Conversion rate gets less attention than it deserves. Traffic is visible, it responds to budget, and it is easy to report on. Conversion rate is quieter. But it is the lever that changes the return on every visit you already get, including every visit you have already paid for.
How to calculate conversion rate
A store with 60,000 sessions and 1,200 orders in a month has a conversion rate of 1,200 ÷ 60,000 = 2.0%.
A few definitions to keep consistent:
- Sessions or users. Most analytics tools default to sessions. Users gives a higher rate because one person often visits several times before buying. Either is fine if you never mix them.
- Orders, not transactions with test orders or cancellations. Exclude internal and test orders, and decide once whether cancelled orders count.
- The same site, the same period. Compare like with like. A month containing a sale is not comparable with one that does not.
What a small lift is worth
The calculator above starts with a store doing 60,000 sessions, 1,200 orders, an $80 average order value and a 40% contribution margin.
That store makes $1.60 of revenue per session. Improve conversion by 10%, from 2.0% to 2.2%, and it takes 120 more orders from exactly the same traffic: $9,600 more revenue and $3,840 more contribution a month, with no extra ad spend.
That last part matters. Buying the same 120 orders through advertising would cost whatever your customer acquisition cost is, every month, for as long as you keep paying. A conversion improvement keeps paying without any further spend.
It also compounds with everything else. In the profit equation, conversion rate multiplies traffic, order value and margin. A better converting site makes every future traffic investment worth more.
What is a good ecommerce conversion rate?
You will see figures between 1% and 3% quoted as typical. That range is real, but it is too wide to be much use, because conversion rate depends heavily on things that have nothing to do with how good your site is:
- Price point. A $30 impulse purchase and a $1,200 considered purchase convert at completely different rates.
- Traffic mix. Returning customers and email traffic convert far better than cold social traffic. Change the mix and the blended rate moves, even if nothing on the site changed.
- Device. Mobile usually converts lower than desktop, and most stores now have more mobile traffic.
- Category. Consumables people reorder behave differently from one-off purchases.
The only benchmark that really matters is your own, segmented. Track conversion rate by device and by channel over time. A drop in one segment tells you something. A drop in the blended number often just tells you the mix changed.
Find where conversions are lost
Conversion rate is an outcome. To improve it, you need the diagnostics that explain it. Break the path to purchase into steps:
| Step | Metric | What a drop usually means |
|---|---|---|
| Visit to product view | Product view rate | Navigation, search or landing pages are not getting people to products |
| Product view to add to cart | Add-to-cart rate | Product page: price, photos, information, trust, stock |
| Cart to checkout | Checkout start rate | Surprise costs, shipping thresholds, cart friction |
| Checkout to order | Checkout completion rate | Form friction, payment options, errors, trust at the last step |
Look at each step by device and by traffic source. The biggest opportunities tend to be wherever one segment performs far worse than the others, like a mobile checkout completion rate well below desktop.
What actually moves conversion rate
Most conversion improvements come from removing reasons not to buy, not from clever tricks.
Clarity on the product page. Good photography, specific descriptions, sizing information and honest delivery times answer the questions people would otherwise leave to go and ask somewhere else.
No surprises at checkout. Unexpected shipping costs are one of the most common reasons people abandon. Show shipping costs and thresholds early, on the product page and in the cart.
Trust. Reviews, clear returns policies and recognizable payment options reduce the perceived risk of buying from a store someone may not know.
Speed and stability. A slow or glitchy mobile experience costs sales quietly, because people rarely complain. They just leave.
Fewer steps. Guest checkout, express payment options and fewer form fields all shorten the distance to the order.
Better traffic. Sometimes the fix is not on the site at all. Traffic that was never likely to buy drags the rate down. Better targeting can raise conversion rate and lower costs at the same time.
Treat changes as tests where you can, and judge them on revenue per session rather than conversion rate alone. A change that raises conversion by pushing people toward cheaper products may not be an improvement. More on this in CRO as a system, not a tactic and our conversion rate optimization service.
Revenue per session: the fuller picture
Conversion rate ignores how much people spend. Revenue per session does not.
In the example, 2.0% × $80 = $1.60 per session. A change that raises conversion rate to 2.2% but drops average order value to $70 gives $1.54 per session, which is worse. Pair conversion rate with average order value and you will not be fooled by that trade.
Common mistakes
- Reading the blended number alone. Mix changes hide real problems and create fake ones.
- Comparing sale periods with normal periods. Promotions inflate conversion rate temporarily and say little about the underlying site.
- Chasing conversion rate with discounts. A discount will raise conversion rate and can lower profit. Check contribution margin before calling it a win.
- Calling tests too early. Small stores need longer tests to reach a trustworthy answer. A week of data on a few hundred orders is mostly noise.
Put it to work
Run your last full month through the calculator above, then break the same month down by device and by channel. The segment with the biggest gap between where it is and where your best segment is, is usually where to start.