The right replacement depends on who builds your experiments. If marketers set up tests in a visual editor, look at a website testing tool. If developers ship changes behind flags, look at an experimentation platform tied to your code and data. Either way, check how each one loads on your pages before you move a single live test.
Mida sells itself as a lightweight way to run website experiments without developers, with personalisation and conversion tracking alongside. If you're moving on, you're probably after one of three things: more depth in how tests are built and read, a tool your developers can own, or something that sits closer to your online shop. Below are seven options, in no particular order, grouped by the job each one does best.
Speed and setup decide more than the feature list
Lightweight is the selling point of the tool you're leaving, and it's the thing most worth protecting. Every testing tool adds a script to your pages, and how that script loads affects what visitors see first. A tool that makes a page flicker, or holds back content while it decides which version to show, can skew the very results you're trying to read.
The second thing to check is how experiments are built and read. Some tools give marketers a visual editor and a results screen they can understand without help. Others expect developers to set up tests in code and read results against data you already keep elsewhere. Neither approach is wrong, but they suit very different teams, so run the same real test through your shortlist before you decide.
- How the script loads, and whether pages flicker while a variant is chosen
- Who builds a test: a marketer in an editor or a developer in code
- Where results are read: inside the tool, or against your own data
- Whether personalisation sits in the same place as testing, or needs a second tool
Three kinds of testing tool
For marketers testing pages
Convert, Webtrends Optimize and Crazy Egg sit closest to the job a lightweight tool used to do. The first two describe themselves around split testing and personalisation on your website. The third adds heatmaps, recordings and surveys, so you can see where visitors get stuck before you decide what to change. These suit marketing teams who want to run tests themselves.
For product and engineering teams
GrowthBook, Kameleoon and Split lean towards teams that ship code. Feature flags, gradual rollouts and experiments sit side by side, so a change can be tested and released from one place. They suit you if developers already own releases and want test results tied to them, and they ask more of your team at setup.
For online shops
Omniconvert pitches itself at Shopify stores, with AI agents that the vendor says find growth opportunities, build campaigns and measure profit. It's the odd one out here, aimed at shop owners rather than at general website testing, so it only makes sense if you sell through Shopify.
Seven experimentation tools side by side
| Tool | Kind of tool | Good for |
|---|---|---|
| Convert | Website A/B testing | A/B tests and personalisation on marketing pages |
| Crazy Egg | Behaviour analysis plus testing | Finding what to test, then testing it |
| GrowthBook | Warehouse-native experimentation | Teams whose data already lives in a warehouse |
| Kameleoon | Experimentation platform | Running tests from first idea through to release |
| Omniconvert | Ecommerce growth for Shopify | Online stores looking past ad metrics |
| Split | Feature management | Safer releases with measured impact |
| Webtrends Optimize | Testing and personalisation | Personalised experiences alongside A/B tests |
The seven, in alphabetical order
In no particular order beyond the alphabet. Each line sums up what the vendor says about its own tool.
Convert
Best for: Marketers who want split testing, personalisation and advanced goals in one tool; the vendor also highlights fast support and monthly billing.
Crazy Egg
Best for: Site owners who want to see what visitors do first, through heatmaps, recordings and surveys, with A/B testing alongside, per the vendor.
GrowthBook
Best for: Product teams wanting experiments, feature flags and product analytics working from their own data warehouse; the vendor says it's open source and transparent.
Kameleoon
Best for: Teams that build, test and roll out experiments in one place, from the first test to final release, according to the vendor; a free plan is named.
Omniconvert
Best for: Shopify store owners after an AI growth engine that, the vendor says, finds opportunities, builds campaigns and measures profit rather than ad return.
Split
Best for: Engineering teams who want feature flags linked to impact data, with alerts that say whether a change made things better or worse, per the vendor.
Want AI doing the heavy lifting in your marketing?
I build the systems that handle the boring 80 percent, so you get your week back. Done properly, with the human kept in.
Webtrends Optimize
Best for: Businesses wanting A/B testing and personalisation together, which the vendor pitches as one platform with no restrictions on what you run.
My advice if your site has modest traffic
Smaller sites send fewer visitors through each test, so results take longer to settle. That makes a tool's reporting more important, not less: you want a results screen that tells you plainly when a test hasn't seen enough visitors yet, rather than one that crowns a version after a quiet weekend.
It's also a good reason to test bold changes, such as a new headline, a shorter form or a different offer, rather than small tweaks that need a lot of traffic to show any difference at all. For ideas on what to change first, see my guide to getting more visitors to convert, and for a wider field, my roundup of A/B testing tools.
A two-week trial plan
- Pick one page that matters, such as your sign-up or demo request page, and write down what you want more people to do on it.
- Install two shortlisted tools on a copy of that page and check whether either one makes the page flicker as it loads.
- Build the same simple test in each, timing how long it takes and noting who on your team had to help.
- Let both run, then read the results screens side by side and ask which one you'd trust to explain to your boss.
- Keep the tool whose setup and reports your team can run without outside help, and remove the other script.
Do you make A/B testing or experimentation software? Tool makers can ask to be considered for a featured place on this list.
Want your tool on this list? Featured spots and reviews here are paid partnerships. Here's how to partner with me.
Frequently asked questions
What's the difference between A/B testing and feature flags?
A/B testing shows different versions of a page or feature to different visitors and measures which does better. Feature flags switch a change on or off for chosen users, which lets developers release safely and test at the same time.
Will a testing tool slow my site down?
It can. Every tool adds a script, and some load in a way that delays or flickers the page. Check this on a real page before moving your live tests across.
How long should an A/B test run?
Until enough visitors have seen each version for the result to settle, and the tool's report should tell you when that point is reached. Don't stop early just because one version is ahead after a few days.
Do I need a developer to run tests?
Not for simple page tests in a visual editor. Code-based experiments, feature flags and tests inside an app usually need a developer to set up.
Do testing tools need cookie consent?
It depends on how the tool stores visitor data and where your visitors are. Rules vary by country, so check with your data protection regulator for your own setup.
Can I move my existing experiments across?
Usually not directly. Export your results and notes, finish or stop running tests first, then rebuild the ones you still need in the new tool.
