The GTM Analytics Stack: What to Measure and How

Why Most GTM Teams Fly Blind

Most companies track too many metrics and understand too few. The typical GTM team has Google Analytics, a CRM, a marketing automation tool, a product analytics platform, and a spreadsheet that someone updates manually every Friday. None of these systems talk to each other properly.

The result: marketing says they generated 500 leads. Sales says only 50 were any good. The CEO wants to know which channel drives revenue, and nobody can answer with confidence. This is not a tools problem. It is a design problem.

The Minimum Viable GTM Analytics Stack

You need four components. Not four tools, four components. Some tools cover multiple components:

• Acquisition tracking. How do people find you? GA4 with proper UTM tracking covers website traffic. LinkedIn Campaign Manager and Google Ads cover paid. Your CRM tracks outbound.
• Pipeline tracking. What happens after someone becomes a lead? This lives in your CRM (HubSpot, Salesforce, Zoho). Define stages clearly: MQL, SQL, opportunity, proposal, closed-won, closed-lost.
• Product analytics. What do users actually do in your product? Mixpanel, Amplitude, or PostHog. Track activation events, feature usage, and retention.
• Revenue attribution. Which activities and channels produce revenue, not just leads? This requires connecting acquisition data to pipeline data to closed revenue. Most companies skip this and make budget decisions based on lead volume instead of revenue impact.

Attribution Models That Work in Practice

There are many attribution models. Here is what actually works for most B2B companies:

First-touch attribution gives full credit to the first interaction. Useful for understanding which channels bring people in. Simple to implement. Undervalues the middle and end of the funnel.

Last-touch attribution gives credit to the last interaction before conversion. Useful for understanding what closes deals. Simple to implement. Undervalues awareness and nurture.

Multi-touch attribution distributes credit across all touchpoints. More accurate but harder to implement and explain. Use a linear model (equal credit) or a time-decay model (more credit to recent touches) as a starting point.

Start with first-touch and last-touch running in parallel. This gives you a range rather than a single number. If LinkedIn shows up as a top first-touch channel but does not appear in last-touch, it is good for awareness but not for closing. That is a useful insight.

Connecting the Dots

The technical challenge is connecting data across tools. Here is a practical approach:

1. Standardize UTM parameters. Every link you publish gets a UTM source, medium, and campaign. No exceptions. Document your UTM taxonomy and enforce it.
2. Push web data to CRM. When a lead fills out a form, the UTM parameters should flow into the CRM record. Most marketing automation tools do this natively.
3. Tag CRM records with product usage. If you have a product analytics tool, push key events (activated, upgraded, churned) to the CRM. This lets you see which acquisition channels produce active users, not just signups.
4. Build a weekly dashboard. One page showing: leads by channel, pipeline by stage, revenue by channel, and CAC by channel. Looker Studio connected to your CRM works fine for this.

Domande Frequenti