In-depth two-product comparison
Heap vs PostHog
A practical Heap and PostHog comparison for B2B SaaS teams: autocapture, governance, warehouse sync, replay, feature flags, experiments, pricing and rollout.
Last updated: 2026-10-08
Short answer
Choose Heap when a B2B SaaS team wants broad autocapture, visual event definition and a more enterprise-oriented Contentsquare relationship. Heap is free for up to 10,000 monthly sessions; its Growth, Pro and Premier prices are pricing on request. Choose PostHog when engineers want an open-source or self-hosted option and one usage-based suite for analytics, replay, feature flags and experiments. PostHog is free within monthly product allowances, then charges by usage. Both capture behavior automatically, but they put the team in different operating models.
Heap
Autocapture and retroactive analysis for a governed, enterprise-leaning workflow.
PostHog
A usage-based, open-source suite for developer-led analytics and releases.
The choice follows the team’s operating model: Heap gives analysts broad autocapture and retroactive discovery, while PostHog gives developers a connected release and analytics suite. Run the same account-level question through both before committing.
Quick comparison
| Decision area | Heap | PostHog | What to verify |
|---|---|---|---|
| Best fit | Product, UX and analytics teams that need broad autocapture and visual investigation. | Developer-led teams that want analytics beside flags, replay and experiments. | Have the people who maintain events and release controls walk through one real product question. |
| Autocapture | Core workflow captures interaction data so teams can define events after collection. | Autocapture is available alongside explicit events and properties. | Install each SDK on a staging flow and inspect clicks, page views, form actions and identity handling. |
| Pricing basis | Free up to 10,000 monthly sessions; paid session pricing is pricing on request. | Monthly free allowance per product, then pay for units used; no fixed public monthly fee. | Model sessions, events, recordings, profile events, flag requests and warehouse rows separately. |
| Data governance | Visual definitions, account analytics and enterprise permissions are a natural fit for a central analytics owner. | Flexible raw events and HogQL require the team to define its own naming, review and ownership rules. | Ask an analyst to correct one property and publish a shared activation definition without engineering help. |
| Warehouse | Heap Connect can sync data to a warehouse automatically and retroactively; confirm the add-on in the quote. | Data warehouse is part of the suite with a free monthly row allowance and paid usage after it. | Trace one account-level metric from product event to warehouse table and downstream report. |
| Session replay | Session Replay is listed as an add-on in the Contentsquare-backed product. | Built-in recordings share a workspace with events and other products. | Replay the same failed workflow and compare retention, debugging context and access controls. |
| Feature flags | Not a core Heap capability in the published product details; plan a separate flag service. | Built-in flags support targeting and rollout alongside the event stream. | Release a flag to one account cohort and record who owns expiry and rollback. |
| Experiments | Treat experimentation as a separate requirement and confirm any Contentsquare scope. | Experiments are billed with feature flags and can use the same product data. | Write the exposure event, success metric, sample rule and analysis owner before a trial. |
| Deployment | Cloud product with enterprise controls and region-specific storage on Premier. | Cloud regions include US (Virginia) and EU (Frankfurt); an MIT-licensed self-hosted option is available. | Give security the required region, retention, access, backup and incident-response answers. |
A fair model uses the same users, sessions, event mix, identified profiles, recordings, flag requests, retention and warehouse rows. Heap needs a quote; PostHog needs a multi-meter usage model.
