Conversion Rate Calculator
With The Sample Size Named

What A Conversion Rate Tells You

Take the conversions, divide by the visits, and write it as a per cent. Those two figures are the whole formula. It is the easiest number in marketing to work out and the hardest to read.

The hard part is the bottom half. People who arrive from a brand search buy far more often than people who arrive from a display ad, so one mixed rate describes neither.

The Nielsen Norman Group puts the usual range for conversion rates somewhere between one and ten per cent, and adds that there is no single right answer.

So read a rate against your own history. An average from other sites tells you what their traffic did, and nothing more.

Value Per Visitor Beats Rate

A rate ignores what a sale is worth. Two per cent at ninety pounds beats four per cent at twenty, and the rate alone hides that.

Value per visit folds both together: the rate times what one sale is worth. Put that figure on your dashboard.

It also makes budget choices easy. A source worth four pounds a visit deserves more than one worth ninety pence, whatever their rates look like.

Why Small Tests Lie

Two versions, forty visits each, one sale apart. Any tool reports a hundred per cent lift, and the number means nothing at all.

Conversions carry the confidence, not visits. As a rough rule, wait for a few hundred per variant before you believe a small difference.

This page says so when your numbers are too small. The lift still shows, since hiding it would be its own kind of lie.

Stopping A Test Early

Watching a test and stopping the moment it looks good is how a team fools itself. Pick the sample size first, then let it run.

Percentage Lift Versus Point Change

Going from 2% to 2.4% is a lift of twenty per cent and a change of 0.4 points. Each claim is accurate, yet they land very differently in a meeting.

Most tools report the larger of the two, which is the lift. Read it beside the plain change in points before you promise anybody anything.

The yearly figure here does the same job in money: the same change, times your traffic, over twelve months.

What Moves A Rate, In Order

Traffic quality first. The same page converts three times better for people who arrived searching your brand than for people who clicked a banner.

Then the offer. Price, delivery cost and the promise in the heading move a rate further than any button colour ever has.

Then friction. A shorter form, a guest checkout and a page that loads in two seconds rather than six all show up in the number.

Design comes last, and usually smallest. That order is the opposite of how most teams spend their week.

Reading A Rate Over Time

A rate has a season. Shopping weeks, holidays and payday all move it, so this month against last month can mislead badly.

Compare the same period a year earlier where you can. Where you cannot, compare a rolling four weeks against the four before it.

Watch the traffic mix as well. A rate that fell while paid traffic doubled has not necessarily got worse at all.

And check the tracking whenever a number jumps overnight. Real changes arrive gradually; measurement changes arrive at midnight.

Where The Rate Hides A Problem

A steady rate with falling sales means traffic dropped, and the rate will never say so on its own.

A rising rate with falling revenue usually means the mix changed. Cheap items convert more easily, so the count goes up while the total goes down.

A rate that looks perfect on desktop and poor on a phone points at one form, one button or one slow image.

So keep the counts on screen beside the percentage. Two numbers explain far more than one ever does.

Segment Before You Celebrate

A whole-site rate hides everything. Phone against desktop, new against returning, paid against organic: the useful number is almost always in one group.