How do I test a business idea before building it?

Put up a one-page site that sells the idea as if it already exists: one promise, one visible price, one call to action. Send it 300 to 500 strangers using $100 to $300 of ads or two weeks of posting where your buyers gather. If 5 to 10% leave an email, or 1 to 2% preorder, the idea has real interest. Below that, change the promise and rerun. The whole test costs under $700.

How much does it cost to validate an idea with a landing page?

Under $400 if you DIY: a page builder runs $0 to $19, a Stripe payment link is free to set up, and $100 to $300 of ads supplies the visitors. Done-for-you, our Overnight Site is $399 with the page live in a day, so the full experiment lands under $700. Compare that with building first: a studio-built MVP runs $8,000 to $18,000, and US agency builds commonly start around $60,000.

How many visitors do I need for a landing page test?

300 to 500 real strangers is the practical range. Under 200, the percentages are noise: one enthusiastic visitor swings your conversion rate by half a point. Past 1,000, you are paying to re-learn an answer you already have. And they must be strangers, because friends, family, and existing followers click out of loyalty and quietly corrupt the test. $100 to $300 of targeted ads or two weeks of posting in the right communities usually gets you there.

What conversion rate means my idea is validated?

Commonly cited smoke-test benchmarks: 5 to 10% of visitors leaving an email signals real consumer interest, and 1 to 2% paying a deposit or preorder signals real demand. Money is the stronger signal by far, since people join waitlists casually but pay reluctantly. Below both thresholds, the promise failed: change the offer, the price, or the audience rather than the ad creative, then rerun for about another $100.

Can I take preorders for a product that doesn't exist yet?

Yes, if you are honest about it. Label it a preorder with a realistic ship date, make deposits refundable, and refund everyone immediately if you decide not to build. Kickstarter normalized paying before the product exists. What crosses the line is pretending the item is in stock, or keeping money for something you quietly cancelled. If a full preorder feels heavy, a small refundable deposit, even $10, still separates real buyers from the merely curious.

What should I do if my landing page test fails?

Pivot the promise, not the ad. A failed test means the offer did not land, so change the price, the audience, or the core benefit and rerun with another $100 of traffic. If two or three rewritten promises all fall flat, kill the idea and be glad: the test just saved you an $8,000 to $18,000 build. That is the test working, not failing. Keep the page, the ad account, and the lessons for idea number two.

Validate a Business Idea With a Landing Page (2026)

The cheapest real test of a business idea is a one-page site that sells the thing before it exists: one promise, one visible price, one button, and $100 to $300 of traffic pointed at it. Strangers either click or they don't, and either way you get your answer in about two weeks for under $700 instead of after an $8,000 to $18,000 build. Almost every guide on this is written by a tool company selling you its page builder. This is the full how-to, plus the part they never offer: we'll build the test page for you, live in a day, for $399

TL;DR. Put up a one-page site that sells your idea as if it already exists: one promise, one visible price, one call to action. Send it 300 to 500 strangers, roughly $100 to $300 of ads or two weeks of posting. Commonly cited benchmarks: 5 to 10% email signups means real interest, 1 to 2% preorders means real demand. Below that, pivot the promise. Total cost: under $700

Why is a landing page the best way to test a business idea?

Opinions are free and priced accordingly. Your friends say the idea is great because they like you. Survey respondents say they would pay $30 because saying yes costs them nothing. A landing page is the first test where strangers spend something real: an email address, a card number, thirty seconds of actual attention. Clicks do not lie to be polite.

This is what marketers call a smoke test: a one-page site that sells the product as if it exists, so you can measure demand before building anything. Here is how it stacks up against the other ways people test ideas:

How you test	Cost	Time	How it fails you
Asking friends and family	$0	A weekend	They like you, so they lie
Running a survey	$0 to $100	1 to 2 weeks	"Would you buy this?" costs nothing to answer with yes
Building the product first	$8,000 to $18,000 studio-built, $60,000+ at a US agency	2 to 6 months	You learn the demand answer after the money is gone
Landing page smoke test	$100 to $700 all in	1 to 2 weeks	Only if you send it the wrong visitors

The landing page wins on one brutal fact: it is the only cheap test where the judge does not know you and gains nothing by being kind. Commonly cited startup post-mortems put "no market need" at the top of the list of reasons companies die. That is a hundred-dollar question, and most founders pay five figures to answer it.

What goes on the page: one promise, one price, one button

That is the entire page. Everything you add after those three muddies what the test measures.

The promise

One sentence at the top: what it does, who it is for, why it beats the current way. If you cannot write that sentence, stop here. You are not ready to test, and no page builder fixes that. Under it, two or three supporting lines and one image or mockup. Done.

The price

Show it. A page without a price tests curiosity, not demand, and curiosity is worthless: everyone is curious. The moment a visitor sees "$49" and still clicks, you have learned something a thousand survey answers cannot tell you. If naming a price feels premature, that is the test already telling you the offer is still vague.

The button

One call to action, as close to money as you can get. Best: a preorder or refundable deposit through a payment link. Weaker but still useful: "get early access" with an email field. Pick one and make it the only thing to do on the page. One number is easier to read than two.

What to leave off: the About section, the twelve-icon feature grid, the nav bar, the footer full of links. Anywhere a visitor can click that is not the button is a leak in your measurement. We wrote a full teardown of one-page website design if you want the layout logic, but for a smoke test the rule is shorter: if it does not sell the promise, it is not on the page.

How do I get 300 to 500 real visitors?

The test needs 300 to 500 visitors who have never met you. Fewer than a couple hundred and the percentages are noise. And they must be strangers: your WhatsApp group will click out of love and quietly ruin the data.

