Analytics & attribution
Your reports should end
where the sale does.
Alchemy IBS Private Limited sets up the tracking and reporting that connect marketing to sales: GA4, Google Tag Manager, Meta and Google conversion APIs, call tracking and CRM source fields, reconciled in Looker Studio. We fix measurement before scaling spend, because a campaign optimised on the wrong conversion only gets better at producing the wrong result.
Three systems, three meanings of "conversion".
Meta counts a conversion inside its attribution window. GA4 counts an event on your site. The CRM counts a sale, if someone recorded it. Side by side, they never agree, and most monthly reports quietly pick whichever number looks best.
We start by agreeing what each number means. Then we decide which one the business should steer by, and send that signal back to the ad platforms.
A home-service business had three records for one job.
Campaign conversions, phone calls and booked jobs lived in different places, so nobody could say which channel produced work. A better dashboard wouldn't have helped. The fix was giving each channel its own tracking number, carrying the source into the job record, and reporting on jobs rather than form fills.
Where we look first depends on what you're seeing.
| What you notice | What we check |
|---|---|
| The platform reports more leads than sales sees | Duplicate tags, form spam, conversion definitions |
| Cost per lead falls while sales stay flat | Which event the campaigns are optimising towards |
| Most buyers call instead of filling the form | Call tracking, and how calls reach the CRM |
| Direct and organic seem to take all the credit | UTM discipline, consent mode, cross-device gaps |
| Enquiries rise but appointments don't | Lead-quality fields and what happens after the enquiry |
The deliverable is a number you can steer by.
An agreed event plan. A GA4 and Tag Manager build or repair. Meta and Google conversion APIs sending qualified-lead or sale signals. Call tracking, CRM source fields, and a Looker Studio report reconciled against the CRM. Alongside it, a written account of what the data cannot see, so nobody mistakes a gap for a finding.
We report attributed revenue as attributed.
Consent choices, multiple devices, phone calls and word of mouth all sit outside what tracking observes. So attribution shows a journey, and we never present it as proof that a campaign caused the sale. When a decision needs stronger evidence, we design a test for it: a holdout, a regional split, a clean before-and-after. We won't promise one true number.
Fix the measurement first. Then spend.
Tracking tells performance marketing what to optimise towards, and the CRM supplies the outcomes that make the numbers mean something. If you're unsure where your reporting breaks, run the growth diagnostic first.