You ask what a good conversion rate is and you get the same answer everywhere: around two to three percent. Then you measure your own number against it and either relax or start worrying.
The problem is that a conversion rate is not a fixed, universal number. It is a ratio you build yourself, and two simple choices can multiply it several times over without a single extra sale.
What is a conversion rate?
A conversion rate is how many out of every hundred people who had the chance to do a particular thing actually did it. The number of people who did it, divided by the base population, times a hundred.
Say ten thousand people come to your shop’s site in a month and two hundred of them buy. Your purchase conversion rate is two percent.
But that is a ratio, not a raw figure, and it has two halves that you choose yourself. The numerator is which action you decided to count as a conversion. The denominator is which population you took the number from. Everything turns on those two choices.
Why “what percentage is good?” has no answer
On one single site, that number changes several times over depending only on where the visitor came from. In ecommerce benchmarks, traffic arriving from email converts at around 4.2 percent on average, while the same measure for paid social traffic is around 1.1 percent.
Those two numbers, nearly four times apart, belong to one site, not two different businesses. All that differs is the population making up the denominator. Someone who clicked a link in your email already knew you and was one step closer to buying, while someone who saw your ad mid-scroll has just met you for the first time.
So when one site’s number depends this heavily on its traffic source, comparing your overall rate with a competitor’s overall rate is like comparing the average marks of two schools when one of them only admits top students and the other admits everyone.
The numerator: which action counts as a conversion?
A conversion is not necessarily a completed purchase. It can be adding to the basket, signing up to a newsletter, filling in a contact form, or even just opening the contact page. On one fixed population, whichever you pick produces a completely different number.
Picture an online sports equipment shop. If you define conversion as “anyone who added a product to their basket”, you might land on ten percent and feel like everything is going well. If you define it as “anyone who completed the payment”, you might land on one and a half percent.
Neither is a lie, but they describe two very different realities. So before you ask what your conversion rate is, decide which action actually matters for the decision in front of you today, and then hold it fixed.
The denominator: which population are you taking it from?
The second half shifts the number even more easily. Is your denominator all visits, or only unique visitors? Do you count each person once, or every time they came back? Do you include everyone who landed on the site, or only those who reached a product page?
To see how much this matters, look at a large stream of traffic. In its 1403 annual report, Digikala said it averages 5.5 million visits a day. Inside that crowd is someone who only came to check the price of a phone and leave, someone who has been hunting one specific fridge for ten days, and someone who does the same weekly shop every week.
A single number over a mixed crowd like that hides all three behaviours under one average. Until you narrow the denominator to one comparable group, the figure is more decoration than a tool for deciding anything.
So what should you do with it?
- Write yourself one fixed definition. State exactly which action is the numerator and which population is the denominator, and always measure it the same way, so that this month’s number and last month’s are genuinely comparable.
- Measure yourself against yourself, not a competitor. You know nothing about their denominator or their traffic mix, so the only fair comparison is your number this month against your own earlier months.
- Break the buying path into steps. Work out the rate for each step separately, which is what a sales funnel does, so you can see exactly where people drop off.
Once you have split the number by step and by source, it stops being a vague grade and turns into a map that tells you where to fix things first.
So stop looking for a universal standard. This is a number you build yourself by choosing the target action and the base population. Define those two clearly, measure yourself over time, and look at each step’s rate separately instead of one overall figure, so you can see where you are losing the most people.