Updated on 9 November 2026 by Nicolas Finet, CEO of Sortlist. This guide has been rewritten for 2026: it explains what conversion rate optimisation (CRO) is, what UK conversion benchmarks look like, how testing works (and the statistical mistakes to avoid), what happened to Google Optimize, what a CRO agency costs and how to choose one.
Traffic is expensive, so converting more of it is one of the most effective ways to grow. CRO is the discipline of finding out why visitors do not act, and testing changes that help them. This guide covers the methods, the tools, the costs and the warning signs when you hire an agency.
Key takeaways
- For most UK websites, a sitewide conversion rate of 1-4% is the realistic range; one 2026 benchmark puts the UK ecommerce median at about 1.85%. Compare yourself with your own history and sector, not a headline average.
- Run tests for at least two full business cycles (about 14 days) and reach the planned sample size before you decide; stopping early inflates false positives.
- UK CRO services range from about £500 for a basic audit to £5,000 or more a month for a full programme, with London agencies charging roughly 20-40% more.
What is conversion rate optimisation?
Conversion rate optimisation is a structured process to increase the share of visitors who complete a desired action: buy, book, sign up, request a quote or download. The conversion rate is the number of conversions divided by the number of visitors (or sessions). CRO combines analytics, user research, psychology, design and experimentation.
Why CRO is important
- Traffic is costly: as paid search and social costs rise, buying more traffic is not always a viable way to grow. Improving conversion means you get more from every visitor. See search engine advertising for UK cost benchmarks.
- Compounding value: a 10% improvement in conversion rate increases results on every channel at once.
- Customer insight: testing shows what your audience values, which informs product, messaging and pricing.
- Lower acquisition cost: the same spend produces more customers.
What is a good conversion rate in the UK?
| Measure | Reported range (2026) |
|---|---|
| Typical UK sitewide range | 1-4% |
| UK ecommerce median | About 1.85% (Statista, as cited in 2026 guides) |
| UK ecommerce top quartile | About 3.6% |
| Global ecommerce averages | About 1.6%-2.9% depending on the source |
Different sources define conversion and traffic differently, which is why published averages vary so much. Use them as context only.
What strategies do CRO agencies use?
Research first
Good CRO starts with evidence, not opinions:
- Quantitative analysis: funnels, drop-off points, device and channel differences in your analytics. See web analytics tools.
- Qualitative research: heatmaps, session recordings, on-site surveys, customer interviews and usability tests.
- Technical checks: speed, bugs, browser and device issues. Slow or broken pages are the cheapest wins.
- Heuristic review: expert review of clarity, trust and friction.
The LIFT model
The LIFT model, developed by WiderFunnel, is a framework for diagnosing a page. It looks at six factors that influence conversion:
- Value proposition: is it clear why someone should choose you?
- Relevance: does the page match the visitor's intent and the ad or link that brought them?
- Clarity: is the message, layout and next step easy to understand?
- Anxiety: what doubts, such as security, price or returns, does the visitor have?
- Distraction: what draws attention away from the goal?
- Urgency: is there a genuine reason to act now?
Be careful with urgency. In the UK, false urgency and scarcity claims, including fake countdown timers, can breach consumer law. See marketing ploys and what UK law allows.
A/B testing
An A/B test shows two versions of a page or element to different visitors at random and compares results. It isolates the effect of one change, such as a headline, a form length or a button. Key rules:
- Have a hypothesis: "Because customers worry about delivery costs, showing them near the price will increase add-to-basket".
- Plan the sample size before you start, based on your baseline conversion rate and the smallest improvement you care about.
- Run at least two full business cycles, about 14 days, and end on the same weekday you started.
- Do not stop early because the result looks significant: peeking inflates the false-positive rate.
- Test one major change at a time unless you have the traffic for more.
“Statistical significance alone is not a stopping condition. You must also reach your pre-planned sample size and run for a minimum of 2 full business cycles (14 days).”
Grow Conversions, How long to run an A/B test
Multivariate testing
Multivariate tests change several elements at once and measure the combinations. They need far more traffic than A/B tests, so they suit high-traffic sites. For lower traffic, run sequential A/B tests or use qualitative methods.
Personalisation and segmentation
Different visitors respond to different messages: new vs returning, mobile vs desktop, paid vs organic. Segment results before concluding that a test did not work: an uplift for one group can be hidden by a drop in another.
