That's the whole recommendation. The rest of this guide shows you how to set it up, what the paid options cost and how to tell when you've outgrown the free ones.
What to use for each job
Your conversion rate is the share of visitors who buy, book or sign up. Tools for raising it look like one crowded category, but they do five different jobs. You only need the ones that match the question you're asking right now.
| The question | The job | Start with | Cost |
|---|---|---|---|
| Where are people leaving? | Analytics (reports on your visitors) | Google Analytics | Free |
| What did they do before they left? | Recordings of visits, and heatmaps (pictures of where people click) | Microsoft Clarity | Free |
| Why did they leave? | A survey on your site, plus watching real people | Contentsquare's free plan (formerly Hotjar) | Free |
| Did my fix actually work? | A/B testing | Nothing yet, unless you have enough visitors to pass the test below | PostHog's free allowance, or from about $30 a month |
| Can I help them before they go? | Questions and guidance on the page | Do the first three first | Varies |
Work down that list in order. Each step tells you what to look for in the next one. Buy a testing tool first and you'll have nothing better to test than button colors.
A word on prices
Prices in this category change often, and some companies don't publish them at all. So every price below is from the company's own page on 4 November 2026. Look again before you pay.
Job 1: find where people leave
Before anything else, you need one number. If you don't know your website conversion rate yet, start there, because everything else in this guide depends on it. Pick the one action that pays your bills: a purchase, a booking or a sign-up.
Google Analytics is free. If your store or booking platform has its own reports, use those too, because they're already set up.
The report you want in Google Analytics is called a funnel exploration. A funnel is just the steps someone takes to do business with you. Go to Explore, choose the Funnel exploration template and add your steps. For a store, that might be product page, add to cart, checkout, purchase. For a service business, it might be service page, contact page, form sent.
Look for the step with the biggest drop. That's your job for the next month. Everything else can wait.
Job 2: see what they did
Analytics tells you 80 people left the pricing page. It can't tell you they kept tapping a price that wasn't a link.
Microsoft Clarity is the obvious place to start. Microsoft says it's free forever with no limit on visitors, and you won't ever be pushed onto a paid version. To install it, go to Settings, then Setup. It has ready-made installs for Shopify, Squarespace, Wix and WordPress, so you may not need a developer.
Once it's running, ignore the dashboard for a week. Then watch 20 recordings of people who reached the page you picked in Job 1. Clarity marks visits with "rage clicks" and "dead clicks." That means someone clicked again and again, or clicked something that did nothing. Start with those, because they're the cheapest bugs you'll ever find.
Three things to know:
- Recordings expire. Clarity keeps them for 30 days, apart from favorites and a random sample that last up to nine months. Mark the good ones as favorites.
- Europe needs permission. Since 31 October 2025, Clarity needs to know that visitors from Europe, the UK and Switzerland have agreed to tracking. Without it, you get limited tracking of those visitors. If that's your market, connect it to your cookie banner.
- Under-18s are out. Microsoft says Clarity shouldn't be used on sites or apps aimed at people under 18.
By default it hides sensitive content, so you aren't watching people type their card numbers.
Job 3: ask them why
Recordings show you what happened. You still have to guess the reason, and your guess will be flattering. So ask.
Put one question on the page where people drop off. Show it after they've been there a while:
What almost stopped you from [buying / booking / signing up] today?
Offer four or five tappable answers plus "something else," because typing on a phone is a chore. Read the "something else" answers first. Those are the reasons you hadn't thought of.
You don't need a survey company for this. Contentsquare's free plan includes basic surveys. Crazy Egg and Lucky Orange both include unlimited surveys on their paid plans.
If you have few visitors, answers will come in slowly. So get five people who could be customers to try it on their phones while you watch and say nothing. Web usability expert Jakob Nielsen has long advised that five people is enough for this kind of test. It's cheap and a little humbling, and you'll probably fix something the same afternoon.
Job 4: test the fix, if you have the visitors
A/B testing tools get the most attention, but they're the least useful for a small team. A test needs a lot of visitors to mean anything.
