Searchlab Analytics & Tracking
Most marketing arguments are actually measurement arguments. Two dashboards show different revenue, nobody trusts either, and decisions get made on instinct. We build the tracking layer properly: a documented event schema, GA4 configured for your business model, server-side tagging where it earns its keep, and consent handled in a way that is both compliant and useful.
GA4 rewards planning and punishes defaults. We start with a measurement plan that lists the questions the business needs answered, then design events and parameters to answer them. Naming follows one convention across the site, key events are marked deliberately instead of flagging every click, and ecommerce or lead events carry value, currency, and identifiers. A data layer specification goes to your developers so events survive the next site release.
Google Tag Manager is where it gets implemented and, more importantly, where it gets QA tested. Every tag has an owner and a trigger you can explain. We remove duplicate tags left by previous agencies, which is one of the most common reasons conversions get double counted. Preview mode, DebugView, and a documented test script confirm each event fires once, on the right action, with the right values attached.
Server-side tagging is worth the extra infrastructure when ad blockers, Safari cookie limits, or iOS behaviour are eating a visible share of your conversions. Running tags through your own endpoint improves data durability and gives you control over what is shared with each platform. It costs money to host and time to maintain, so we recommend it when the loss is measurable, not by default.
Attribution deserves honesty. With third-party cookies restricted and cross-device journeys normal, no model is exact. We combine platform reported conversions, GA4 attribution, modelled conversions, self-reported source on forms, and periodic holdout or geo tests. Consent Mode is configured so denied consent still contributes modelled data lawfully. The output is a directional picture that is good enough to allocate budget, presented as such rather than as false precision.
Everything in Analytics & Tracking
Measurement plan
A written document linking business questions to specific events, parameters, and reports, agreed before any code is deployed.
GA4 configuration
Property and stream setup, key events, custom dimensions, internal traffic filters, cross-domain rules, data retention, and audiences built for your funnel.
Tag manager implementation
A clean GTM container with named tags, triggers, and variables, duplicate legacy tags removed, and version notes describing every published change.
Data layer specification
Developer facing documentation of every event, its trigger, and its payload, so tracking survives redesigns and new feature releases.
Server-side tagging
A server container deployed on your own subdomain where data loss justifies it, with routing to Google Ads, Meta, and analytics endpoints.
Consent Mode and privacy setup
A consent banner wired to Consent Mode v2 signals, with documented defaults and behaviour for both granted and denied states.
CRM and offline conversion linking
Click identifiers captured at form submission and passed to your CRM, then imported back to ad platforms so optimisation follows revenue, not form fills.
Reporting dashboards
Looker Studio dashboards covering channel performance, cost per qualified lead, and funnel drop-off, with definitions written next to each metric.
The process
Tracking audit
We inventory every tag, event, and conversion action currently firing, compare reported numbers against the CRM and payment records, and list exactly where the gaps are.
Design the schema
Events, parameters, naming conventions, and identifiers are defined in a single document and reviewed with your developers before anything ships to production.
Implement and validate
Tags are built in GTM, tested in preview and DebugView against a written test script, then published in a versioned release with rollback notes.
Report and maintain
Dashboards go live, definitions are documented, and tracking is rechecked after each significant site change so the data does not quietly break.
Questions about Analytics & Tracking
Why does GA4 show different numbers to Google Ads?
They measure differently. Google Ads credits the conversion to the click date and uses its own attribution and modelling, while GA4 credits the session date and applies its own model. Some difference is expected and normal. A gap beyond roughly fifteen to twenty percent usually points to a tagging or configuration fault worth investigating.
Do we actually need server-side tagging?
Only if you can measure the loss. If a large share of your traffic is Safari or iOS, or client-side conversions clearly undercount against your CRM, it pays for itself. For a small site with modest spend, the hosting cost and maintenance overhead usually outweigh the recovered signal.
How does consent affect our data?
When a visitor declines, tags run in a restricted mode that sends anonymous pings instead of identifiers, and Google models the missing conversions from consented traffic. You lose granularity, not the whole picture. Configured properly, Consent Mode v2 keeps reporting usable while respecting the choice the visitor made.
Can you fix tracking without our developers?
Much of it, yes, through Tag Manager using existing page elements. Reliable ecommerce, logged-in state, and lead value tracking need a proper data layer, which does need developer time. We write the specification so that time is measured in hours rather than in a long discovery project.
Ready to talk about Analytics & Tracking?
Tell us where you are stuck. We reply within the hour on WhatsApp, usually sooner.
Or email Searchlabtools@gmail.com