What CRO is, and what it is not
Conversion rate optimisation is often sold as button colours and headline tests. That is the smallest part of it. Most websites lose enquiries for bigger reasons: the visitor cannot quickly tell whether you serve someone like them, the price question is avoided, the form asks for too much, or the next step is unclear.
CRO is not a redesign for its own sake, and it is not a set of persuasion tricks. Countdown timers that reset, fake scarcity and pre-ticked boxes may lift a number for a week, but they erode trust and can fall foul of consumer protection rules and the ASA's advertising codes. For a company that competes on quality, the aim is the opposite: remove doubt by being clear.
Start with research: where and why visitors drop out
Before anything changes, we want to know what is actually happening. Typical sources:
- Analytics. Which pages and traffic sources bring visitors who convert, and where the others leave.
- Funnel and form data. Which steps of a checkout or enquiry form lose people, and on which devices.
- Session recordings and heatmaps. Used sparingly and with consent, to see confusion that numbers cannot show.
- Your own team. Sales and front-desk staff hear the same objections every week. Those questions belong on the website.
- Customers. A few short conversations or a one-question survey often explain more than a month of data.
A website audit can be a useful starting point when technical problems such as slow pages or broken forms are part of the picture.
From findings to priorities
Research produces more ideas than anyone can act on. Each finding becomes a hypothesis: because we saw this, we expect that changing this will lead to that. We then rank them by likely impact, confidence and effort, so the first changes are the ones with the most evidence behind them. Opinions from the loudest person in the meeting do not make the list without data.
Test, implement or simply fix
Not every change needs an A/B test. Choosing the right route saves time and avoids false conclusions.
| Route | When it fits | Example |
|---|---|---|
| Just fix it | The problem is plainly broken or confusing | A form error that wipes all fields, a missing delivery cost, a dead link |
| Implement and monitor | Strong evidence, but too little traffic for a reliable test | Adding pricing guidance or a clearer process section to a service page |
| A/B test | Enough traffic and conversions to reach a reliable result | Two versions of a product page layout on a busy online shop |
Being honest about traffic and testing
Many UK service businesses do not have enough monthly conversions for statistically sound A/B testing. A test that is stopped early, or run on too few conversions, often shows a winner that is really noise. We say so plainly, and in those cases we work from research, best practice and careful before-and-after comparison instead.
We also do not promise a percentage uplift. The result depends on your starting point, your offer, your traffic and how much changes. What we can promise is a clear method and an honest report of what changed and what we learned.
Where service websites lose enquiries
For solicitors, clinics, accountants, consultancies and other professional firms, the same issues come up again and again:
- The homepage talks about the firm instead of the client's problem.
- No indication of cost or how fees work, so serious buyers go elsewhere to find out.
- Generic stock photography instead of the real people a client will deal with.
- A single contact form with eight required fields and no other option.
- Slow mobile pages, especially for visitors arriving from paid search.
Often the answer is a dedicated page for one service and one audience. Our landing page design work is built on exactly these principles, and pairs well with campaigns run by our PPC agency service.
Conversion optimisation for online shops
In ecommerce, the common leaks are different: delivery costs and times revealed only at checkout, unclear returns, weak product images, forced account creation, missing payment methods and slow pages on mobile. For orders to the EU, unexpected import charges on delivery can undo a sale after it has been made. Many of these are fixed at the build stage, which is why our Shopify web design projects plan delivery, returns and payments before design starts.
Measuring with respect for privacy
Good CRO depends on good data, and in the UK collecting it is covered by UK GDPR, the Data Protection Act 2018 and PECR, which sets rules for cookies and similar technologies. In general terms, non-essential cookies need consent, and the rules have recently been updated, so the current guidance from the ICO is the reference point. Practically, this means a consent banner that offers a real choice, analytics configured to respect it, and accepting that your data will never be complete. We set up measurement with that in mind and recommend you confirm your approach with your data protection adviser.
How a CRO engagement works
- Strategy session. A paid 50-minute conversation with the founder about your goals, your website and whether we fit.
- Proposal. A fixed scope and fixed price if we take the work on.
- Measurement check. We make sure conversions are tracked correctly before drawing any conclusions.
- Research and priorities. Findings, hypotheses and a ranked list of changes.
- Changes and evaluation. Fixes, implementations and tests, each reported with what we expected and what happened.
If the research shows the site itself is the problem, a broader web design project may be the more honest recommendation.