Here's how long one test takes with two versions of a page, worked out with Evan Miller's free calculator:
| Visitors a month to that page | 2 in 100 buy, and you hope to get 50% more buyers | 2 in 100 buy, and you hope to get 20% more buyers | 5 in 100 buy, and you hope to get 20% more buyers |
|---|---|---|---|
| 1,000 | 7 months | 40 months | 15 months |
| 5,000 | 6 weeks | 8 months | 3 months |
| 20,000 | 10 days | 2 months | 3 weeks |
| 40,000 | 5 days | 1 month | 12 days |
Getting 20% more buyers is already a very good result, and plenty of real changes do less. For a 10% gain when 2 in 100 buy, you'd need about 78,000 visitors for each version.
A good rule of thumb: if you can't finish a test on your main page within about two months, don't pay for a testing tool yet. That's roughly 20,000 visitors a month to that page if 2 in 100 buy, or about 8,000 if 5 in 100 buy.
What to do if you have fewer visitors than that
Change things anyway. You just can't prove it to a statistician, and you weren't trying to.
- Fix what's clearly broken without testing it. A dead button doesn't need an experiment.
- Make big changes. A new offer or a rewritten first screen may make a big enough difference to show up. A new button color won't.
- Change one thing, then compare. Look at the four weeks before and the four weeks after, using the same steps. Write down anything else that changed, like a sale or a new ad campaign. It's weaker proof than a test, but it beats opinion.
Once you know where people drop off, a conversion rate optimization checklist helps you choose what to change first. The fixes themselves are a bigger subject, starting with how to increase website conversion rate from the visitors you already have.
When you are ready to test
Decide how many visitors you need before you start, then wait. It's tempting to stop as soon as one version looks like a winner. But that raises the chance of a false win, where a test says a change worked when it didn't. Evan Miller showed it can go from 5% to as much as 26.1% in the worst case. Most of us find this very hard, and the dashboard is right there.
Also, Google Optimize closed on 30 September 2023, so any old advice to start there is out of date.
What the paid tools cost
Here's what you'd get for the entry price, for the tools a small team could buy without a sales call.
| Tool | Free option | First paid plan | What's included | Watch out for |
|---|---|---|---|---|
| Microsoft Clarity | Free forever | None | Recordings, heatmaps, AI summaries | No A/B testing listed |
| Contentsquare (Hotjar) | 200,000 visits a month | $39 a month billed yearly, $49 monthly, 7,000 visits | Recordings, heatmaps, funnel reports, surveys | Free plan keeps only 1 month of data. Paid keeps 13. |
| Zoho PageSense | Up to 5,000 tracked users | From $12 a month billed yearly | Reports, funnel reports, heatmaps, recordings, form reports | A/B testing starts on Enterprise, from $29.42 a month billed yearly |
| Crazy Egg | 30-day trial | $29 a month billed yearly, 5,000 page views, 50 recordings | Heatmaps, recordings, surveys, funnel reports | A/B testing starts on Plus at $99 a month |
| Lucky Orange | 7-day trial | $32 a month billed yearly, $39 monthly, 3,500 visits | Heatmaps, recordings, surveys, live chat, funnel reports, form reports | Keeps data for 60 days on the first plan. No A/B testing listed. |
| Unbounce | 14-day trial | $22 a month billed yearly, 5 pages, 500 visitors | Landing page builder | A/B testing starts on Experiment at $112 a month billed yearly, 30,000 visitors |
| PostHog | 1 million events (actions it tracks), 5,000 recordings, 1,500 survey responses a month | Pay as you go | Reports, recordings, surveys, A/B tests | Best with a developer around to set it up |
Compare these carefully, because they charge for different things. Visits (also called sessions), page views and tracked users aren't the same. One visitor can view many pages. Work out your own numbers before you compare prices.
Before you buy, divide the monthly price by your profit on one sale. A $99 plan with $30 profit per sale needs a little over three extra sales a month to pay for itself. If nobody on your team will open the tool every week, it won't bring you any.
