Google Ads reports one number. Meta reports another. The CRM shows something smaller, and the finance team trusts only the bank statement. For many marketing teams in Bangalore, the monthly review has turned into an argument about whose number is right.
The teams that have moved past that argument didn’t find a perfect dashboard. They changed what they measure, what they feed back to the ad platforms, and who decides whether a campaign worked. Here is what that shift tends to look like.
- Budget decisions follow the CRM, not the ad dashboard
Ad platforms count what they can see: a form fill, a click, a purchase pixel firing. They can’t see whether the lead answered the phone, visited the site or paid. That is why platform-reported conversions usually exceed the qualified opportunities sales records.
When the two disagree, the CRM wins. Platform numbers stay useful for diagnosis, not as the scoreboard.
- “Qualified” gets defined before the campaign goes live
Marketing and sales rarely mean the same thing by “lead”. Teams making this shift agree in writing on the stages that matter, such as contacted, qualified, site visit done and booked. Every lead in the CRM then carries both a source and a stage. Without this, no attribution model can help.
- Cost per qualified outcome replaces cost per lead
CPL rewards volume. Widen the targeting and CPL falls, while close rates fall with it. The replacement metric depends on what the business sells:
| Sector | Instead of | Track |
| Real estate | Cost per lead | Cost per site visit or booking |
| Healthcare | Cost per enquiry | Cost per attended appointment |
| Education | Cost per enquiry | Cost per enrolment |
| B2B / SaaS | Cost per MQL | Cost per opportunity or closed deal |
| D2C | Platform ROAS | Return after product, shipping and return costs |
CTR and CPL still help diagnose creative and targeting, but they stop being the headline.
- CRM outcomes are sent back to Google and Meta
Measuring revenue is half the job. The other half is showing the bidding algorithms what a good lead looks like, which CRM integration makes possible.
On Google, the recommended route is enhanced conversions for leads, set up through Google Ads Data Manager. It matches CRM outcomes to ad clicks using the click ID and hashed customer details such as email or phone number. Google reports that advertisers who added this first-party data to offline imports saw a median 10% increase in recorded conversions compared with standard imports.
On Meta, the Conversions API for CRM, paired with the Conversion Leads performance goal, lets campaigns optimise toward leads that progressed in the pipeline. It currently works only with Meta’s native Instant Forms.
- The optimisation target is a stage the algorithm can learn from
Optimising for closed sales sounds right, but in long-cycle businesses it often backfires: sales happen too rarely and too late for the algorithm to learn from. The practical choice is the deepest stage that still happens often and soon after the lead comes in, usually “qualified” or “site visit done” rather than “booked”. Revenue stays the reporting metric even when it isn’t the optimisation event.
- Platform attribution settings stop deciding channel budgets
Attribution windows and models are rules set by the platforms, and the platforms change them. Meta, for example, has removed its longer view-through windows from reporting. When rules like these change, reported conversions can shift even if customer behaviour doesn’t.
In Google Ads and GA4, the older rule-based models (first click, linear, time decay and position-based) are gone, leaving data-driven and last click as the main options. Use them for in-platform optimisation; judge channels by the CRM over the full sales cycle.
- Awareness spend is judged on its own terms
Last-click attribution gives little credit to the ads that started the journey. Someone buying an apartment, an MBA seat or enterprise software often sees several ads over weeks before enquiring. Cut awareness because it shows few direct conversions, and lead quality can slip later.
Better tests: check whether leads reached by earlier campaigns qualify at a higher rate, or, where budgets allow, pause awareness in one region and compare results.
- D2C brands reconcile revenue before scaling
Platform ROAS and store revenue rarely match. Two platforms can claim the same order, and returns are not deducted unless the brand sends adjustments back. In India, cancelled or undelivered cash-on-delivery orders widen the gap when purchases are tracked at the time of order.
Before increasing spend, these brands reconcile platform figures with Shopify or backend data and work out the ROAS they need to break even on margin, not just on revenue.
- What happens after the form gets tracked
Leads often move to phone calls or WhatsApp soon after enquiring, and that is where the tracking trail usually ends. If sales updates the CRM inconsistently, every report built on it suffers. The fix is mandatory stage fields, logged call outcomes and a weekly lead-quality review with sales.
- Consent is built into the data flow
Sending hashed emails and phone numbers to ad platforms is likely to count as processing personal data under India’s DPDP Act. The DPDP Rules, notified in November 2025, require clear, standalone consent notices explaining why data is collected, and the notice and consent obligations apply from May 2027.
Update lead forms and consent language while setting up attribution, not afterwards, and have your legal team review the specifics.
Where to start
Start with three moves:
- Compare platform-reported conversions with CRM outcomes for the last 90 days, campaign by campaign.
- Agree on lead stages with sales and make source and stage mandatory. Even a shared Google Sheet works at first; Google’s Data Manager can import conversions from one.
- Connect one platform’s offline conversions and optimise toward a qualified stage.
Whether your accounts are run by an in-house team or a performance marketing agency in Bangalore, ask for reporting on cost per qualified outcome, not just CPL and platform ROAS. If those numbers can’t be produced, start there.
FAQs
What is revenue attribution in performance marketing?
It means judging ad spend by outcomes that earn money, such as qualified leads, bookings, enrolments, closed deals or profitable orders, rather than by clicks, CPL or platform-reported ROAS. The CRM records what each lead became, and that data guides both channel decisions and the platforms’ bidding.
Why do Google Ads and Meta report more conversions than my CRM?
Each platform credits conversions using its own window and model, so both can claim the same lead. They also count duplicates, spam and form fills that never qualify.
How does CRM integration improve ad performance?
It sends lead stages back to Google through enhanced conversions for leads and to Meta through the Conversions API for CRM, so the platforms learn which clicks become qualified leads rather than just form fills.
Which attribution model should we use in Google Ads?
Google recommends data-driven attribution, the default for most conversion actions, with last click as the main alternative. The outcomes you feed back matter more than the model, and cross-channel budget decisions should rest on CRM data.