Pay for them: $100 to $300 of ads

A small Meta or Google Ads budget is the standard smoke-test play, and $100 to $300 is the commonly used range: for consumer ideas that typically buys several hundred clicks. Write the ad with the same promise as the page, word for word. If the ad promises something the page does not, you are testing your ad-writing, not your idea. B2B clicks cost several times more, so if your buyer is a business, lean on the second option.

Earn them: two weeks of posting where your buyers already are

Subreddits, Facebook groups, niche forums, LinkedIn if you sell to businesses, local groups if the idea is local. Show up, be useful, mention the page where the rules allow it. Slower than ads and it costs evenings instead of dollars, but the visitors are better matched, and the comments tell you things a bounce rate never will. One caution: respect each community's self-promo rules. Getting banned from the one forum where your buyers gather is an expensive way to save $200.

What does not count: friends, family, your existing followers, and cheap bulk traffic from sites that promise 10,000 visitors for $20. Those clicks are bots or bored people, and they will hand you a confident answer to a question nobody asked.

How do I know if my idea passed?

Read the numbers against commonly cited smoke-test benchmarks, not against hope. These come from common industry practice, not our data:

5 to 10% of visitors leave an email: real consumer interest. Worth another round, possibly worth building.
1 to 2% put down a deposit or preorder: real demand. Strangers paying for a thing that does not exist yet is the strongest signal a pre-launch idea can produce.
Below both: the promise failed. Not the font. Not the button color. The promise.

Keep the reading honest about sample size. This is a smoke test, not a clinical trial: you are looking for a clear signal, not statistical significance. Twenty emails from 400 visitors is a 5% signal worth acting on. One email from 40 visitors is a rounding error, so keep the traffic coming until the base is real.

When a test fails, pivot the promise, not the ad. Founders instinctively fiddle with headlines and images because that feels like progress. But 400 people saw the offer and shrugged, so change the actual offer: a different price, a different audience, a different core benefit. Then spend another $100 and rerun. A full pivot-and-rerun cycle costs about a week.

Zero signups from 400 real visitors is not bad luck. It is an answer.

Why not just skip the test and build the product?

Because the build is the expensive part, and it costs the same whether the idea is good or not.

Our MVP Sprint runs $8,000 to $18,000 fixed and takes about 8 weeks. A comparable app at a US or UK agency commonly runs $60,000 to $200,000 and up. A freelancer might do it for $5,000 to $15,000 with coordination risk on top. Whichever door you pick, an unvalidated idea walks through it at full price.

Now the test math. A done-for-you smoke-test page is $399. Add $300 of traffic and the whole experiment is $699, under a tenth of a studio-built MVP. Run it three times on three different ideas for about $2,000 and you can kill two duds and greenlight one idea with actual demand behind it, for less than most builds take as a deposit.

And a failed test is not the method failing. It is the method working. The $399 that tells you no in two weeks is worth more than the $8,000 to $18,000 build that tells you the same thing eight weeks and one quiet launch later.

Done-for-you version: the $399 Overnight Site is your validation page live in a day, copy and capture included, and you own it. Cheaper than a month of tool subscriptions, and it exists tomorrow.

Get the $399 page → See the real prices

Should I build the test page myself or pay someone?

Either works. This is not about tools, and anyone insisting the test needs a specific platform is selling you the platform.

The DIY path, fully spelled out

Write the one-sentence promise and pick a price.
Build the page in an afternoon. Carrd is about $19 a year, Framer and Wix have free tiers, and honestly a Google Site works. The page is disposable, so do not spend a week making it pretty.
Wire the button: a Stripe payment link for preorders or deposits takes ten minutes and no code. For the email version, any form tool feeding a spreadsheet.
Point $100 to $300 of ads at it, or spend two weeks posting where your buyers are.
Read the results against the benchmarks above. Pivot the promise and rerun if it fails.

Total: under $400 and two weekends. The common failure is not the build, it is the stall: the page sits at 90% for a month because polishing it is more comfortable than testing it.

The done-for-you path: $399, live in a day

Our Overnight Site is literally this product: a one-page site written and designed to sell one promise, live in a day, $399 fixed. You send the idea and the price, we build and ship the page, you point traffic at it. You own every file from day one, and here is how a site goes live in a day if the speed sounds suspicious.

Pay the $399 if your weekends are worth more than the fee, if you have been "about to launch the test" for over a month, or if you want a page credible enough that a $49 preorder feels safe to a stranger. Skip it if you have a free afternoon and any comfort with a page builder. The tool matters that little.

Either way: if the test passes and the idea is software, the next step is our $500 Week-1 Build Audit, credited against the build. You leave with a one-page spec, a fixed price, and a go or no-go.

When is a landing page test the wrong move?

Honest routing, because this test is not for everyone:

You cannot describe the offer in one sentence. Then you are not ready to test, you are ready to think. Go back one step: our guide on what to do first when you want to start a business covers the ordering, and step one is not a website.
You sell a service and already know 20 potential buyers. Skip the page. Pitch ten of them directly this week. One pre-sold client is validation and revenue in a single move, and it costs $0.
Your idea is an app. Still test the promise on a landing page first, it stays the cheapest signal. But before any build money moves, read website or app first: plenty of "app ideas" are a website for their first year.
Long-cycle B2B or regulated products. When a purchase takes six months and four stakeholders, a $100 ad test measures the wrong thing. Ten real conversations with buyers beat any conversion rate.

One reframe to take with you: if you run the test and it fails, the test worked. Most founders never find out whether anyone wanted the thing. You found out for less than a phone bill, and the build budget is still yours to point at idea number two.

Validate a Business Idea With a Landing Page Before You Spend Build Money