A real A/B test needs around 1,500 conversions per variant. Very few sites have that — and nearly every agency sells it anyway. We do the maths on your traffic before you pay, and when testing can't work, we say so and do the work that does.
€990 once, credited in full against your first month. The report is yours either way.
How long until the result means anything. Each division is one year.
The maths nobody else shows you
How much traffic an A/B test needs isn't an opinion. It's a formula. Here it is, for a site with a 2% conversion rate, at 95% confidence and 80% power.
Visitors needed per variant · 2% baseline · 95% confidence · 80% power. The smaller the effect you're looking for, the more brutal the maths — and you never know in advance how big the effect will be, so you have to plan for the small case.
| Uplift sought | Visitors per variant | Total per test | At 5,000 visits a month |
|---|---|---|---|
| 5% | 309,928 | 619,855 | 124 months |
| 10% | 78,039 | 156,078 | 31 months |
| 20% | 19,784 | 39,568 | 8 months |
| 50% | 3,292 | 6,585 | 1.3 months |
Halve the effect you're looking for, and the traffic you need quadruples. At 5,000 visits a month, a single test for a 10% uplift takes two and a half years. Anyone still selling you “A/B testing with statistical significance” is selling you something that cannot exist at your scale.
Speero — the agency behind CXL, at $10,000 a month — publicly turns away every client under 100,000 visitors a month and sends them to conversion research instead of testing. Turning down that much revenue on principle is the most credible thing anyone in this field does.
We use the same threshold. The difference is that we also do the research work, rather than handing you back to the market.
Visitors a month. The published minimum before a serious shop will run a test for you — and roughly where we draw the line too.
Not sure which side of that line you're on? We'll tell you for free.
Your numbers, not a brochure's
Four sliders. It works out how long one clean test would take at your traffic — and what the uplift you're hoping for would actually be worth if it happened.
Testing can't work here
One clean test would take
62 monthsOver five years for a single answer. Long before then your site, your offer and your market have all moved.
Sample size from Kohavi's rule of thumb (16 × variance ÷ effect², 95% confidence, 80% power) — it agrees with Evan Miller's calculator to within a few percent across the range above. The revenue figure is arithmetic on your own numbers, not a forecast: it shows what the uplift would be worth if it happened, which is exactly the number nobody can promise you in advance.
The second uncomfortable number
Share of A/B tests that actually improve the target metric — at companies with in-house statisticians and millions of users.
“We have been involved in tens of thousands of A/B tests that ran at Airbnb, Booking, Amazon, and Microsoft, and have never seen any change that improves conversions anywhere near this amount.”
Two paths
Not by your budget. We'll tell you which in the first call — even when the cheaper one earns us less.
Testing can't work for you, so we don't measure — we decide well.
You have the volume for real tests, so we run them properly.
When we can't measure, we can still be systematic. Every page in the funnel is read against four principles, in order — because a page almost never fails at the last one.
I understand it
What is this, who is it for, what happens next. Most drop-off is here, in the first four seconds.
I believe it
Proof, faces, specifics. Without it the form reads as a risk rather than a next step.
I want it
The offer stated in the visitor's terms, with the outcome ahead of the feature list.
I act
One next step, no competing links, a form short enough that nobody stops halfway.
Want to know which path your traffic puts you on?
The cheaper question first
Most enquiries that reach us asking for a new site don't need one. They need the one screen where people are leaving fixed. Here's the honest comparison — including when a rebuild genuinely is the answer.
Transparency
These three are standard practice in CRO. Every one of them produces numbers that look good and mean nothing.
“We'll lift your conversion by 87%” isn't a forecast, it's a sales line. Kohavi has never seen anything of that magnitude across tens of thousands of tests — realistic is 0.1–1% per winning test. We guarantee method and transparency, not a number nobody can know.
Checking a running test and stopping at “95% significance!” is how most agency wins are made. It pushes the true false-positive rate from 5% to 26.1% — one in four “winners” is then noise. We fix the sample size in advance and stick to it.
From Microsoft's research: test five treatments and the false-positive rate goes from 2.5% to 12%. Six rounds of five and the chance of a “significant” result exceeds 50% — even when the change provably does nothing. That is how success stories get manufactured from dead hypotheses.
Your investment
In our research, not one German-speaking CRO agency published a real rate card. Some run a page titled “What does CRO cost?” that contains no figure at all. Here are ours.
Conversion audit
€990one-time, credited against your first month
Afterwards you know where the losses are — and whether testing is even possible for you.
Expert-led retainer
€1,290per month, cancel monthly
Expert-led, no testing promises. The audit is credited against your first month.
Testing programme
€2,900per month, cancel monthly
Real A/B tests. Only sensible from ~100,000 visitors a month — we check that first, not you.
