Almost every account we audit is optimising against a number that is wrong. Not slightly wrong — structurally wrong, because the conversion counts page views, or double-counts, or misses a fifth of purchases to iOS. Every decision made on top of that is a guess wearing a chart.
Conversion tracking and analytics covers GA4 configuration, server-side event tracking, ad platform conversion setup, CRM attribution and A/B testing. Its purpose is that the numbers in your reports match reality closely enough to make budget decisions on.
Across client accounts, deduplicated and reconciled against CRM records rather than counted twice.
The share of conversions browser-only tracking commonly misses to iOS restrictions and blockers.
Measured after the tracking was fixed, which is usually the reason it improved.
Cost per qualified lead, or ROAS reconciled to revenue. Everything else is diagnostic.
Tracking audit and gap list
Measurement plan
Tracking implemented
Verified event map
Reconciliation report
Test log and results
Meta says ninety leads. The CRM shows sixty-one. GA4 shows a hundred and four. Nobody has reconciled them, so the monthly meeting is spent arguing about whose number is right rather than about what to do next.
The metric that matters differs by model. Measuring the wrong one confidently is worse than measuring nothing.
Optimising for form fills, so the algorithm finds people who fill forms and never answer the phone.
Qualification from the CRM sent back as the conversion event.
Platform-reported ROAS at double what the bank statement supports.
Server-side purchase events, deduplication, and a blended view reconciled to actual revenue.
Trial signups counted as success while activation and retention go unmeasured.
Product events joined to acquisition data on one user id, so the funnel is continuous.
A six-month cycle measured in a thirty-day attribution window, so half the pipeline looks like it came from nowhere.
Offline conversion imports and a longer window, with lead source persisted through the whole journey.
Calls and walk-ins from advertising that leave no digital trace at all.
Call tracking numbers per location and per campaign, with duration thresholds so wrong numbers do not count.
Supply and demand acquisition measured together, so neither is understood.
Separate funnels, separate events, separate reporting for each side.
It is unglamorous and it is consistently the highest-return engagement we run on a new account.
Better conversion signal means the same budget finds better customers. No creative change produces that effect as reliably.
Server-side events hand back the fifth of conversions that browser tracking loses. Those sales already happened; you were simply not being credited for them.
One reconciled number that finance, sales and marketing all accept turns the monthly meeting from a dispute about data into a decision about budget.
You cannot run a meaningful A/B test on a metric you do not trust. Measurement is the prerequisite for every optimisation programme that follows.
Every event the business cares about, defined — name, trigger, parameters, and which platform needs it. Written before any tag is created.
Events, conversions, audiences, custom dimensions and cross-domain tracking, configured deliberately rather than left on defaults.
Meta Conversions API and Google Enhanced Conversions with event deduplication and match quality monitored.
Source, campaign and creative persisted onto the CRM record, and outcomes exported back to the ad platforms.
Tracked numbers with duration thresholds, and store visits or walk-ins captured where they matter.
Consent mode configured so tracking is lawful and modelling still works where consent is declined.
Spend, leads, qualification and revenue in one view — see [Data & BI Dashboards](/services/data-analytics-dashboards).
Structured tests with a hypothesis, a sample size calculation and a decision rule set before launch.
Almost every account we audit is optimising against a number that is wrong. Not slightly wrong — structurally wrong, because the conversion counts page views, or double-counts, or misses a fifth of purchases to iOS. Every decision made on top of that is a guess wearing a chart.
GA4 configuration, server-side event tracking, ad platform conversion setup, CRM attribution and structured A/B testing. Its purpose is narrow and important: that the numbers in your reports match reality closely enough to make budget decisions on.
They attribute differently — Meta uses view-through and click windows across devices, GA4 uses last non-direct click by default. They will not match, and chasing that is wasted effort. What matters is that each is internally consistent and that one of them reconciles to your CRM.
The single highest-return change on most Indian ad accounts. Conversions sent from your server rather than the browser, deduplicated against the pixel by a shared event id, with as many hashed customer parameters as you lawfully hold. The mechanics are in server-side tracking setup.
Roughly two hundred conversions per variant within a month. Below that, most tests will not reach significance in any sensible timeframe, and you are better served by fixing the obviously broken things and watching ten session recordings — which is usually more informative than a test that never concludes.
In one place everyone reads, with spend, leads, qualification and revenue in the same table. That is a dashboard problem once the measurement is trustworthy — and an unsolvable one before.
Every stage ends in something you can hold — a document, a build, a live account. If a stage cannot name its output, it is a meeting, not a stage.
Every existing tag, event and conversion catalogued, with duplicates and misfires identified.
The measurement plan: what to track, where it goes, and what each event means in business terms.
Client-side and server-side tracking built, with consent handling.
Every event triggered manually and confirmed in every destination. Nothing is assumed to work.
Platform numbers compared with the CRM and with finance, monthly, until the differences are explainable.
A structured CRO programme against a metric everyone now trusts.
Everything here is part of the engagement at no extra cost. We do not itemise them on an invoice and we do not withhold them if you leave.
Every tag, event and conversion catalogued with the duplicates and misfires identified. Free, and it frequently changes what people think their campaigns are doing.
Every event defined — name, trigger, what it means to the business — so your team can read the analytics without a translator.
Triggered manually and confirmed in every destination. Nothing is assumed to work because it was installed.
Platform numbers compared with what actually happened, until the differences are explainable rather than argued about.
So tracking is lawful and modelling still works where consent is declined.
Longer answers to the questions people ask before they hire anyone for conversion & analytics.
Built by us, free, no signup, nothing uploaded to a server. Take them whether or not you ever become a client.
An audit and measurement plan is a small fixed project; the full build with server-side tracking, CRM attribution and dashboards is larger; an ongoing CRO programme is monthly and only makes sense above a certain traffic level. The audit usually pays for itself in wasted spend identified before you commit to anything else.
Tell us what you have now and what you are trying to reach. We will audit it and tell you what we would do, what it would cost and whether you need us at all. The audit is free and yours to keep.