Too small to test.
Too important to guess.

Too small for research. Too small for A/B testing. Too important to guess. That’s what Prior.Run is for.

Every consumer team has two ways to learn what customers think.

• —A study — eight people, three weeks, real budget.
• —Or a test — real traffic, six weeks, a statistically detectable lift.

Both have a floor. Below it, the evidence costs more than the change is worth, so the page ships on whatever the team caught clicking through it the night before.

And most of what goes wrong on a page isn’t a matter of opinion. A promo code that wipes the address. An email code that won’t paste. A delivery date you only see after you pay. Somebody just has to go through it before your customers do.

1. 01Paste a link. A live page, a checkout, a signup flow, or a design prototype from Figma, Framer and the like.
2. 02Set it up. Pick desktop or mobile. Choose what to comment on — the words, how it works, or both. Then tell your panel what to do, like “use the yearly plan” or “skip the quiz”.
3. 03Your panel runs it. In a browser, they click every button, open every tab, and fill every form with test details. They take the main path to the end, then the other one.
4. 04You get the page back, marked up. In about 45–55 minutes. Each pin sits on the spot it’s about, with what went wrong, why it matters, and the change to make.

Pins come sorted: fix first, worth fixing, small, and keep — the things that already work, so nobody “fixes” them.

A/B tests tell you which. Never why. And they need traffic, so the worse version sits in front of real customers for weeks while you wait.

Studies answer why beautifully — and cost three weeks. Worth it for a new product, absurd for a change to the shipping step.

QA checks that the page does what the ticket says. It doesn’t ask whether a person understands it. A button can work perfectly and still look switched off.

A Run takes under an hour, before a single customer sees the page. It clears the bar for the changes none of the others can profitably touch.

Don’t trust it. Check it.

Every finding is pinned to the exact spot it’s about. Open the page, look where the pin is, and you’ll know in seconds whether it’s right. Nothing in the report asks you to take a number on faith.

The best first run is a page you already know has a problem — the form support keeps hearing about, the step where people drop off. See if it finds it.

What it holds itself to

• —It never presses the button that costs money. No card details, no Place order, no Subscribe, and no Send on a contact form. Forms get filled with test details, so nothing lands in your orders or your inbox.
• —It says when it was stopped. A login you haven’t connected, a “press and hold” check, a Continue button that never turned on — a screen it couldn’t reach is listed and marked as blocked, never quietly skipped.
• —It doesn’t guess about price. It won’t call a price steep unless it can name what a rival charges.
• —It knows a prototype isn’t finished. Links that go nowhere and screens you haven’t built yet don’t count as faults.

For online stores

• —Checkout: promo codes, shipping choices, address fields, and whether the total changes at the last step.
• —Product page on mobile: the size picker, where the price sits, what’s hidden in the tabs, and whether Add to Cart is in reach.
• —Cart: changing amounts, removing an item, and whether the upsell gets in the way of paying.

• —Signup and onboarding: password rules, email codes, and whether you can skip the steps that don’t matter yet.
• —Pricing page: whether the plans are easy to compare, what each one really includes, and where the fine print lives.
• —Your product, signed in: the upgrade flow, settings, or the first screen a new account lands on.
• —Demo or contact form: every field filled, every error message seen — and nothing ever sent.

• —A prototype before it’s built: paste the Figma or Framer link and find the confusing step while it’s still cheap to change.
• —A landing page from an ad: whether the page keeps the promise the ad made, on the device the ad was seen on.
• —A redesign: run the new version before it replaces the old one, and keep what the pins say already works.

A link. No code to install, no access to your analytics, nothing to set up on your site.

If the page is behind a login, sign in once in a browser window we open for you. We keep you signed in for the Run, and never keep your password.

Pick a ready-to-go audience, or build one from your own customers. The audience tells your panel who the page is for — a first-time buyer and a returning customer trip over different things.

Your first Run is free. After that, see the plans.