A good website conversion rate is usually somewhere between 2% and 5% of sessions, but only if you compare like with like.
Ruler Analytics' 2026 study of more than 110 million sessions puts the cross-industry average at 5.13%, with retail and ecommerce at 2.4% and legal at 7.9%.
A dedicated landing page usually converts higher: Unbounce's 2024 report gives a median of 6.6%.
So "good" depends on what you count as a conversion, which traffic you send and what you sell.
This page sets out the published benchmarks by site type, explains where they come from, and shows real before-and-after rates from tests in our database so you can see what moved them.
What website conversion rate means (and how to calculate it)
Website conversion rate is the share of visits, or visitors, that complete the action you care about: a purchase, a form submission, a call, a trial signup. The formula is simple:
conversion rate = conversions / sessions (or users) x 100
The denominator matters more than most people think.
In Google Analytics 4, Google defines session key event rate as "number of sessions in which a key event happened / total number of sessions", and user key event rate as users who performed a key event divided by total users.
The user-based rate is always the higher of the two for the same site, because one person can visit several times before converting.
| Basis | Example | Result |
|---|---|---|
| Per session | 300 leads / 12,000 sessions | 2.5% |
| Per user | 300 leads / 8,000 users | 3.75% |
| Per landing page visit (paid traffic only) | 300 leads / 4,000 ad clicks | 7.5% |
Same site, same month, three honest numbers. Before you compare yourself to any benchmark, check which denominator it uses.
Our conversion rate calculator does the arithmetic, and the conversion rate glossary entry covers the definition in more detail.
What is a good conversion rate? Benchmarks by site type
Here are the most-cited current benchmarks, with what each one actually measures. They are not interchangeable.
| Source | What it measures | Headline figure |
|---|---|---|
| Ruler Analytics, 2026 | Whole-site sessions to qualified leads or sales, 13 industries, 110M+ sessions | 5.13% average |
| Unbounce, 2024 | Visitors to conversions on 41,000+ landing pages | 6.6% median |
| LocaliQ / WordStream, 2026 | Search ad clicks to leads (calls, forms, chats, emails) | 8.18% average |
The gap between these numbers is mostly about scope.
A whole-site rate includes blog readers, job seekers and returning customers who were never going to convert.
A landing page rate counts only people who arrived with intent, often from an ad. An ad-click rate is narrower still.
By industry, the Ruler Analytics data shows how wide the spread is:
| Site type (Ruler Analytics, 2026) | Average conversion rate |
|---|---|
| Legal | 7.9% |
| Automotive | 7.9% |
| Software | 7.6% |
| Finance | 6.3% |
| Education | 6.3% |
| Marketing and advertising | 6.2% |
| Professional services | 6.1% |
| Construction and engineering | 4.9% |
| Beauty and cosmetics | 3.5% |
| Real estate | 2.8% |
| Retail and ecommerce | 2.4% |
| Health and social care | 2.3% |
| Travel | 1.9% |
Ruler's figures are for qualified leads or sales across online and offline conversions, so a store that only tracks checkout completions should compare against the retail line, not the headline average.
For store-specific numbers, see our ecommerce conversion rate benchmarks, and for subscription software, our SaaS conversion rate benchmarks.
The full cross-industry breakdown lives on conversion rate by industry.
What real stores in our database started from
Published averages hide how much individual sites vary.
These are real baselines from ecommerce tests in LiftArchive:
- Simple Modern ran a seasonal collection landing page test that went from 0.31% to 0.47% conversion. Well under 1%, and still a clear win.
- VWO's case with Grene moved the homepage conversion rate from 2.64% to 3.18% by reworking the USP icons.
- SplitBase's Fontana Candle redesign lifted a store already at 4.88% to 5.43%.
- Oliver Kenyon's Purau landing page redesign went from 2% to 7% conversion.
A 0.31% page and a 4.88% site can both be healthy.
What counts as good is whether your rate is rising against your own baseline at a cost you can afford.
