Growth Marketing / CRO
Conversion Rate Optimization
More revenue from the traffic you are already paying for.
The problem
If 1% of your visitors convert and you double that to 2%, you have doubled your revenue without spending an additional dollar on traffic. This is why CRO is the highest-leverage growth activity most businesses never prioritise.
Most businesses do not know their conversion rate. They do not know where visitors drop off. They have never watched a session recording of a real visitor trying to complete a purchase. The data is available — they just have not looked at it.
Our approach.
What the CRO research phase covers
Guessing is the enemy of conversion work, so the first month is research, not redesign. We watch session recordings of real visitors moving through your key flows and log every point of hesitation, backtrack, and rage-click. We analyse heatmaps and scroll maps to see what visitors actually notice versus what the page assumes they notice. We map the full funnel in analytics to find the exact step where the biggest share of value leaks out. And we run moderated user tests — real people from your target audience attempting real tasks while thinking aloud.
The research converges on a diagnosis: not "the page could be better" but "visitors abandon at the delivery-cost reveal" or "mobile users cannot find the specification they need to justify the purchase". Specific problems produce testable solutions.
How hypotheses are prioritised
Every proposed change is written as a falsifiable hypothesis — "if we do X, metric Y will improve, because research finding Z" — and scored on three axes: expected impact, strength of supporting evidence, and effort to implement. The roadmap runs highest-score first.
This discipline matters because CRO budgets die on pet theories. Scoring forces every idea, including ours and yours, to compete on evidence. A stakeholder's favourite redesign idea and a finding from session recordings enter the same queue and are ranked by the same rules.
How tests are run and judged
Before any test launches, we calculate the required sample size and minimum run time, and we agree the decision rule in advance. Tests run full weeks to avoid day-of-week bias, and we do not peek early and call winners — the two practices that make most agency A/B "wins" evaporate after rollout.
A test concludes in one of three ways: a winner at 95% confidence (rolled out to all traffic), a loser (documented, because knowing what does not work is worth money), or an inconclusive result (retired, and the hypothesis refined). Every outcome feeds the learnings log, so the programme gets sharper with each cycle.
Beyond the button: structural CRO
Some conversion problems cannot be solved by testing variations of an existing page — the page itself is built on the wrong argument. When research points there, we redesign the page structure: the order of information, the objections addressed, the proof presented, and the action requested. The redesign is then validated as a test against the original, so even structural changes are held to the same evidential standard.
Where the friction lives in forms and checkout, we go field by field: removing questions that are not used, fixing error messages that blame the user, and reviewing the mobile experience on real devices — because that is where most of your traffic is, and where most conversion damage hides.
Deliverables.
CRO audit and research
Session recordings, heatmaps, funnel analysis, form analytics, and user testing synthesis. We find out why visitors are not converting before guessing at solutions.
Prioritised test roadmap
Hypotheses ranked by expected impact and ease of implementation. We run the highest-leverage tests first.
A/B testing implementation
We set up, run, and analyse A/B tests using Google Optimize or a similar tool. Statistical significance required before declaring a winner. No gut-feel conclusions.
Landing page redesigns
Where testing reveals that a page needs a significant structural change, we design and build the new version. Not just a button colour change.
Form and checkout optimisation
Friction reduction in your highest-value flows. Every unnecessary field removed. Every error message made human. Mobile UX reviewed on real devices.
Winning implementation and monitoring
Winning variants rolled out to 100% of traffic. We monitor for regressions and unexpected interactions with other page elements.
What happens when.
Month 1 — Research
- Session recording analysis (2 weeks)
- Heatmap and scroll map review
- Funnel drop-off analysis
- User testing (5 participants)
- Hypothesis generation and prioritisation
Month 2 — Test
- Top 3 hypotheses implemented as A/B tests
- Weekly data review
- Early stopping criteria applied
- First winning variant identified
Month 3 — Iterate
- Winning variant shipped
- Next round of tests launched
- Conversion rate improvement documented
- Ongoing test roadmap updated
Pricing
Every engagement is scoped to the account. What you pay depends on the number of markets, products and campaigns in play, and on how much needs rebuilding before we start. You get a number on the first call, in writing, before you commit to anything.
Get a scoped quoteCRO questions answered.
A minimum of 1,000 monthly visitors per variation being tested for statistically valid results. Below this threshold, tests take too long to reach significance and results are unreliable — in that case we use the research-led sequential approach instead of A/B testing, so lower-traffic sites still get evidence-based improvements.
We use VWO, Google Optimize (where still available), or Optimizely depending on your tech stack and budget. For e-commerce, we often use native Shopify A/B testing or a headless alternative.
Yes. Shopify stores are one of our most common CRO environments. We work within Shopify's constraints and know which elements have the highest leverage for e-commerce conversion.
We require 95% statistical confidence before declaring a winner. We use a sample size calculator before launching every test to know how long it needs to run.
No. CRO is hypothesis-driven — some tests win, some lose. What we can tell you is our average conversion rate improvement across engagements longer than 90 days is 34%. We show this data openly.
Both in parallel, ideally. CRO makes every traffic source more efficient. If you are spending heavily on ads, a 50% conversion rate improvement doubles your ROAS without changing the ads.
We switch methods rather than force invalid tests. For lower-traffic sites we rely on the research toolkit — session recordings, user testing, heuristic review, and form analytics — and ship improvements sequentially, measuring before-and-after over longer windows. Less statistically pure than A/B testing, but far better than doing nothing or trusting tests that can never reach significance.
Part of Growth Marketing
Growth Marketing overviewRelated services