GTM tracking: the exact setup every startup should have live before their first user arrives

There is a window that every startup only gets once.

The first hundred users. The first real behaviour data. The first signal about what is working and what is not.

Most startups throw this window away. Not deliberately — they just do not have the tracking in place to capture it. They launch, users arrive, and by the time someone realises the analytics are not configured properly, the most valuable data set they will ever have is gone.

This is the exact GTM tracking setup that should be live before a single user lands on your product.

LAYER ONE: ANALYTICS FOUNDATION

Google Analytics 4 is the baseline. But "installed GA4" is not the same as "have working analytics."

The common failure mode: the GA4 snippet is on the page, pageviews are recording, and the founder thinks they have analytics. They do not. They have a page view counter.

What actually working GA4 looks like:

Conversion events defined and firing. Every meaningful action a user can take — signup, first login, completed onboarding, first payment, upgrade, share — should be a named conversion event in GA4. If it is not a named event, it is invisible.

User properties set. At minimum: acquisition source, signup date, plan tier. These let you segment every analysis by cohort rather than looking at aggregate numbers that hide everything important.

Enhanced measurement configured and verified. Not just enabled — verified. Use GA4 DebugView to confirm every event fires correctly before launch. Events that look configured but do not actually fire are the most common analytics problem we find in audits.

LAYER TWO: ATTRIBUTION

You need to know where your users come from. Not roughly. Exactly.

UTM parameter structure: every link that sends a user to your product should have a UTM. Consistent naming across every channel. Source, medium, campaign — all defined in a document before you send a single link. UTMs added ad hoc by different team members create attribution chaos within weeks.

Attribution model: choose before you spend. Last-touch attribution (the default in most platforms) overvalues the final touchpoint and undervalues the channels that created awareness. Decide whether last-touch, first-touch, or linear attribution best fits your business model — then configure GA4 and your CRM to report against that model consistently.

Cross-channel identity: if a user sees an organic post, clicks a paid ad three days later, and converts via an email — who gets credit? Define your answer before your first campaign. Without a clear policy, every channel claims full credit and your total attributed conversions exceed your actual conversions by 300%.

LAYER THREE: ONBOARDING FUNNEL TRACKING

Your onboarding flow is your most important funnel. Every drop-off point is either a product problem or a messaging problem. You cannot tell which without the data.

Map every step in your onboarding as a named event. Step one complete. Step two complete. Step three complete. Profile finished. First core action taken.

Then instrument the drop-offs: time between steps, steps where users abandon, steps where users return after abandoning. This is the data that drives your first iteration priorities.

Add session recording to your onboarding flow specifically. Not everywhere — that is noisy. Just onboarding. Tools like Hotjar or Microsoft Clarity on the onboarding pages show you exactly where users get confused, what they click that does not work, what they read twice before giving up.

LAYER FOUR: CRM FROM DAY ONE

Every user who signs up should be in your CRM from day one — not exported in a batch later.

The minimum viable CRM setup for launch:

Acquisition source on every contact record. Where did this person come from? This should be set automatically from UTM parameters, not filled in manually.

Lifecycle stage tracking. Lead, signup, activated, paying, churned. Every transition should be automatic.

Triggered email sequences. At minimum: a welcome email on signup, an activation nudge if they have not completed onboarding within 48 hours, and a check-in at day seven if they have not taken the core action.

This is not complex to set up. HubSpot's free tier handles all of it for early-stage startups. The setup takes a week. The data it produces from your first hundred users is worth more than any amount of post-launch survey work.

Most of the startups we audit have some version of this in place. Almost none have all of it, and the gaps are never evenly distributed — they are always in the places that matter most.

If you want your GTM tracking set up correctly before you launch: foundersbar.com/gtm-for-startups