@oliverkenyon
SplitBaseWhen a benchmark is the wrong yardstick
- Your traffic mix changed. A spike in cold social traffic lowers your rate without anything on the site getting worse. Ruler's channel data shows organic social converting at 2.2% against 5.4% for paid search.
- Your price point is high. A $3,000 product will convert lower than a $30 one, and should.
- You count micro-conversions. Newsletter signups or "add to cart" inflate the rate compared with benchmarks that count leads or sales.
- Your volume is small. At 50 conversions a month, week-to-week swings of a full percentage point are normal noise.
Lead conversion rate: two different numbers
"Lead conversion rate" gets used for two separate things, and mixing them up causes bad decisions.
- Visitor-to-lead rate: the share of visits that become a lead (form fill, call, quote request). This is what the benchmarks above measure.
- Lead-to-customer rate: the share of leads that become paying customers. This lives in your CRM, not your analytics tool, and depends on lead quality and your sales process.
The two often move in opposite directions. Make a form easier and the visitor-to-lead rate goes up, but some of the extra leads may be weaker.
Add a qualifying question and you may get fewer, better leads.
A MarketingSherpa case found that adding an extra field to an Act-On incentive form caused a significant drop in leads, which is the cost side of that trade.
The useful metric is the product of both: visits to customers.
If you can, report a lead-to-customer rate alongside every visitor-to-lead test result.
What lead pages in our database started from
Lead-gen baselines in LiftArchive range more widely than ecommerce ones:
- Seer Interactive's water treatment client had a contact form converting at 1.16%, and showing the form open lifted it to 1.65%.
- Venture Harbour's multi-step enquiry form went from 0.96% to 8.1% conversion.
- Open Mile's lead page masthead redesign took the lead rate from 3.95% to 13.11%.
- Ultraviolet Digital's Kaya Skin Clinic test, which reworded the form CTA around getting an expert opinion, went from 4% to 9.5%.
- A KlientBoost client's landing page already converted at 33.22%, and adding social proof and testimonials moved it to 43.52%.
A 1% contact page and a 33% landing page are not comparable.
The second gets warm, single-offer paid traffic; the first serves everyone who lands on a corporate site.
Lead generation conversion rate: how to raise yours
For lead-gen sites, the dedicated landing page median of 6.6% from Unbounce and LocaliQ's 8.18% average for search ad clicks are reasonable reference points for paid traffic.
For the whole site, the 5.13% Ruler average is closer.
Below those, the tests in our database point to a few patterns that repeatedly moved the number.
Break long forms into steps
Apexure's work for Reclaim My PPI Tax is a good example of compounding gains.
A shorter landing page went from 1.02% to 1.77%.
A four-stage claim form then took it from 1.77% to 4.33%, and a gateway page from 4.33% to 6.09%.
Three tests, roughly six times the starting rate.
More examples are on our multi-step forms page.
Put the offer and form where people can act
TheHOTH's minimal homepage signup form went from 1.39% to 13.13% sign-ups.
Warner Music Group's signup form placement test on artist sites went from 6.2% to 18.4%.
Both cut the distance between arriving and converting.
Ask for less, or ask better
Field count and field design are the most-tested levers in lead gen.
Removing a field you never use in sales follow-up is usually a safe first test; removing one your sales team relies on can raise the lead rate while lowering the close rate.
Browse the form optimization tests and the full lead-gen hub for what worked and what didn't.
KlientBoost
ApexureApexureApexureTest it properly before you call it a lift
Low baselines need a lot of traffic. Take a lead page converting at 3%.
To detect a 20% relative lift (3% to 3.6%) with 95% confidence and 80% power, the standard two-proportion formula gives:
n per variant = (1.96 + 0.84)^2 x [0.03 x 0.97 + 0.036 x 0.964] / (0.006)^2 = about 13,900
That is roughly 27,800 visitors across both versions.
At 2,000 visitors a week, it's a 14-week test.
If you don't have that traffic, test bigger changes (like the form and page rebuilds above) instead of button colors, or measure on a higher-volume step.
Plug your own numbers into the A/B test sample size calculator.
For a broader view of what lead-gen tests tend to deliver, see our lead-gen A/B test results research, or search all results in the test database.