree. Wanting to see it working before committing is completely reasonable and I would ask for the same thing.
2. Explain. The building part is quick. What takes the time is learning how your process actually works, and that is the same work whether we call it a proof of concept or the first week of the project.
3. Give. Here is a walkthrough of something close to what you are describing, built for another business, so you can see the shape of it today.
4. Offer. If you want it on your own data, we can run a two week pilot at a fixed fee. You get a working version of the riskiest piece, and if you go ahead the fee comes off the full build.

Then stop talking. The offer needs a decision, and filling the silence with reassurance turns a firm proposal back into a negotiation. If the buyer counters by asking for the pilot free, you have learned something useful: this is a budget problem wearing a risk costume, and no amount of free work will fix it.

Keep the tone even throughout. Nothing about this is a confrontation, and builders who deliver the line apologetically get talked out of it. The steadiness is the persuasive part, which is the same muscle described in [presenting your price without flinching](/blog/present-your-price-without-flinching).

Convert It Into a Paid Pilot

The paid pilot is the highest-converting answer to this request. Small fee, narrow scope, one week or two, aimed squarely at the thing the client is most worried about, with the fee credited against the full engagement if they proceed.

It works because it changes who is invested. A paid pilot has a budget line, an internal sponsor and a date, which means somebody at the client is now motivated to make it succeed. Free work has none of those things. The fee is not really about the money at that stage. It is about creating the conditions in which a decision gets made.

• Scope it to the riskiest component, not to a shrunken version of everything. Prove the hard part and the rest stops being in question.
• Fix the fee and fix the deliverable in writing. An open-ended pilot is free work with an invoice attached.
• Credit the fee against the full build if they proceed, and say so upfront. It removes the feeling of paying twice.
• Write the next step into the pilot agreement. On the last day you present the result and both sides decide, with a date already in the calendar.
• Price it so it is easy to approve without a committee. Most businesses have a threshold under which one person can just say yes, and a pilot belongs beneath it.

The full mechanics, including how to present the result so it leads naturally into the main engagement, are in [running a paid pilot before a full engagement](/blog/run-a-paid-pilot-before-a-full-engagement). The related version for clients who are not ready to build anything yet is [the audit offer](/blog/the-audit-offer-foot-in-the-door).

When Free Is Actually the Right Call

There are narrow cases where building something for free is a good decision, and they share one feature: you would have built it anyway. If the work only exists because this client asked for it, the exception does not apply.

• It becomes a reusable asset. You can strip the client specifics and show it to every future prospect in the same industry, which makes it marketing you happened to get a warm lead from.
• It is genuinely an hour. Something small enough that scoping it would cost more than doing it, and small enough that nobody expects a second one.
• You have no proof at all yet. Very early on, evidence is worth more than the fee, and that trade is deliberate rather than accidental. The structured version is [the anti-portfolio](/blog/the-anti-portfolio-proof-with-no-clients).
• You are testing an offer. Building free for two businesses in one vertical to learn what the real problem is can be a sound investment, as long as you know that is what you are buying.

Even in those cases, put a boundary on it. One deliverable, one deadline, one round of feedback, and a clear statement that this is a sample rather than the start of the project. Free work without a boundary is the most reliable way to acquire a client who never pays you, and the containment rules are the same ones in [a scope of work that prevents scope creep](/blog/scope-of-work-that-prevents-scope-creep).

What to Do If They Say No

Some buyers will decline the pilot and repeat the request for free work. Treat that as information rather than a setback, and do not counter by lowering the fee to nothing.

A buyer who will not spend a small fixed amount to remove their own risk is telling you one of three things: there is no budget, there is no authority in the room, or they are collecting free work from several suppliers. None of those improve if you work for free. The first two get resolved by time and qualification. The third is a pattern you want to be out of early.

Leave the door open in a way that costs you nothing. Send the generic walkthrough, say the pilot offer stands whenever they are ready, and set a reminder to check in. Deals that stall at this point often close months later when a budget cycle turns, and the builder who stayed friendly and did not discount is the one who gets the call. The follow-up structure is in [reactivation emails that win back past clients](/blog/reactivation-email-win-back-past-clients).

The Real Cost of Saying Yes

The unbilled hours are the smallest part of the cost. The expensive parts are positional and they persist long after the prototype is forgotten.

1. You have set your rate at zero for the first phase, and every later number is compared against it.
2. You have become a supplier who works before terms are agreed, which invites the same pattern at every milestone.
3. You have spent your available capacity on an unqualified buyer while qualified ones waited.
4. You have removed the client's need to make a decision, because nothing was at stake for them.
5. You have taught yourself that the way to handle hesitation is to work harder rather than to sell better.

The builders who charge properly are rarely the ones with the best code. They are the ones who put a small, easy payment between the conversation and the work, every time, without apologising for it. If the underlying discomfort is about your own rate rather than about this client, the honest diagnosis is in [why builders undercharge](/blog/why-builders-undercharge).