All prices excl. VAT · The audit (€990) is credited in full against your first month if you continue with a retainer · Cancel monthly, no minimum term, no setup fee · Swiss clients are billed in CHF.
The verifiable DACH benchmarks: €1,500–3,000 a month at entry level (Löwenstark) and a recommendation to plan with at least €3,000 a month (suxeedo). Speero in the US charges $10,000. We sit below all of that — for three reasons, none of which is “less quality”.
Retainers at a time, maximum. Three of those slots are open this quarter — which is also why we can afford to turn work down.
Start with the audit. If it says don't bother, you've saved a year of retainer.
Who does the work
The difference between an amateur and a professional in CRO isn't who tests more. It's who knows when testing is nonsense.
“Most agencies sell A/B testing to everyone because it sounds like science. For most clients it measures noise. I'd rather tell you your traffic isn't enough — and then do the work that actually moves things.”
Why there are no result numbers here
Numbers like that are on almost every CRO page. They're nearly always made the same way: a test with too few participants, stopped early, at the right moment. At low statistical power a significant result systematically exaggerates the effect — it's called the winner's curse. A meta-analysis of 115 public CRO tests found a mean uplift of 3.77%, only 46% significant at all, and around 70% underpowered. The “winners” in it averaged nearly three times the true typical effect.
This page itself carried “87% average conversion increase” in its description until recently. That wasn't provable, and we removed it — rather than hiding it behind small print, as is customary in this industry.
What we offer instead: the maths on your own numbers, before you pay. If your traffic is enough, we test — and you get real results at honest magnitudes. If it isn't, we say so and do the work that helps without testing. What we won't do is promise you a number nobody can know.
FAQ
Because it's maths, not opinion. To reliably detect a 10% uplift at a 2% conversion rate you need roughly 78,000 visitors per variant — over 156,000 per test. At 5,000 visits a month, a single test would finish in two and a half years. By then your site, your offer and your market have changed three times over. A test that runs shorter isn't measuring your offer, it's measuring chance — it produces a number but no information. This is exactly why Speero ($10,000 a month) turns away clients under 100,000 visitors.
Because nobody can know that before the work is done. Ronny Kohavi, who built the experimentation platform at Microsoft and Bing, writes after tens of thousands of tests at Airbnb, Booking, Amazon and Microsoft that he has never seen magnitudes like that. Realistic is 0.1–1% per winning test — and only about one test in ten wins at all. Anyone quoting you a percentage upfront either invented it or derived it from an underpowered test that systematically exaggerates it. We guarantee method, not numbers.
Honest answer: informed design decisions, not measurement. We see in session recordings where people drop out, we test with real people from your audience, and we know the patterns that make funnels fail — broken forms, unclear offer, missing trust, load time. These are real, fixable problems that need no statistics. What we won't claim: that we can prove to you afterwards exactly how much it gained. Anyone promising both at your traffic — impact and proof — is promising something impossible.
Rule of thumb: from around 100,000 visitors or 1,000 conversions a month. Conversions matter more than visitors — the threshold sits at roughly 1,500 per variant for a 10% effect, almost regardless of your baseline conversion rate. We run that calculation with your real numbers in the audit rather than applying a blanket rule. If you're just below, there are ways through: test bigger changes rather than smaller ones (at a 50% target effect the requirement drops to a twenty-fourth), or measure conversion to the next funnel step rather than the final sale.
At low traffic: honestly, fairly close. The difference is the evidence base — we don't guess at what looks nice, we look at where users actually drop out and work through patterns we know from hundreds of funnels. At high traffic it becomes a discipline of its own, because then you can measure instead of believe. Anyone selling CRO at 3,000 visitors as “data-driven science” is selling you web design with a dashboard in front of it.
Because it's real work: funnel analysis, reviewing recordings, expert review, power calculation on your numbers, a prioritised action list. A free audit is either superficial or a sales call in disguise. If you book a retainer afterwards, we credit the full €990 against your first month — so it was effectively free. If you don't, the report is still yours, action list included, to implement yourself or with another agency. And “conversion work isn't worth it for you right now, put the money into reach” is a perfectly legitimate audit outcome.
Yes. No notice period, no exit fee, no contract. One email and you're out at the end of the month. In an industry where 12-month lock-ins are standard, that's unusual — and it's a very direct incentive on our side.
Yes — and those are often the clients where real testing becomes possible, because the traffic and conversions are there. For shops we have a dedicated page under online shops. For B2B SaaS with long sales cycles, combining conversion work with AI visibility usually makes more sense than testing alone — there the shortlist decides, not the button colour.
Conversion audit for €990, credited against your first month. You get the funnel analysis, the prioritised action list and the power calculation on your real numbers. Afterwards you know whether testing is even possible for you — and the report is yours, whether you continue or not.
Three retainer slots open · cancel monthly · reply within 24 hours.
The audit — credited in full against your first month. The report is yours either way.
