What It Actually Means
Simple arithmetic, and both numbers in it are chosen by whoever produces the report, which is why the same underlying performance can be presented as excellent or poor without anybody lying.
How The Same Month Becomes Two Different Numbers
The denominator. All visitors, or only those who reached a particular page? Including or excluding returning visitors? Bots filtered out or not? Counting sessions or people? Each choice moves the figure, sometimes by a multiple.
The numerator. What counts as a conversion. A form submission, or a form submission that turned out to be a real enquiry? A newsletter signup counted alongside a sales enquiry inflates the number and tells you nothing about revenue.
The window. Someone who visits today and enquires next week is counted in some models and not others.
None of this is necessarily dishonest. It does mean a conversion rate quoted without its definition is not information, and asking for the definition is a completely reasonable request that occasionally produces an interesting silence.
What Is A Good One
The honest answer is that industry averages are close to useless for a specific business, because the number depends on what you count, what you sell, what it costs, and who is arriving.
A site selling something considered, expensive and infrequent will convert far lower than one selling something cheap and habitual, and neither figure says anything about the other.
The only comparison that means anything is against your own previous figure, measured the same way. Which is why the definition needs pinning down before the improvement work starts, not after.
What It Hides
A conversion rate is an average, and averages conceal exactly what you need to see.
Traffic from different sources converts at wildly different rates. Combined into one figure, a channel converting well subsidises one converting terribly, and the aggregate looks acceptable while money is being wasted. The useful view is always segmented: by source, by device, by page, by whether they had been before.
Mobile against desktop is the split that most often reveals something actionable. A large gap usually points at something specific and fixable rather than at a general difference in intent.
What To Ask
- What exactly counts as a conversion, and what is the denominator? Before discussing whether the number is good.
- What does this look like by traffic source? The aggregate is nearly always hiding something.
- How many of these became real enquiries? Form submissions are not leads. The gap is where junk traffic lives.
- What is the rate on mobile against desktop? A large difference is usually a fixable problem rather than a fact of life.
The Trap Worth Naming
Conversion rate can be improved by reducing traffic, and this happens accidentally more often than people expect.
Cut a channel that brought many visitors and few enquiries, and the rate rises immediately while the absolute number of enquiries stays flat or falls. The chart improves and the business does not.
That is why the rate should never be looked at alone. The number that pays wages is enquiries, and the rate is only useful for explaining why that number moved. Any report leading with a percentage and not showing the count alongside it is presenting the flattering half.
More terms are in the glossary.
