How product marketers use PostHog for GTM signals

PostHog is on-site truth for GTM signals, not a vanity dashboard.

Ehrenberg-Bass / LinkedIn B2B Institute: about 95% of B2B buyers are not in-market. Most product events are noise relative to purchase. I use PostHog as an example of product analytics to find a buying window, not to decorate a weekly review. Activation, pricing page, feature use, and invites show what happened. They do not show who to email until you pair them with identity.

PostHog is on-site truth. Without identity and a send, it is a dashboard hobby. Pair behavior with who, then touch the window.

What PostHog signals matter for GTM?

Four event families. Keep the list short.

1. Activation. First value moment. Project created, first tracking installed, first insight saved. Someone crossed from tire-kicker to user. That is a warmer window than a random homepage hit.
2. Pricing and plan. Pricing page views, plan compare clicks, billing portal opens, trial-to-paid attempts. Intent to spend lives here.
3. Feature depth. Repeat use of the feature you sell as the wedge. If your wedge is session replay and they watched three replays in a day, that is a signal. If they bounced after one, that is a different note.
4. Expansion tells. Invite teammate, create second project, hit a limit banner. Those are Grow seeds. Still treat them as Signals first: who did it, how hot, what to say.

Most PostHog events will still be noise relative to that 95% out-of-market reality. Your job is to score the few that look like a window, not to celebrate total event volume.

How do I pair PostHog with identity?

Website de-anonymization tools such as Vector or RB2B answer who was on the site. PostHog answers what they did once they were in product or on key pages.

Workflow shape:

1. De-anon resolves company or person on marketing site.
2. PostHog (or your product analytics of choice) shows activation, pricing, or feature events for that account when they sign up or when you can stitch known users.
3. Enrichment adds role, tech, and path to a real inbox.
4. First touch names the behavior. "Saw you hit pricing twice after inviting two teammates" beats " Congrats on exploring modern product analytics."

Do not invent a Product Marketer Pro analytics product. PostHog, Amplitude, Mixpanel, or your warehouse can fill the behavior seat. The motion is the product: signal to enrichment to send.

How do I score windows without a huge TAM?

Same logic as scoring buying windows without a huge TAM.

Weight recency and intensity. A pricing view today plus an activation yesterday beats a feature click from three weeks ago. Cap the queue. Ten sharp sends beat two hundred soft ones. Salesforce still puts about 27% of reps at quota. Flooding the other 73% with "you used feature X" notes does not raise the number.

If you cannot say why this account is in-market in one line, do not send.

What mistakes kill this motion?

• Dashboard tourism. Looking at funnels weekly without a send SLA.
• Event spam. Tracking everything, scoring nothing.
• Identity skip. Behavior without who becomes a slide, not a meeting.
• Generic copy. A note that could have gone to anyone does not go.
• No kill path. If a segment misfires, you need a stop button before the domain burns.

Monday.com CTR from 2.94% toward 0.84% was channel fatigue. Product-triggered outbound fatigues the same way when every click gets a pitch. Thresholds matter.

How do I run this in one week?

Day 1: pick three PostHog events tied to money (activation, pricing, invite).

Day 2: confirm you can join those events to account or person via de-anon or known-user stitch.

Day 3: write three first-touch templates that name the event and one outcome.

Day 4 to 5: send to a capped list. Count replies and meetings. Forrester-style ~84-day cycles mean you will not close this week. You should still see whether the open earns a reply.

Would you rather a PostHog board that looks busy, or ten named windows you actually touched?

Adapt or fail. Instrument the events that mean money. Attach identity. Send while the window is open. That is Signals.

Start Signals, Convert, Grow

FAQ

Is PostHog required for this playbook?

No. It is an example of product analytics. Use whatever gives you trustworthy activation, pricing, and feature events. The motion is behavior plus identity plus first touch.

What if most of my traffic is anonymous?

Lean harder on Vector, RB2B, or Factors.ai for company identity on the marketing site. Use PostHog heavily once users are authenticated. Anonymous pricing-page spikes still feed account-level outbound when de-anon lands.

How is this different from product-led sales dashboards?

PLS dashboards often stop at "PQL created." This playbook demands a signal-matched send and a meeting metric. If nobody got a note, you did not do GTM. You did reporting.

Where does this sit in Signals, Convert, Grow?

Signals owns event choice, identity join, and scoring. Convert owns the note and the ask. Grow owns expansion events (invites, limits) once the core motion books meetings.