Testing tools in 2026
Google Optimize, which many teams used because it was free and connected to Google Analytics, shut down on 30 September 2023, and there is no Google-branded successor. Alternatives include:
- Paid platforms: VWO, Optimizely, AB Tasty, Convert and Adobe Target.
- Open-source and developer-focused tools: GrowthBook and Statsig.
- Platform-native testing: many ecommerce and email platforms have built-in A/B tests.
Use Google Analytics 4 for measurement and make sure that consent management does not distort test data. Under UK rules, analytics and testing cookies may require consent, which can reduce the tracked share of visitors; ask your agency how they handle it.
What kind of agency provides the best CRO services?
Past experience
- Case studies with before and after numbers, the number of tests run and the duration, in sectors like yours.
- Clarity on what failed as well as what worked. Many tests do not win, and agencies that hide this are unrealistic.
- References you can call.
Process and expertise
- A research-led approach: they audit before they test.
- Statistical rigour: sample size calculations, test duration and how they treat segments.
- A mix of skills: analyst, UX researcher, designer and developer.
- Reporting on revenue and profit, not only conversion rate.
Partnerships
Partnerships with testing platforms (VWO, Optimizely and others) show experience with the tools but are not a guarantee of quality. Ask which tools they use and whether you will own the accounts and data.
Red flags
- Guaranteed uplifts, such as "we will double your conversion rate".
- Jumping straight to design changes without research.
- Recommending tests on too little traffic, with results declared after a few days.
- No access to data or tools for the client.
- Reliance on dark patterns such as fake scarcity, hidden costs or pre-ticked boxes, which are risky under UK consumer law.
- Long lock-in contracts with no exit.
How much does hiring a CRO agency cost?
| Service | Typical UK cost (2026 guides) |
|---|---|
| Basic audit | From about £500 |
| Full CRO audit with recommendations | Often £1,500-£5,000 |
| Ongoing programme | Up to £5,000 or more a month |
| London agencies | About 20-40% higher than regional equivalents |
A UK guide notes that at larger scale a £2,500 monthly retainer can be an excellent return if conversion gains are meaningful. The cost makes sense only if you have enough traffic and conversions to test, and a business value per conversion that justifies it. If you have fewer than a few hundred conversions a month, start with research, fixes and qualitative testing before you commit to a testing programme. See conversion cost, landing page price and what a conversion optimisation specialist does.
Common CRO quick wins
- Speed up pages, especially on mobile.
- Clarify the main headline and call to action.
- Shorten forms and remove unnecessary fields.
- Show delivery costs, returns and total price early.
- Add social proof: reviews, ratings, recognisable client names.
- Fix technical errors and checkout failures.
- Make contact details and support easy to find.
For more on driving action, see how to drive conversions.
Final thoughts on the importance of a CRO agency
CRO is a long-term discipline, not a quick trick. The right agency helps you ask better questions, test properly and learn from results, whether the test wins or loses. Choose one that is clear about method, honest about uncertainty and focused on your profit, and keep the learning in-house so that it compounds.
Frequently asked questions about CRO
What is a good conversion rate?
For most UK websites, 1-4% is a realistic range, and a 2026 benchmark puts the UK ecommerce median at about 1.85%. Compare against your own baseline.
How long should an A/B test run?
At least two full business cycles (about 14 days) and until the planned sample size is reached.
What replaced Google Optimize?
Nothing from Google. Teams moved to tools such as VWO, Optimizely, AB Tasty, Convert, GrowthBook or Statsig.
How much does a CRO agency cost in the UK?
From about £500 for a basic audit to £5,000 or more a month for a full programme, with London agencies about 20-40% more expensive.
Do I have enough traffic for A/B testing?
You need enough conversions in the tested pages to reach significance in a reasonable time. If not, start with research and fixes.
Can CRO be risky?
Yes, if it uses misleading tactics. UK consumer law restricts false urgency and hidden costs.
Looking for a CRO partner? Post your project on Sortlist and compare UK CRO agencies for free.
Sources: Grow Conversions, Omniconvert, Mida, Google Optimize shutdown, No Worry, UK conversion benchmarks, Kirro, CRO agency UK, Marketing Sales Management, CRO costs. Benchmarks and costs are indicative.
