GuideTOFU
The marketing tracking guide
The short answer
If you can't measure it, it doesn't exist — and if you measure it wrongly, it's worse than not measuring at all. Ad platforms optimise toward whatever conversion data you send them. Send duplicated, mis-valued or incomplete data, and they'll confidently optimise toward the wrong outcome.
This guide covers the tracking stack a modern Indian brand needs, in the order to build it.
The layers of a tracking stack
- Analytics — GA4 for site and app behaviour
- Ad platform tracking — Meta pixel and Conversions API, Google Ads conversions with enhanced conversions
- Server-side layer — server-side Google Tag Manager to route events reliably
- CRM and offline data — lead outcomes and offline sales
- Attribution and reporting — dashboards that reconcile everything against your back end
Step 1: Write a measurement plan
- List the business outcomes that matter — purchases, qualified leads, bookings.
- Define the events and parameters for each.
- Decide which events are conversions for which platforms.
- Set naming conventions and a UTM policy.
- Document it where everyone can find it.
Step 2: Set up GA4 properly
- Recommended eCommerce events with item data
- Key events that match business outcomes
- Cross-domain tracking for payment gateways and booking tools
- Referral exclusions for payment gateways
- Internal traffic filters
- Maximum data retention and BigQuery export
Step 3: Meta pixel and Conversions API
Install the pixel with standard events, then send the same events from your server through the Conversions API with a shared event ID so Meta can deduplicate. Include hashed customer details where consent allows to improve match quality. Check Event Match Quality in Events Manager.
Step 4: Google Ads conversions
Use Google Ads conversion tracking or import from GA4, enable enhanced conversions, and import offline conversions from your CRM for lead-generation businesses.
Step 5: Server-side tagging
Route events through a server container on your own subdomain. It improves data completeness, gives you control over what's shared, and can speed up your site by removing browser scripts.
Step 6: UTMs and campaign hygiene
- Lowercase, consistent values
- A shared naming convention across teams and agencies
- UTMs on every non-Google-Ads campaign link
- No UTMs on internal links
- Links tested through redirects
Use a UTM builder to keep it consistent.
Step 7: Offline conversions
For lead-generation businesses, the most important data lives in your CRM: which leads were qualified, visited, bought. Send those outcomes back to Meta and Google so they optimise for revenue, not form fills.
Step 8: Reconcile and monitor
Compare platform-reported conversions with your store or CRM weekly. Expect differences — platforms use different attribution — but investigate large or sudden gaps. Set up alerts for sudden drops in event volume.
Common mistakes
- Purchase events firing twice
- Values including tax and shipping inconsistently
- Pixel without Conversions API
- Payment gateways stealing credit as referrals
- UTMs stripped by redirects
- No offline conversions for lead-gen
Consent and privacy
India's Digital Personal Data Protection framework requires clear notice and consent for processing personal data, with rights to access and erasure. For marketing tracking, that means a clear privacy notice, consent collection where required, and tracking configured to respect choices — including Google's Consent Mode for Google tags. Server-side tagging must respect the same consent decisions as browser tags.
A tracking QA checklist
- Place a test order or lead and confirm exactly one conversion appears in each platform
- Confirm values match the back end, net of discounts
- Confirm event IDs match between pixel and server events
- Test through the payment gateway and back
- Test with UTMs and confirm they survive to the thank-you page
- Test on mobile and in in-app browsers
- Confirm internal traffic is excluded
Dashboards that support decisions
- Top line — revenue or pipeline, total spend, MER, new-customer CAC
- Channels — platform-reported and reconciled numbers side by side
- Funnel — conversion rates by stage
- Creative — performance by concept
- Health — tracking and data freshness alerts
When to add incrementality testing
Once spend across channels is large enough that a wrong allocation decision costs real money — and once basic tracking is trustworthy. Start with a geo holdout on the channel you're least sure about.
