Your funnel leaks. We find where.
Roughly seven in ten carts are abandoned, and about two thirds of the checkout sessions that begin end without a sale. That is not a traffic problem. It is a measurable one.
Stop buying traffic to cover a leak.
Almost every ecommerce team we meet has the same instinct when revenue goes flat: buy more traffic. It is the fastest lever to pull and the easiest one to justify. It is also the most expensive place to look for growth, because the traffic is usually not the problem.
A funnel leaks in specific places, for reasons that can be measured. Most of those reasons never show up in a monthly report, because the report averages them away. A four percent conversion rate hides the product page that converts at nine and the one that converts at one.
We instrument the funnel stage by stage, find where the money is actually going, fix the highest value leak first, and prove the fix worked with statistics that hold up in a board meeting. Where the evidence is thin, we say so rather than dressing it up.
Funnel instrumentation
GA4 and event tracking that measures every step properly: product view, add to cart, checkout start, purchase, and the drop between each.
Behavioural analytics
Session replay, scroll and click behaviour and form analytics, read for cause rather than collected as anecdotes.
Testing with real statistics
Sample size and duration worked out before the test starts, so you know in advance what a result will and will not be able to tell you.
Attribution you can defend
Where last click and platform reporting disagree, we show you which number is closer to the truth, and why the gap exists.
The monitoring system behind the work
CRO usually stops when the report is delivered. Ours keeps running, on a monitoring system we built for exactly that.
It connects to GA4, captures a baseline window before a change goes live, then polls every fifteen minutes and compares what is happening now against that baseline. Twelve metrics are tracked across revenue, conversion, funnel, engagement and lead generation: revenue per session, session to add to cart, session to purchase, product page to add to cart, add to cart to checkout, checkout to purchase, landing page bounce, form views, lead submissions and contact clicks.
Every metric carries its own threshold: a minimum sample size, a minimum effect and a significance level. An alert therefore means a real deviation rather than a quiet Tuesday, and it lands in Slack when it fires. Around it sit diagnosis, trends, segments, personas, page level views, goals, funnels, experiments and release tracking, plus a client report and a monthly report.
GRECA members: four months on the platform, free.
We are putting the CRO monitoring platform described above in the hands of GRECA members for four months at no cost. Not a demo account on sample data. Your GA4 connected, your own baseline captured, your twelve metrics tracked against it and the alerts running, on your real store and your real traffic.
Four months is deliberate. One month tells you the tool works. Four months covers a seasonal swing, at least one campaign cycle and enough sample for a change to prove itself or fail honestly.
There is no card held, no retainer attached and nothing that renews on its own. If it earns its place you keep it. If it does not, you still walk away with four months of clean funnel data.
Mention GRECA in the form and we will verify membership against the association’s published member list.
The Conversion Lab
A practical CRO guide to reading the whole funnel, from the first click to the checkout. What share of A/B tests really win. How much traffic you need before a test can tell you anything. Why last click attribution and platform reporting disagree, and which one is closer to the truth. Which behavioural signals are worth acting on today. Built from client work and the primary literature, cited so you can check it.
Frequently asked
How much traffic do we need before CRO is worth doing?
Diagnostic work pays off at almost any volume, because finding the leak only needs enough sessions to see the pattern. Split testing is different: below a few hundred conversions a month, most tests cannot reach significance in a reasonable window, so we prioritise evidence-led fixes over testing and say so upfront.
How long does a test need to run?
Long enough to hit the sample size the expected effect requires, and never less than two full business cycles so weekday and weekend behaviour are both represented. We calculate that number before the test goes live rather than watching the graph and stopping on a good day.
Why does your revenue number disagree with our ad platform?
Platforms count conversions they can claim, on their own attribution window, and they each claim generously. Analytics counts sessions and orders. Both are measuring something real. We reconcile them once, document which source answers which question, and then report against a single agreed number.
Do you need development resource from our side?
For tracking and for larger template changes, yes, and we scope that in the plan. Many of the highest value fixes are copy, layout, and merchandising changes that we can implement or hand over as a ready specification.
We are a GRECA member. How does the four month trial work?
You get the CRO monitoring platform on one store for four months at no cost: your own workspace, GA4 connected, your baseline captured and the twelve metrics tracked with alerts running. We need GA4 access and roughly an hour to set the baseline up with you. No card is held, no retainer is attached and nothing renews automatically. Mention GRECA in the form and we will verify membership against the association’s published member list.
Where do you usually find the biggest leak?
Most often between the product page and the cart, and then again inside checkout, where cost surprises and form friction do the damage. But the point of instrumenting first is that we do not assume. The funnel tells us.
Three guides. No charge, no filler.
Each one is the working method behind a service, written the way we brief a client rather than the way an agency writes a brochure. Sources are cited so you can check the claims.
The Search Evolution
Why a top ten ranking no longer guarantees a citation, what the research says earns one, and the audit and reporting steps to act on it.
Download now
The Conversion Lab
What share of A/B tests really win, how much traffic a test needs, and why last click attribution and platform reporting disagree.
Download now
Predictive UX
What saliency models genuinely know and where they fail, the three signals worth reading, and a four gate pre-launch playbook.
Download nowEcommerce work from the portfolio
Funnel work is only as good as the traffic and the creative feeding it. These are engagements where both sat with us.
See all casesShow us the funnel
Send the store URL and your GA4 access. We will come back with where we think the revenue is leaking and what it is worth fixing first.
