Validation Experiment Calculator

How much does this experiment need?

Use this before you launch. It tells you the floor, not the target. Budget above the number it gives you, never below.

What you are planning

Variants you are comparing Two audiences, or two messages. Not both at once, or you cannot attribute the result to either. Target impressions per variant A thousand is enough to read a click-through rate. Reaching 100 clicks and 5 signups usually needs three to four times that. Expected CPM (cost per 1,000 impressions, USD) Broad audiences in Lebanon and the wider region tend to run between 2 and 6. Use your own account history if you have any. Expected click-through rate (%) 1.2 is a fair planning assumption for a cold audience. Expected landing page conversion (%) 6 sits mid-range for a waitlist or a lead form. Days you will run it Five is the minimum. The first 48 hours are the learning phase and tell you very little.

What that buys

Total impressions: —
Minimum total spend: —
Daily budget: —
Expected clicks: —
Expected signups: —
Cost per signup: —

Working backwards

Impressions per variant for 100 clicks: —
Impressions per variant for 5 signups: —

The second number is usually the one that binds.

The CPM and the two rates above are planning estimates you entered, not measured values. Swap them for your own numbers as soon as you have a full run behind you.

What is a signup worth to you?

Work backwards from what a customer is worth. If you cannot fill these in, that is worth knowing before you spend anything.

Your economics

Revenue per customer per year (USD) A contract, a subscription, or an average order. Whichever fits how you actually sell. Average customer lifetime (years) Be conservative. Two years is a reasonable default for something unproven. Gross margin (%) Revenue minus what it costs you to deliver. Signup to customer conversion (%) Of the people who join a waitlist, how many eventually buy? One in four is optimistic at an early stage. Share of lifetime value you will spend to acquire (%) Thirty is a common ceiling. Above that the economics get fragile.

Your numbers

Lifetime value: —
Most you can pay per customer: —
Most you can pay per signup: —
LTV to CAC ratio: —

The third number is your threshold. Put it in your hypothesis and in box 5 of the canvas, then do not move it.

Every input here comes from you. None are benchmarks. If you are guessing at the signup to customer rate, say so on the canvas and treat the threshold as provisional.

What did it actually say?

Do not open this until the campaign has finished running. Reading partial results is how founders talk themselves out of a clear answer.

What happened

Measure	Variant A	Variant B	Combined
Impressions			—
Clicks			—
Spend (USD)			—
Signups			—
Of those, the right buyer			—

Measure

Variant A

Variant B

Combined

Impressions

—

Clicks

—

Spend (USD)

—

Signups

—

Of those, the right buyer

—

Read every submission by hand before you fill in the last row. That row is the one people skip.

The four numbers

Metric	Variant A	Variant B	Combined	Reference
Click-through rate	—	—	—	0.8–2% cold
Landing page conversion	—	—	—	3–10% waitlist
Cost per signup	—	—	—	vs your threshold
Share who are the right buyer	—	—	—	above 50%

Metric

Variant A

Variant B

Combined

Reference

Click-through rate

—

0.8–2% cold

Landing page conversion

—

3–10% waitlist

Cost per signup

—

vs your threshold

Share who are the right buyer

—

above 50%

The rule you wrote before launch

At or below this, continue (USD) At or above this, stop (USD) Anything between the two means change one thing and run it again.

These have to be the numbers you wrote down before launching. If you are editing them today, that is the failure this whole thing exists to prevent.

Data adequacy

The call

Reference ranges

Starting points, not targets. Your own account history beats every number on this page.

Metric	Rough range	If yours sits below it
Click-through rate	0.8–2.0% cold	Either the message is wrong or the audience is. Test the problem framing before you touch anything else.
Landing page conversion	3–10% waitlist	The page promised more than the ad did, or the ask is too big. Cut form fields first.
Cost per 1,000 impressions	USD 2–6 in Lebanon	A high CPM usually means the audience is too narrow. Widen it before you raise the budget.
Cost per signup	Against your own threshold	There is no universal number. If it costs more than a customer is worth, that is a fact about this market, not a campaign problem.
Share who are the right buyer	Above 50%	You bought attention, not demand. Tighten the audience or make the offer less generically appealing.
Signup to customer	10–30% early stage	Interest is not becoming purchase. Look at the gap between what you promised and what you deliver.

Metric

Rough range

If yours sits below it

Click-through rate

0.8–2.0% cold

Either the message is wrong or the audience is. Test the problem framing before you touch anything else.

Landing page conversion

3–10% waitlist

The page promised more than the ad did, or the ask is too big. Cut form fields first.

Cost per 1,000 impressions

USD 2–6 in Lebanon

A high CPM usually means the audience is too narrow. Widen it before you raise the budget.

Cost per signup

Against your own threshold

There is no universal number. If it costs more than a customer is worth, that is a fact about this market, not a campaign problem.

Share who are the right buyer

Above 50%

You bought attention, not demand. Tighten the audience or make the offer less generically appealing.

Signup to customer

10–30% early stage

Interest is not becoming purchase. Look at the gap between what you promised and what you deliver.

Before any of that is worth reading

1,000+ impressions per variant. Below this, click-through rate swings on chance alone.
100+ clicks in total. Needed before a conversion rate means anything.
5 to 10 conversions. Two signups is a coincidence. Ten is the start of a signal.
48 hours minimum. Delivery is still calibrating, so early cost figures usually mislead.
5 to 7 days for the full run. Shorter and the learning phase dominates everything.

These ranges come from campaign work in Lebanon and the region rather than from published industry figures. Treat them as a place to start.