The bigger platforms
VWO and Optimizely are the big names. VWO has merged with AB Tasty and now sits under the Wingify name. Its Growth plan is described as for small teams, but its page asks you to schedule a demo and shows no price. Optimizely doesn't publish prices either, so you have to ask for a quote. Both are serious products for teams with a lot of visitors. A small team can come back to them later.
Which setup fits your number of visitors
| Visitors a month to your main page | Use | Monthly cost |
|---|---|---|
| Under 5,000 | Google Analytics, Clarity, one survey question, five people watched | $0 |
| 5,000 to 20,000 | The same, plus a paid plan only if you want surveys, chat or longer history in one place | $0 to about $100 |
| Over 20,000 | Add a testing tool, if someone has time to run a test every month | Around $100 to start, rising with traffic |
The last line depends on how many of your visitors buy, so check it against the table above. If 5 in 100 buy, you get there sooner.
Several of these tools now offer AI summaries of recordings or heatmaps, including Clarity and Crazy Egg, and AI conversion rate optimization is becoming its own subject. A summary can save you some watching. Somebody still has to decide what to change and then change it.
Helping people while they're still there
Everything so far reports on people after they've gone. Even the survey only collects an answer. If visitors keep telling you they aren't sure which option suits them, they needed a reply while they were on the page.
That's what we built Possible to do, so take this with the obvious pinch of salt. Possible is our own product. It's an AI built on real sales psychology that asks each visitor the right question at the right moment. It remembers the answers and guides them to a sale, sign-up or booking. It's free to start, and you can read more about our AI sales assistant for online stores. It's worth a look if your visitors really have to choose between options. If one of your steps has a broken button, fix that first.
Your first month
- Today: Install Google Analytics and Clarity. Pick the one action you care about most.
- Week 1: Build your funnel report and find the biggest drop.
- Week 2: Watch 20 recordings of people on that page. Write down what you see, then fix anything broken.
- Week 3: Add your one question to that page. Watch five people try to buy from you.
- Week 4: Make the biggest change the answers point to. Write down the date, so you can compare the four weeks on each side.
After that, keep 30 minutes a week for it in your calendar. That habit will do more for you than any plan upgrade, and it's cheaper too.
Questions about conversion rate optimization tools
What are conversion rate optimization tools?
They're software that helps you find out why visitors don't take the action you want, and whether your changes helped. The main types are reports on your visitors, recordings and heatmaps, on-site surveys, A/B testing, and questions or popups on the page.
Is conversion optimization software different from conversion optimization tools?
No. People use conversion optimization software, conversion optimization tools and CRO tools to mean the same thing. Some companies say "platform" when one product does several of the jobs.
What's the best free CRO tool?
For most small teams, Microsoft Clarity, used alongside Google Analytics. Clarity shows you what people did, and it's free with no limit on visitors. Google Analytics shows you where they left. If you'd rather have a survey tool in the same place as your recordings, look at Contentsquare's free plan.
How much do CRO tools cost?
You can start at nothing. Paid plans from smaller companies start at around $12 to $40 a month, usually with low visitor limits and yearly billing. A/B testing usually costs more, from about $30 to $112 a month on the plans above. PostHog includes A/B tests in its free allowance. The biggest platforms give prices only on request.
Can I do A/B testing with low traffic?
You can run the software, but you won't get a reliable answer unless the change is huge. With 5,000 visitors a month and 2 in 100 buying, spotting a 20% gain takes about eight months. Below that, fix what's clearly broken, make bigger changes and compare before and after.
How long should an A/B test run?
Until it reaches the number of visitors you chose before starting. Use a calculator, then don't stop early because the result looks good. It's also sensible to run for whole weeks, because many businesses get different visitors at the weekend.
Is Google Optimize still available?
No. Google says it ended on 30 September 2023. Google now points people to other companies' testing tools that work with Google Analytics.
Do I need a developer to use these tools?
Usually not to get started. Clarity has ready-made installs for the common website platforms. You may want help with a cookie banner setup, and with PostHog.