Feature comparison
| Feature | Heap | PostHog |
|---|---|---|
| Autocapture depth | Heap’s defining move is to collect interaction data first, then let a team label meaningful events from the history. That reduces the risk of discovering an important path after the relevant release has passed. | PostHog can autocapture, while explicit events and properties give engineers a deliberate contract for business-critical actions. The team must decide which raw interactions deserve stable names. |
| Metric ownership | A product analyst can own visual event definitions and account views, with access controls and review conventions around shared reports. | HogQL and raw event access are powerful, but a developer-led team needs a dictionary, query review and clear owners before dashboards become company metrics. |
| B2B identity | Use account analytics when the product question concerns workspaces as well as individual users. Verify how account attributes, merges and historical identities appear in the contracted plan. | Identify users and groups deliberately, then test person-profile and group analytics charges. A workspace model should be part of the implementation brief, not added after the first bill. |
| Access and residency | Premier includes advanced user permissions and region-specific storage, which suits a procurement-led review of roles and data location. | Cloud customers choose US (Virginia) or EU (Frankfurt); SSO, audit logs, custom roles and project permissions are available in platform packages. Self-hosting changes who operates the controls. |
| Session replay | Heap lists Session Replay as an add-on and connects it to the wider Contentsquare digital-experience context. Include retention, access and add-on cost in the proposal. | PostHog recordings sit beside captured events, console and network context. The shared developer workflow is useful when a replay needs to explain a release regression. |
| Feature flags | Heap is primarily an observation and analysis layer here. A team that needs progressive delivery should budget for a separate flag system and an integration between exposure data and Heap. | Flags are a first-class product with targeting and rollout controls. Pair every flag with an owner, expiry date, exposure event and rollback procedure. |
| Experiments | Heap can analyze behavior around a separately managed experiment, but confirm which experimentation workflow is included in the Contentsquare relationship before assuming a native replacement. | PostHog connects experiments with flags and product analytics. The advantage is a shorter path from assignment to readout; the trade-off is that the team owns experiment hygiene. |
| Warehouse use | Heap Connect is suited to teams that want product data available in an existing warehouse and want the sync to include historical data. | PostHog includes a warehouse product and row allowance. Check whether its warehouse is the destination you want or an additional store alongside Snowflake, BigQuery or another system. |
| Who installs it | Engineering adds the snippet, then analysts and PMs can investigate interactions without waiting for a new event taxonomy to be deployed. | Engineering typically owns SDKs, identity, flags and self-hosting, while analysts and PMs use the resulting workspace for questions and experiments. |
| What happens when requirements change | Retroactive definitions make it easier to revisit a newly important path in existing sessions. Confirm history length: Free lists six months and Growth lists twelve months. | Raw flexibility supports new questions, but a changed definition may require a query update, backfill decision or explicit event in the next release. |
| Enterprise procurement | Heap is part of Contentsquare, and Premier includes dedicated customer success and region-specific storage. Ask for the exact scope and service terms. | PostHog offers a self-serve usage path, annual plans and volume discounts through sales. Larger organizations should confirm platform packages, support and contract terms. |
Pricing and operating cost
Heap
Heap Free is $0 monthly for up to 10,000 monthly sessions, one project and six months of data history. Growth, Pro and Premier are pricing on request; Growth shows an estimate only after the free snippet is installed.
Paid session pricing is not published, so procurement needs a traffic model and a written definition of a session. Session Replay and warehouse sync may be add-ons; include retention, storage, support and implementation in the quote. A separate feature-flag or experimentation service may be needed if those capabilities are central to the release process.
PostHog
PostHog Free is $0 monthly within its product allowances: 1M analytics events, 5K session recordings, 1M feature-flag requests, 1,500 survey responses and 1M data-warehouse rows are examples listed on the published plan. Usage stops at the free limits. Pay-as-you-go adds usage pricing and seven-year event retention.
A workload can use several meters at once: events, person profiles, recordings, flag requests, surveys and warehouse rows. Self-hosting transfers upgrades, monitoring, backups, scaling and incident response to the engineering team. Custom retention for logs is listed at $0.05 per GB per month retained; confirm which data is covered by the contract.
| Tier or usage case | Heap | PostHog |
|---|---|---|
| Free allowance | $0 monthly for up to 10,000 sessions Core analytics, unlimited enrichment sources, guides integrations, six months of history and one project. | $0 monthly within each product allowance Unlimited team members, one project, one-year retention and community support; each product has its own limit. |
| Growing B2B SaaS | Pricing on request for Growth; no public monthly list price. Growth lists unlimited users and reports, chart customization, CSV exports and twelve months of history. | Pay-as-you-go after the free allowance; exact total depends on product usage. Six projects, seven-year event retention, email support and per-product billing limits; recordings have a separate meter. |
| Enterprise controls | Pricing on request for Pro or Premier; custom session pricing. Pro adds account analytics and Session Replay is an add-on; Premier adds warehouse integration, advanced permissions, dedicated customer success and region-specific storage. | Annual plans and volume discounts are available through sales; pricing on request for a specific mix. Platform packages can add SSO, audit logs, custom roles and project permissions. |
The practical difference is observation versus an operating suite
Heap is the more straightforward choice when the first question is “what did users do?” Install the snippet, let autocapture collect the interaction stream and let analysts define meaningful events from that history. A product manager investigating a confusing activation drop can explore clicks, form actions and paths without waiting for a new release to add an event. That is particularly useful for a B2B SaaS with a large surface area and a small analytics team. Heap’s position inside Contentsquare also points toward a broader digital-experience and enterprise buying motion.
