Conversion & Analytics
Know which pages produce revenue.
Event instrumentation, funnel analysis and structured experiments, so budget follows evidence instead of instinct.
Analytics installed is not analytics implemented
Most sites have a tag firing and almost no usable data. Pageviews are recorded, conversions are not defined, forms submit without an event, the checkout has no funnel, and traffic from AI assistants is filed under direct. The dashboard is full and it cannot answer which page produced revenue last month.
That gap is expensive in a specific way: it makes every marketing decision an argument about opinion. Without instrumentation there is no evidence, and without evidence budget goes to whoever is most confident in the meeting.
We implement measurement as engineering — a measurement plan, named events, verified firing, and consent handling that does not quietly break the data or the law.
What we do
- Analytics implementation
- GA4 configured properly, server-side where accuracy or ad-blocking justifies it, with data quality checked rather than assumed.
- Event & conversion tracking
- A measurement plan first: the actions that matter, named consistently, then implemented and verified firing on real user journeys.
- Funnel & drop-off analysis
- Where users leave, on which device, from which source — quantified per step instead of inferred from a bounce rate.
- Heatmaps & session review
- Microsoft Clarity and session replay for the qualitative half: what people tried to do before they gave up.
- A/B & multivariate testing
- Experiments with a defined hypothesis, a pre-agreed sample size and an honest read, including the ones that lose.
- Landing page optimization
- Structure, message hierarchy, proof placement and form design, tested against conversion rather than against taste.
- KPI dashboards
- One place where marketing, sales and leadership see the same numbers with the same definitions, so meetings argue about the decision instead of the figure.
- Consent-compliant measurement
- Consent Mode v2 and a banner that does not fire analytics before permission — the first thing a European prospect checks, and frequently the first thing they find broken.
How a measurement engagement runs
- 01
Write the measurement plan
Every event, its definition, its trigger and the question it answers. Agreed before anything is implemented.
- 02
Implement & verify
Tracking built and then tested against real journeys on real devices, because unverified events are worse than no events.
- 03
Analyse
Funnels, segments and drop-off. The output is a ranked list of where revenue leaks, not a dashboard tour.
- 04
Test & iterate
Experiments against the biggest leaks, on a monthly cycle, with results reported whichever way they land.
Our method, published
The twelve events every B2B site should track
This is the starting measurement plan we bring to a B2B engagement. It is deliberately small: twelve events implemented correctly beat sixty implemented approximately.
- Qualified page view (service & pricing pages)
- Scroll depth at 75% on long-form pages
- Primary CTA click, by page and position
- Form start & form submit, tracked separately
- Form field abandonment
- Booking flow start, step complete & confirm
- Document or resource download
- Outbound contact — email, phone, WhatsApp
- FAQ expansion, by question
- Internal search query & zero-result search
- Return visit before conversion
- Assistant referral source, segmented from direct
StackGA4·Microsoft Clarity·PostHog·Consent Mode v2·Metabase
Analytics, answered directly
Having GA4 installed and having it configured are different states. The common pattern is pageviews recorded, conversions undefined, forms untracked and funnels absent. The question is not whether a tag fires — it is whether you can name the page that produced revenue last month.
Where this leads next
Ecommerce Development
Shopify, WooCommerce and custom commerce, with the pricing rules, subscriptions and integrations your operation actually runs on.
Growth & Demand Generation
Paid acquisition, content and lifecycle automation run as one system, against one pipeline number.
Data & Business Intelligence
Once measurement works, the next question is usually a warehouse — which is where our data practice starts.
Every capability in this practice: Web & Digital Growth
Get your measurement in order
Describe what you have and what it needs to do. We will tell you what it takes.