How to Validate an MVP Before Spending Money on Development

How Do You Validate an MVP Before Building It?

Climb four rungs in order, and stop the moment one fails. Rung one: fifteen problem interviews with people who have the problem — if you cannot find fifteen, that is itself the result. Rung two: a landing page stating the promise and price, with traffic sent to it and sign-ups counted. Rung three: money — pre-orders, deposits, or a concierge version delivered manually. Rung four: the smallest build that automates what rung three proved people pay for.

The ladder works because each rung is an order of magnitude cheaper than the mistake it prevents. An interview costs an hour. A landing page test costs about a thousand dollars. A wrong build costs thousands and — worse — months. The discipline is refusing to let excitement skip rungs: a founder who jumps from idea to build is betting the whole budget on the rung the interviews would have tested for free.

RUNG 1  interviews (free)      → is the problem real and painful?
RUNG 2  landing page (~$1,100)  → do strangers want the promise?
RUNG 3  pre-sales / concierge   → will they pay real money?
RUNG 4  smallest build          → does usage survive reality?
         ↑ stop and rethink at the first rung that fails

What Counts as Real Evidence — and What Is Vanity?

Evidence is anything that cost the giver something: money, time, reputation or data. Vanity is everything free. A deposit is evidence; a compliment is not. A stranger completing your concierge workflow twice is evidence; a friend saying they would 'definitely use it' is not. Rank every signal you collect by what it cost the person giving it, and weigh your decision accordingly.

How Do You Run Problem Interviews That Do Not Lie to You?

Ask about the past, never the future. 'Would you use this?' invites politeness; 'what did you do the last time this happened?' invites facts. Fifteen interviews with people who genuinely have the problem will tell you three things no survey can: whether the problem is frequent enough to matter, what people currently do about it, and what they already pay — in money or hours — for the workaround.

Listen hardest for the workaround. Someone maintaining a spreadsheet, paying a virtual assistant, or gluing three tools together has pre-validated your idea with their own effort; someone who shrugs and says 'yeah, it is a bit annoying' has invalidated it just as clearly. And record who the strongest interviews came from — those specifics become your first landing page audience and, later, the scope decisions in what to build in the MVP.

How Do You Run a Landing Page Test — and What Do the Numbers Mean?

Publish a page that makes the promise, states the price, and asks for a sign-up — then buy traffic and count. The full arithmetic, every input visible: a purpose-built SEO landing page costs roughly $500–$1,000 to have made properly; add $300 of paid traffic at about $1.50 a click for 200 targeted visitors. Total: around $1,100 for a real read on demand.

Benchmarks tell you what the count means. Unbounce's conversion benchmark analysis of 41,000 landing pages puts the median conversion rate at 6.6 percent across industries, and 3.8 percent for SaaS specifically. So 200 visitors converting at the SaaS median yields seven or eight sign-ups — meaning single digits are normal, not failure. Beating the median meaningfully with a stated price on the page is a genuine signal; converting at one percent with no price shown is a warning taken early, for about the cost of a phone.

Two upgrades sharpen the test. Put a price and a 'reserve your spot' deposit button on the page — even a refundable $10 deposit converts intent into evidence. And add a simple lead-capture chat widget so visitors can ask questions; the questions they ask are free interview data from exactly the strangers you need.

What Do Pre-Sales and Concierge Tests Prove?

Pre-sales prove price; concierge tests prove the workflow. A pre-sale — charging before the product exists, with an honest delivery date — is the strongest demand evidence available to a founder, because it removes every hypothetical from the question. Ten pre-orders at $49 is not $490; it is ten strangers overruling your doubt with their cards.

A concierge test delivers the service manually before automating it: matching candidates by hand, compiling the report yourself, scheduling the appointments over WhatsApp. It feels unscalable because it is — deliberately. What it buys is a complete map of the real workflow, including the steps you did not know existed, before any of them is hardened into software. Founders who concierge first consistently commission smaller, sharper builds, which is exactly the budget logic covered in how much to spend on an MVP.

How Much Should Validation Cost in Total?

Under $1,500, in most cases, and mostly time rather than money. The table below is the whole ladder priced honestly. Note what it adds up to: less than the median software build, for evidence that decides whether the build should happen at all.

Who Needs Validation — and When Should You Just Build?

The ladder is for founders guessing at someone else's problem — which is most founders. If your idea came from an article rather than a lived workflow, every rung applies, in order, no exceptions.

But validation has honest limits, and pretending otherwise wastes months. Skip ahead when you live the problem daily — a recruiter building a recruiting tool has run years of unpaid interviews already. Skip ahead when an existing audience is asking to pay: pre-sell to them this week instead of surveying them. And reconsider the arithmetic when the build itself is cheap: PINCLER is an AI-first custom software development studio, and across PINCLER's 79 documented projects a validation-grade build starts at $500 — when six weeks of testing costs more than the software, shipping the real thing to twenty users is sometimes the better experiment. What no situation justifies is skipping the price signal: whatever you ship first, charge for it.

What Are the Common Validation Mistakes?

Validation fails through comfort more than through method. These six patterns account for most of the false positives that later become expensive builds.

• 1. Interviewing friends — polite data is worse than no data, because it feels like evidence.
• 2. Asking about the future ('would you use…') instead of the past ('what did you do when…').
• 3. Running a landing page with no price — you validate curiosity, then build for buyers who never existed.
• 4. Treating a big waitlist as demand — inboxes are free; only conversion to money confirms it.
• 5. Stopping at one rung — interviews without a traffic test validate the problem but never the product.
• 6. Validating forever — the ladder has four rungs, not fourteen; past pre-sales, only shipping teaches.

PINCLER's Perspective: a Build Studio That Tells You Not to Build Yet

It is odd for a development studio to publish a guide about not paying for development, so here is the self-interested version of the argument: validated clients scope better, phase better and come back for phase two. Across PINCLER's 79 documented projects — all fixed-price between $500 and $2,500, median $1,450 in 13 days — the cheapest tier exists precisely for validation: 11 of the 79 project types start at $500, and websites, the natural landing-page category, run a median of $1,025 and 8 days. The full dataset is at our research page.

The pattern we see in inbound briefs is consistent: founders arriving with pre-sales or concierge experience ask for less software, not more — because they know which steps the workflow actually needs. That usually means a smaller phase-one quote, and it always means version one meets its first users with the hardest question already answered.

The Bottom Line

Validate an MVP by buying evidence in ascending price order: free interviews, a ~$1,100 landing page test, then real money via pre-sales or a concierge run — and only build once a rung has been paid for by strangers. The whole ladder costs less than the software it protects you from building wrongly.

When a rung passes and it is time to build, keep the same discipline on scope: one workflow, fixed price, code you own. A free 30-minute call turns whatever your validation found into a written quote within one working day.