PostHog starts from a wider developer workflow. Its cloud product and open-source MIT-licensed option put analytics beside session replay, feature flags, experiments, surveys and a warehouse. A developer can capture an explicit billing event, expose a new navigation design to one account cohort, inspect a replay and query the result in one workspace. That breadth can remove several integrations, but it means the team must own event names, flag expiry, experiment design and access rules. Flexibility becomes useful only when someone is responsible for keeping it coherent.
Both tools can capture automatically, so the decision is not simply whether a snippet exists. It is about who should control meaning after collection and who should control behavior before release. Put an analyst and an engineer in the same evaluation. Ask them to find a failed onboarding step, connect it to an account, gate a fix for a pilot cohort and explain the result to a customer-success leader. The handoffs in that exercise reveal more than a feature checklist.
A worked scenario: 12 million events, 8,000 active users and 300,000 identified events
Consider a B2B SaaS with 8,000 monthly active users, 12 million product events each month and 300,000 identified-user events inside that total. It has four engineers, two product managers and one data analyst. The team needs account-level reporting, replay for support investigations, feature flags for staged releases and two experiments each quarter. It wants to keep a warehouse copy for finance and customer-success reporting. This is a planning scenario, not a vendor quote; its purpose is to show which quantities need a written answer.
Heap’s Free allowance covers up to 10,000 monthly sessions at $0, but this company is likely to exceed that boundary depending on how its 8,000 users create sessions. Growth, Pro and Premier have no public list price, so the buyer should send Heap the monthly session distribution, desired history, account analytics requirement, replay retention and warehouse destination. The quote should say whether Session Replay and Heap Connect are included, how a session is counted, what happens above the limit and which Contentsquare support terms apply. If the team values retroactive analysis and analyst-led governance, those workflow requirements belong in the commercial conversation.
PostHog supplies 1M analytics events free each month, then bills usage. Its published pricing lists a base rate of $0.0000500 per event for the 1–2M tier and $0.0000343 per event for the 2–15M tier. Identified events also add person-profile charges: the first 1M profile events are free, then the listed 2–15M rate is $0.0000697 per event. The exact bill depends on how the 12M events are distributed between anonymous and identified traffic and whether group analytics is enabled. Recordings, flag requests and warehouse rows have their own allowances. Set product-level billing limits before the first production rollout, then model the same mix in the calculator or order form.
The staffing implication is as important as the meter. With Heap, the analyst can lead event definition while engineering maintains the snippet and warehouse connection; a separate flag service still needs an owner. With PostHog, engineering likely owns SDKs, identity, flags and any self-hosted deployment, while the analyst governs the event dictionary and PMs own experiment questions. Heap is the better first proposal when the main risk is inconsistent metrics. PostHog is the better first proposal when the main risk is a fragmented developer stack and the team can operate usage limits and infrastructure.
Data governance, warehouse and privacy questions
Start with the account model. A B2B product rarely has a one-user-to-one-customer relationship: several people may share a workspace, one person may belong to several workspaces and a parent company may own many accounts. Define the user, group, workspace and parent-company identifiers before comparing dashboards. Then test an identity merge, a user who changes workspaces and an account that is deleted. A tool that looks easy in a single-user demo can become expensive to govern when account joins are left implicit.
