Free tool
Time-to-outcome checker
Ranking form variants on what closed instead of what completed is a better idea than it is a practical one, and whether it works for you is an arithmetic question rather than a philosophical one. Volume, close rate and cycle length decide it. For a lot of funnels the honest answer is no, and this will say so.
Runs in your browser. Nothing you type is sent anywhere, stored, or logged — there is no request to send it in. Built by the team behind the dishonest dashboard.
How this is calculated
Every figure above comes from the arithmetic below. No weighting, no model, no numbers of ours mixed into yours.
How fast outcomes arrive
gradeable per month = submissions × gradeable share per variant per month = gradeable per month ÷ variants
Submissions that never get a disposition cannot be part of a test, however real they were. A CRM full of leads nobody ever marked won or lost has a gradeable share near zero.
How many you need
pA = close rate pB = close rate × (1 + lift) n = ( z₀.₉₇₅ × √(2p̄(1−p̄)) + z₀.₈ × √(pA(1−pA) + pB(1−pB)) )² ÷ (pB − pA)²
The standard two-proportion sample size at 95% confidence and 80% power. Note how it scales: halving the lift you want to detect roughly quadruples the sample.
How long that takes
months accumulating = n ÷ per variant per month months of lag = median days to disposition ÷ 30.44 total = months accumulating + months of lag
The lag is added once rather than per submission. Submissions resolve while later ones are still arriving, so only the tail of the sample is still open when collection ends — adding one median cycle at the end is the reasonable approximation, and it is an approximation.
Where the defaults come from
The defaults describe a small B2B funnel: 600 submissions a month, a 3% close rate, a month to disposition, and 70% of leads eventually getting marked one way or the other. Nothing about it is a benchmark — it exists so the page does something before you touch it.
The 20% default lift is the smallest improvement most people would bother shipping a form change for. Try lowering it to 10% and watch what happens to the sample size; that behaviour is the most useful thing on this page and it is a property of the statistics, not of our opinion.
The verdict bands — under six weeks, under four months, under a year — are our judgment about what a marketing team can actually wait for, not a finding. Someone with a different testing cadence would draw them elsewhere, and the number above them is the part that matters.
What this cannot tell you
It assumes your traffic, offer and sales team hold still for the duration. Over a test measured in quarters they will not, and every change is a confound. This is the main reason a long answer here should be read as “no” rather than as “yes, but be patient.”
It also treats disposition lag as a single median. Real time-to-outcome distributions have long tails, and a median of 30 days often hides deals that take six months. If your distribution is skewed, the answer here is optimistic.
And it says nothing about whether outcome data is worth collecting — only whether it can adjudicate a split test. Those are different questions with different answers. Knowing which of your traffic sources produce leads that close is useful at any volume, including volumes far too low to test on. If the answer here is a flat no, that is the version to build.
Once you do have two variants with outcomes attached, the outcome-weighted split test calculator is where you find out whether the difference holds up.
Why we built this
Every number on this page is one your form builder could have told you and didn’t.
Endpoint Forms is an open-source form builder for marketers: forms built to convert, data that goes wherever you need it, and every submission carrying what it turned out to be worth. It is not shipped yet. The waitlist is where we tell you when it is.