A sample measurement plan
| Event | Trigger | Key parameters | Key event? | Sent to |
|---|---|---|---|---|
| view_item | Product page view | item_id, price, category | No | GA4, Meta, Google Ads |
| add_to_cart | Add to cart | item_id, value | No | GA4, Meta |
| begin_checkout | Checkout start | value, items | No | GA4, Meta |
| purchase | Order confirmed | transaction_id, value, currency | Yes | GA4, Meta CAPI, Google Ads |
| generate_lead | Valid form submission | form_id, lead_type | Yes | GA4, Meta, Google Ads |
| qualified_lead | CRM status change | lead_id, stage | Yes (offline) | Google Ads, Meta CAPI |
Reconciliation routine
- Weekly — compare tracked purchases or leads against orders or CRM records.
- Investigate gaps that change suddenly — usually a broken tag, checkout change or consent change.
- Check deduplication — Meta event match quality and duplicate counts.
- Audit UTMs — inconsistent naming and untagged links.
- Document every tracking change in a change log.
Tracking for Indian businesses: special cases
- Payment gateway redirects — prevent them from appearing as referrers
- COD orders — track placement and later delivery status
- WhatsApp conversions — log chats and outcomes in the CRM
- Phone calls — call tracking with minimum duration
- Marketplaces — separate reporting; use attribution tools where available
- Offline store sales — loyalty or billing data matched with consent
Key tracking terms, defined
- Data layer — a structured object on your site that passes event data to tags.
- Deduplication — preventing the same conversion being counted twice from pixel and server events.
- Consent mode — adjusting tags based on users' consent choices.
- Offline conversions — sales or qualified leads uploaded from your CRM to ad platforms.
- UTM parameters — tags in URLs that identify campaign sources.
- First-party data — data you collect directly with consent.
A worked example: reconciling three sources of truth
A brand's back end showed 1,000 orders; GA4 showed 820; Meta claimed 900 and Google 700. We confirmed the back end as the source of truth, fixed duplicate GA4 tags and missing purchase events on the payment-gateway return page, added server-side events with deduplication and aligned conversion windows. The gaps narrowed to explainable differences — attribution windows and cross-device journeys — and the team stopped debating whose numbers were right.
The tracking QA checklist
- Test purchases or leads recorded once in every tool
- Values and currencies correct
- Consent choices respected
- UTMs consistent across campaigns
- Offline conversions uploading on schedule
- Alerts for sudden drops in tracked conversions
Privacy and India's data-protection framework
India's Digital Personal Data Protection framework requires notice and consent for processing personal data, with obligations phasing in through 2027. Marketing teams should audit what personal data they collect, why, where it goes and how consent is recorded — and minimise data shared with ad platforms to what's needed.
Further reading on PMG
Frequently asked questions
Why doesn't GA4 match Meta?
Different attribution models, windows and data. Reconcile each against your back end.
Is server-side tracking necessary?
For businesses spending meaningfully on ads, it usually pays for itself through better data.
How often should tracking be audited?
After every site change, and fully at least quarterly.
What should we track first?
The events that represent revenue or qualified pipeline — purchases and qualified leads — with correct values and deduplication.
Do we need a data warehouse?
Not at first. It becomes useful when you combine many sources, need longer history or want custom attribution and LTV analysis.
Who should own tracking?
One named owner — internal or agency — responsible for the measurement plan, changes and data quality.
Which tool should be our source of truth?
Your back-end system — orders or CRM — with analytics and ad platforms used for detail and optimisation.
Is server-side tracking necessary for small businesses?
Not always, but even small businesses benefit from accurate conversion setup; server-side becomes more valuable as spend grows.
How long does a full tracking setup take?
From a few days for simple sites to several weeks for complex, multi-platform setups.
More guides
Services behind this
Terms used on this page
Free growth audit
Find out where your budget is leaking.
Send us your ad account, analytics or site. Within 72 hours you get a written audit: tracking gaps, wasted spend, and the three moves we'd make first. No pitch deck.