Product Intelligence

What is the job to be done?

"Help me understand what users do, why they do it, and what to build next."

• Understand how users navigate your product and where they get stuck
• Identify which features drive retention and which are ignored
• Get qualitative context for quantitative patterns (the "why" behind the numbers)
• Validate product hypotheses with experiments before committing engineering resources
• Collect direct user feedback at key moments in the product experience
• Track business outcomes (revenue, expansion) tied to product usage
• Act on insights by guiding users through onboarding and feature adoption directly inside the product

This is our bread and butter. Most accounts start here. The risk is they stay here as a single product analytics customer and never expand. The opportunity is that Product Intelligence naturally creates demand for the other use cases once teams start acting on what they learn.

What PostHog products are relevant?

• Product Analytics (core) — Funnels, retention, cohorts, lifecycle analysis, trends, user paths. The quantitative foundation for understanding what users do.
• Session Replay — The qualitative layer. Watch what users actually do when the numbers say they're dropping off. Bridges the gap between "40% drop at step 3" and "oh, the button doesn't render on mobile Safari."
• Surveys — Direct feedback loop at key moments. NPS after onboarding, CSAT after support, "why did you cancel?" on churn. Ties qualitative signal to quantitative behavior data.
• Revenue Analytics — Business outcome tracking. Connect product usage to MRR, expansion revenue, LTV, and churn. Lets PMs prove that the feature they shipped actually moved the revenue needle. (Dashboard)
• Experiments — Validate hypotheses with statistical rigor before committing to a full build. A/B test changes against real metrics, not gut feel. Requires Feature Flags for implementation.
• Workflows — Onboarding sequences, activation nudges, lifecycle engagement. The action layer: once you've identified a drop-off in analytics, Workflows lets you do something about it automatically. (Email drip campaigns)
• Product Tours alpha — In-app guided onboarding, feature adoption prompts. The in-product complement to Workflows' out-of-product engagement. Guide users through the right path at exactly the right moment. (Creating tours)
• AI Evals — For products with AI features: proactively surface where users are struggling based on AI output quality. This is product intelligence driven by AI observability. A bridge product to the AI/LLM Observability use case.
• PostHog AI — Natural language querying and insight discovery. Lowers the bar for non-technical product stakeholders to self-serve. A PM asks "why did retention drop last week?" instead of building a custom query. (Example prompts)

Adoption path and expansion path

Entry point

Usually Product Analytics. Customer starts tracking events, builds dashboards, creates their first funnel. Then they hit the ceiling of quantitative data alone: "I can see that users drop off, but not why."

Primary expansion path

Product Analytics → + Session Replay → + Surveys → + Experiments → + Revenue Analytics → + Workflows / Product Tours

The logic of each step:

• Product Analytics → Session Replay: They know what is happening (40% drop at step 3). They need to see why. Session Replay gives them the qualitative context that numbers can't.
• Session Replay → Surveys: They're watching replays and forming hypotheses about why users struggle. Surveys let them ask users directly at the moment of friction, then tie responses back to behavior data.
• Surveys → Experiments: They've identified the problem through analytics, replay, and feedback. Now they want to test a fix. Experiments require Feature Flags, which gets engineering involved (multithreading moment).
• Experiments → Revenue Analytics: They've validated changes with experiments and want to prove business impact. Revenue Analytics connects product usage to MRR, expansion, and churn.
• Revenue Analytics → Workflows / Product Tours: They've identified what drives value. Now they want to guide users toward those high-value behaviors automatically, through in-app tours and behavior-triggered engagement sequences.

Alternate expansion paths

B2B accounts with Group Analytics: B2B SaaS companies almost always need company-level analytics alongside user-level. If they're B2B and not using Group Analytics, that's a significant upsell opportunity. Group Analytics lets them answer "which companies are most engaged" not just "which users."

Starting from Session Replay: Some accounts come in through Session Replay first (debugging, QA, customer support use cases). They realize they need Product Analytics to quantify what they're seeing qualitatively. The expansion path reverses: Replay → Analytics → Surveys → Experiments.

Product teams that ship AI features: If the product has AI components, AI Evals can proactively surface where users are struggling based on output quality. This bridges Product Intelligence into AI/LLM Observability.

Business impact of solving the problem

This is the use case with the largest existing install base. Most PostHog accounts start with Product Analytics. The expansion opportunity isn't convincing them to adopt PostHog. It's convincing them to go beyond a single product and use the full Product Intelligence stack.

The Workflows and Product Tours close-the-loop story is powerful. You identify a drop-off point (analytics), you understand why users leave (session replay, surveys), and now you can actually fix it by guiding users through the right path (product tours) or re-engaging them when they disengage (workflows). That's a complete insight-to-action cycle that no competitor offers in one platform.

Product Intelligence creates demand for other use cases. Once the product team is deep in PostHog, they pull in the growth team (Growth & Marketing use case) for acquisition and activation. Once they're running experiments, engineering gets involved in rollouts (Release Engineering). This is the gateway use case.

