What Is a Good Conversion Rate? Averages, Benchmarks and Formulas

A good conversion rate is one that beats your own last month and pays for the traffic that produced it. The averages you find online can't tell you that. Each one measures a different thing, on a different group of businesses, and they disagree by a factor of four. This guide, part of our series on how to increase your conversion rate, shows what the numbers mean and how to set your own benchmark.

Why every published average answers a different question

Search "average conversion rate" and you'll get answers from 1% to 10%. They aren't all wrong. They're measuring different things.

Three differences explain most of the gap:

• What counts as a conversion. For a store it's an order. For a landing page it's often a form: an email, a download, a demo request. A form is a much smaller step than a purchase, so its rate is much higher.
• What counts as a visitor. Some tools divide by sessions, some by unique visitors. The same person visiting three times is one visitor but three sessions, so the session-based rate looks lower.
• Who is in the sample. Every benchmark comes from a vendor's own customers. Their customers are not the whole market, and they change from year to year.

Here's what three published sources say, with what each one actually measures:

All three are vendor-reported: each company publishes data from its own customers. That doesn't make them useless. It means a store converting at 1.4% and a landing page converting at 6.6% may be doing equally well, because they aren't doing the same thing.

Before you compare your rate to any average, check what that average counts as a conversion and as a visitor.

The conversion rates people mean (and how to calculate each)

"Conversion rate" is one formula used for many different steps. Before you compare anything, name which one you mean. The formula is always the same: the people who did the action, divided by the people who had the chance to do it, times 100.

A worked example: 2,000 people visited your sales page last month and 30 bought. Your sales conversion rate is 30 ÷ 2,000 × 100 = 1.5%. If 120 of those visitors left their email and 12 of them bought later, your lead conversion rate is 6% and your lead-to-customer rate is 10%.

The website rate is the least useful of the six, because it mixes blog readers, job seekers and returning customers with the people who came to buy. The page-level rates tell you where to work. You can run your own numbers in the free conversion rate calculator.

Name the rate before you compare it: a 1.5% sales rate and a 6% lead rate can describe the same month.

What "good" looks like for the way you sell

The useful question isn't "what's the average?" but "what's normal for a business that sells like mine?" Here is what to look at for four common ways of selling online.

An online store

The conversion is an order, divided by sessions. Littledata's Shopify data is one of the few sources that publishes its sample size and dates, and its median sits at 1.4%. Its mean is 1.9%, higher because a few stores convert far above the rest. Price matters a lot here: a store selling $15 socks and one selling $900 furniture should not expect the same rate, since an expensive purchase takes more visits and more thought.

A course or digital product

The conversion is a purchase from the sales page. We didn't find a public benchmark for course sales pages that publishes its sample and method, only vendor blog posts quoting ranges without a source. Treat any number you see with care. The practical benchmark is your own launch-to-launch or month-to-month rate, measured on the sales page only.

A service sold through a call

The conversion is a booked call, and the sale happens later. This rate is usually higher than a purchase rate, because booking a free call is a smaller step than paying. What matters is the second number too: how many calls turn into clients. A page that books lots of calls with the wrong people can look great and sell nothing.

A SaaS product with a trial

The conversion is a trial signup, and then a paid account. First Page Sage's data shows the tradeoff clearly. Trials with no card get more signups (8.5% of visitors) but fewer of them pay (18.2%). Trials that ask for a card get fewer signups (2.5%) but more of them pay (48.8%). Their own note adds that a single paid month counts as a conversion, so the second number says nothing about churn. Neither model is "better" in general; it depends on what happens after the trial.

Compare yourself only with businesses that sell the same way, at a similar price, and even then treat the number as a rough range.

How to measure your own rate correctly

Your own number is the only benchmark that fits your offer, your price and your traffic. It's only useful if you measure it the same way every time.

1. Pick one conversion. An order, a booked call, a trial signup. Don't mix them in one number.
2. Pick one page. Divide the conversions that came through that page by the visitors to that page. Sales from your email list or from people who never saw the page don't belong in it.
3. Pick one denominator and keep it. Sessions or unique visitors, whichever your tool reports reliably. Switching between them changes the rate without anything changing on the page.
4. Use full weeks. People buy differently on a Monday and a Saturday. Compare four full weeks with four full weeks, not 10 days with 17.
5. Split by traffic source. Visitors from your email list may buy at five times the rate of visitors from a cold ad. A blended number hides which source is changing.
6. Remove the noise. Exclude your own visits, your team's and obvious bots.

Then write the number down, with the date range and how you measured it. Next month's number is only comparable if the method is the same.

A rate measured the same way every month tells you more than any industry average.

The benchmark that matters: does it pay?

There's a second test that no average can answer: does your rate cover the cost of getting visitors?

A simple example, with made-up numbers. Say a visitor costs you $1 in ads, and each sale leaves you $50 after costs. You need one sale for every 50 visitors to break even, which is a 2% conversion rate. At 1% you lose money on every visitor you buy. At 3% you make $0.50 on each one. The same 2% would be poor for someone whose visitors cost $3, and excellent for someone whose visitors are free from their email list.

So "good" has two parts:

• Better than your own last period, measured the same way.
• High enough that the traffic pays for itself, given what a visitor costs you and what a sale is worth.

If your rate passes the second test, you can buy more traffic. If it doesn't, more traffic only multiplies the loss, and the page is where to work first.

A good conversion rate is the one at which every visitor you pay for earns back more than they cost.

When your rate is low: where to look

If your rate is below your own past numbers, or too low to pay for traffic, the cause is usually one of five things. We covered them in detail in why your sales page isn't converting: visitors can't tell what it is, can't tell if it's for them, don't believe it, can't see the next step, or aren't ready yet.

The last two show up most in the numbers. When a page sells several offers, or speaks to several kinds of buyer, many visitors leave because they can't work out which option fits them. That's the problem Snacked works on: a few questions in front of the page, then each visitor goes to the offer that fits. It counts the clicks to each offer out of everyone who opened the funnel, on every plan, and with a short snippet on your thank-you page it counts the sales too. It doesn't bring new visitors; it works on the ones you have.

Find which of the five reasons is costing you the most before you try to raise the number.

Recap

• Published conversion rate averages measure different things on different samples, and they are vendor-reported. Read what they count before comparing.
• Measure your own rate on one page, for one conversion, over full weeks, the same way every time. Last month is your real benchmark.
• A good rate is better than your last one and high enough that the traffic pays for itself.

For more on raising the number once you know it, read our guide to increasing your conversion rate without more traffic. To see which of the five reasons your page fails, run it through the free sales page grader.