Analytics
GA4 that actually tracks what matters to your business.
Universal Analytics was sunset in July 2023. Most Indian businesses migrated to GA4 by copying their old setup across, which means they now have a GA4 property that is misconfigured, tracking duplicate events, and reporting numbers that do not match their ad platforms or CRM. GA4 is not UA with a new interface. It is a fundamentally different data model that requires a proper implementation, not a migration of old habits.
Why most GA4 setups are reporting bad data.
GA4 migration was rushed for most Indian businesses because Universal Analytics stopped processing data on a deadline. Here is what that rush left behind.
GA4 is installed but 40% of events are duplicated.
The GA4 tag is firing from both GTM and the direct site code. Every pageview, every purchase, every form submission is counted twice. Reports show double the actual traffic and revenue. Nobody noticed because the old UA numbers stopped being available for comparison.
Conversions are set to the wrong events.
The purchase conversion is tracking an add-to-cart event. The lead conversion is tracking a page view of the thank-you page rather than the form submission. The CTA click is marked as a conversion when a button click without a successful form submission means nothing. Bad conversion definitions corrupt every campaign optimisation that references them.
The data layer was never built, so custom events are missing.
Scroll depth, video plays, PDF downloads, quiz completions, and checkout funnel steps are all invisible to GA4. Only the default automatic events are firing. The business is making content and UX decisions without knowing whether any of it is being engaged with.
Cross-domain tracking is broken.
The business has a primary website and a separate checkout domain or booking system. Sessions are breaking at the domain boundary. Every user who completes a purchase appears as a new session from a direct source, making acquisition attribution completely unreliable. The checkout is the most important step in the funnel and it is invisible.
UA goals are not in GA4 as conversions.
The old Universal Analytics had 12 goals configured over 4 years by different people. Zero of them were recreated as GA4 conversion events during the migration. The business has no conversion data in GA4 and cannot measure campaign performance, funnel drop-off, or channel contribution.
GA4 data does not match the CRM or ad platforms.
GA4 shows 450 leads this month. The CRM shows 310. Google Ads claims 380. Nobody knows which number is correct. Without data validation across systems, every business decision is made on contested numbers and every meeting starts with a fight about whose dashboard is right.
How we implement GA4 correctly.
GA4 implementation is a data engineering project. We design the event taxonomy before writing a single tag, and we validate every number before handover.
Audit & architecture design
- Existing GA4 property audit, duplicate events, misconfigured conversions, and missing events identified
- GTM container audit, firing triggers, tag sequencing, and variable definitions reviewed
- Event taxonomy design, every event the business needs mapped with parameters before any tag is created
- Conversion definition, which GA4 events constitute a conversion and why, agreed with the business
- Cross-domain requirements, subdomains and external checkout or booking systems mapped
- Data layer requirements, what custom data needs to be available in the data layer for event parameters
Implementation
- GA4 property configuration, data streams, data retention (14 months minimum), and internal traffic filters
- GTM GA4 configuration tag, single canonical configuration tag with correct measurement ID
- Deduplication, all duplicate tags removed, single event source confirmed per event type
- Data layer build, developer brief for data layer pushes required for custom events
- Custom event tagging, all events in the taxonomy tagged via GTM with correct parameters
- Conversion event marking, correct GA4 events marked as conversions with appropriate counting method
- Cross-domain configuration, linker parameter added and cross-domain domains listed in property settings
- Audience creation, remarketing and funnel audiences built for ad platform use
Data validation
- GA4 DebugView testing, every event verified in real time before publishing
- GTM Preview mode, every tag trigger and variable confirmed before production release
- Conversion validation, every conversion event verified against actual user actions, not just tag fires
- Cross-system reconciliation, GA4 session and conversion counts compared to CRM and ad platform data
- Source attribution audit, UTM parameters verified to be passing correctly through the funnel
- Anomaly baseline, a 7-day baseline dataset established so future data anomalies are detectable
Reporting & handover
- GA4 exploration reports, custom funnel, path, and segment reports built for the business use cases
- Standard reports configuration, default channel grouping and attribution model set correctly
- Looker Studio connection, GA4 data connected for dashboard use
- Team training, GA4 interface walkthrough, how to read attribution reports, and how to find specific events
- Documentation, event taxonomy, conversion definitions, and GTM structure documented for future reference
Everything included in a GA4 setup engagement.
Property Setup
- GA4 property configuration
- Data stream setup
- Internal traffic filters
- Data retention settings
- Custom dimensions and metrics
- Referral exclusion list
Event Tracking
- Full event taxonomy design
- GTM tag implementation
- Data layer specification
- Custom event parameters
- Scroll and engagement events
- Ecommerce or lead events
Conversions
- Conversion event definition
- Correct counting method per event
- Funnel step events
- Cross-domain purchase tracking
- Ad platform conversion import
- Audience list creation
Validation
- DebugView testing
- Cross-system reconciliation
- UTM attribution audit
- GA4 exploration reports
- Looker Studio connection
- Team training and documentation
This is right for you if:
- Businesses that migrated from Universal Analytics in 2023 and have never audited the GA4 setup
- E-commerce companies where GA4 revenue does not match Shopify or WooCommerce revenue
- B2B companies with no conversion tracking in GA4, making campaign attribution impossible
- Businesses running Google Ads without GA4 conversion import, leaving Smart Bidding without signal
- Marketing teams who cannot trust their GA4 data and are avoiding it because the numbers are wrong
Not the right fit if:
- Businesses with fewer than 500 monthly sessions, GA4 setup cost is not justified at very low traffic volumes
- Companies already running a validated GA4 implementation who are satisfied with their data quality
Frequently asked questions.
Is GA4 actually better than Universal Analytics?
GA4 is more powerful than UA for specific use cases: cross-device tracking, event-based flexibility, BigQuery export, and predictive audiences. It is also harder to configure correctly because the event model is more flexible, meaning more ways to get it wrong. The fact that most Indian businesses now have bad GA4 data is a configuration problem, not a platform problem. A correct GA4 setup is genuinely more useful than UA was.
How is GA4 data different from Universal Analytics data?
UA was session-based: every interaction was grouped into sessions, and metrics like bounce rate were session-level. GA4 is event-based: every interaction is an event, including page views. This means GA4 metrics are not directly comparable to UA metrics. Bounce rate in GA4 is "engaged session rate" and is calculated differently. Sessions, users, and conversion counts will all differ between the two platforms even if both are correctly configured.
What is a data layer and do we need one?
A data layer is a JavaScript object on your website that stores structured data for use by analytics and tag management tools. Without a data layer, GTM can only access information visible on the page surface, like text and URLs. With a data layer, GTM can access transaction amounts, product IDs, user attributes, form field values, and anything else your development team pushes into it. Any meaningful GA4 implementation for e-commerce or lead generation requires a data layer.
Can we keep using Universal Analytics data for historical comparison?
Universal Analytics stopped processing new data on 1 July 2023. The historical data was accessible in the UA interface until July 2024, after which Google deleted it. If you exported your UA historical data before the deletion deadline, that data exists in your own storage. If not, the historical UA data is gone. We recommend building a 12-month GA4 data baseline before making year-on-year comparisons in GA4.
Ready to know whether your GA4 data is actually accurate?
Book a 30-minute call. We will run a live audit of your GA4 property and show you exactly where the data is breaking before you commit to anything.
Book a call