Personas to target

Signals in Vitally & PostHog

Vitally indicators this use case is relevant

PostHog usage signals

Command of the Message

Discovery questions

• How does your product team decide what to build next? What data informs that decision?
• When you see a drop-off in a funnel, how do you figure out why users are leaving?
• How do you measure whether a new feature is successful after launch?
• Do you collect direct user feedback inside the product? How is that connected to your analytics?
• When you have a hypothesis about user behavior, how do you validate it? Do you run experiments?
• How do you prove to leadership that a product investment drove business results (revenue, retention)?
• How many tools does your product team use to understand users? (Analytics, replay, surveys, experiments — how many vendors?)
• Can your PMs answer their own questions, or do they depend on data/engineering for every query?

Negative consequences (of not solving this)

• Product decisions are based on gut feel or incomplete data because the team can't connect behavior to outcomes
• PMs can see that users drop off but not why, leading to guesswork about what to fix
• Experiments are rare or nonexistent because the tooling is disconnected from analytics, so every test requires a separate setup
• User feedback (surveys) lives in a separate tool, disconnected from behavior data, so you can't answer "what happened right before this user gave us a 3/10?"
• Product team can't prove business impact, making it hard to justify investment or prioritize
• Insights are identified but never acted on because there's no automation layer to guide users or re-engage them

Desired state

• One platform for the full cycle: measure behavior → watch sessions → collect feedback → test changes → measure revenue impact → act on insights
• PMs can self-serve answers without waiting for a data team
• Every product change is measured against real retention and revenue metrics
• Onboarding and feature adoption are guided automatically based on user behavior
• Product team and engineering share the same platform, reducing tool fragmentation

Positive outcomes

• Faster product decisions: cycle time from "we see a problem" to "we've validated a fix" drops significantly
• Higher retention from catching and addressing drop-off points systematically
• Better resource allocation: experiments prove what works before engineering commits to a full build
• Product team can demonstrate revenue impact to leadership, strengthening their influence
• Tool consolidation: replace separate analytics + replay + survey + experimentation vendors with one platform

Success metrics

Customer-facing:

• Feature adoption rates improve for targeted features
• Retention curves flatten or improve for key cohorts
• Experiment velocity increases (more hypotheses tested per quarter)
• Time from insight to action decreases

TAM-facing:

• Customer expands from Product Analytics-only to multi-product (Session Replay, Surveys, Experiments)
• Multiple product team members (PMs, designers) are active, not just engineers
• Experiment usage grows (indicates the product team is using PostHog for decisions, not just reporting)
• Workflow or Product Tour usage starts (they're closing the insight-to-action loop)

Competitive positioning

Our positioning

• Quantitative + qualitative in one platform. Product Analytics and Session Replay together. No switching between Amplitude and Hotjar. Filter replays by funnel drop-off, cohort, or event.
• Insight to action, not just insight. Most analytics tools stop at the dashboard. PostHog lets you act on what you find with Workflows, Product Tours, and Experiments.
• Experiments built into the analytics workflow. See a drop-off in a funnel, right-click to create an experiment, measure the result in the same tool. No separate experimentation platform.
• PostHog AI makes analytics accessible. PMs who aren't comfortable with SQL can ask questions in plain English.

Competitor quick reference

Honest assessment: Our strongest position is the breadth of the platform. No competitor offers analytics + replay + surveys + experiments + workflows + product tours in one tool. We're weaker against Amplitude in very large enterprises where their ML features and enterprise sales motion are more mature. We're weaker against Hotjar/Pendo for non-technical product teams who want a simpler, more opinionated UX. Our sweet spot is technical product teams at companies with engineers who value depth, flexibility, and not paying for 5 separate tools.

Pain points & known limitations

Getting a customer started

What does an evaluation look like?

• Scope: Instrument their core product flow: signup → key activation event → retention-defining action → conversion/upgrade. Build the primary funnel and retention analysis. Enable Session Replay on key flows.
• Timeline: 1 to 2 weeks to see value from analytics and replay. Experiments need enough traffic for statistical significance, so timeline varies.
• Success criteria: Can you answer: "Where do users drop off in our core flow, and why?" Can you see the full retention curve by cohort? Can you watch a replay of a user who dropped off?
• PostHog investment: Product Analytics free tier covers 1M events. Session Replay free tier covers 5K recordings. Surveys, Experiments have generous free tiers.

Onboarding checklist

• Install PostHog SDK with autocapture enabled (getting started)
• Define and instrument core conversion events (signup, activation, key feature usage, upgrade)
• Build the primary conversion funnel
• Set up retention analysis for key activation events
• Enable Session Replay on production (configure sampling if needed)
• Watch 5+ replays filtered by funnel drop-off to get qualitative context
• Build a "Product Health" dashboard for the product team (funnels, retention, trends)
• Introduce PostHog AI to PMs for self-serve querying
• Set up one Survey at a key friction point (post-signup NPS, feature feedback)
• Plan first Experiment targeting a known drop-off point

Cross-sell pathways from this use case

Internal resources

• Product Analytics docs: Funnels · Retention · Lifecycle · Cohorts · User paths · SQL
• Session Replay docs: Session Replay
• Surveys docs: Creating surveys
• Experiments docs: Experiments · Exposures
• Revenue Analytics docs: Getting started · Dashboard
• Workflows docs: Getting started · Email drip campaigns
• Product Tours docs: Getting started · Creating tours
• Feature Flags docs: Getting started
• PostHog AI docs: Enable PostHog AI · Example prompts
• Group Analytics docs: Group Analytics

Appendix: Company archetype considerations