For Heap, ask how account analytics, visual event labels and historical changes are exposed to the analyst who owns the metric dictionary. Ask whether Heap Connect includes the warehouse tables you need, whether the sync is retroactive for the selected history and how regional storage affects the contract. For PostHog, ask which product meters apply to the warehouse, how group analytics is priced and whether cloud or self-hosted deployment meets the security team’s backup and deletion requirements. PostHog’s selectable US and EU cloud regions are concrete choices; self-hosting is a larger operational commitment than a checkbox.
Write a governance policy before either pilot starts. Name the event owner, the person who approves a definition used in executive reporting, the retention period for recordings, the process for deleting personal data and the conditions for changing a flag. Include a warehouse reconciliation: one activation metric should be computed in the product tool and from exported data, with an agreed tolerance and explanation for late or dropped events. These controls make a usage-based suite and a custom enterprise quote comparable on the work they create.
Migration and rollout checklist
Moving from Heap to PostHog is more than replacing one snippet. Preserve the Heap event dictionary, visual labels, account definitions and dashboard filters, then decide which interactions should become explicit PostHog events. Map Heap’s retroactive questions to PostHog queries and mark any history that cannot be reproduced without a new event. Recreate flag, replay and experiment workflows only after the identity model is stable.
Moving from PostHog to Heap reverses the emphasis. Export event names, HogQL queries, cohorts, flag exposure data, experiment assignments, recordings policy and warehouse jobs. Use Heap autocapture to cover broad discovery, then create durable virtual events for the five metrics that leadership relies on. Plan a replacement for flags and experiments, and keep PostHog read-only long enough to reconcile account counts and conversion definitions.
- Inventory events, properties, user and group identifiers, dashboards, warehouse jobs, recordings, flags and experiments.
- Choose five business-critical metrics and document the exact filters, owner, retention window and account join for each one.
- Run a parallel staging flow through both SDKs; compare page views, identity merges, account membership and dropped data.
- Pilot one replay investigation and one staged release. Record who can answer the question and who can roll the release back.
- Set PostHog billing limits per product or obtain a Heap proposal that spells out sessions, history, replay, warehouse and support.
- Keep the old dashboards available while the new definitions complete one full reporting cycle; then archive, rather than silently overwrite, the old queries.
Questions to ask before signing
- For Heap, what session definition, history length, replay allowance, account analytics scope and warehouse sync are included in the quoted tier?
- For PostHog, which products are covered by the free allowance, what is metered after the allowance and can a billing limit stop usage at a chosen spend?
- Can an analyst define a trusted event from existing autocaptured data, and who reviews a definition before it becomes a shared metric?
- How are workspace, user, group and parent-account identities represented in reports and in warehouse exports?
- Which team owns feature-flag expiry, experiment analysis, recording retention, deletion requests and incident response?
- Does security require EU storage, self-hosting, audit logs, SSO, custom roles or a documented backup and recovery process?
Pros and cons
Heap
Pros
- Broad autocapture supports discovery before a complete event taxonomy exists.
- Visual event definition and retroactive analysis suit analyst-led teams.
- Contentsquare ownership, account analytics and regional storage options support an enterprise procurement path.
- Free plan is $0 for up to 10,000 monthly sessions.
Cons
- Growth, Pro and Premier prices are pricing on request.
- Session Replay is listed as an add-on and warehouse sync needs scope in the quote.
- Feature flags and a complete experiment workflow need separate validation or tooling.
- Session volume can move a team out of the free allowance quickly.
PostHog
Pros
- Analytics, replay, feature flags, experiments, surveys and warehouse share one suite.
- Open-source MIT-licensed deployment and cloud region choices give infrastructure flexibility.
- Free monthly allowances and product-level billing limits support a controlled first rollout.
- Explicit events and HogQL give engineers a direct path to complex account analysis.
Cons
- Several usage meters make a realistic volume model necessary.
- The team must create conventions for event names, metric review and flag cleanup.
- Self-hosting transfers upgrades, monitoring, backups and incident response to engineering.
- Identified events, person profiles and group analytics can change the cost of the same event volume.
Who should choose what?
Analyst-led product organization
Several squads need shared activation and retention definitions, while engineering time for instrumentation is limited.
Recommendation: Heap. Start with Heap because autocapture and visual, retroactive definitions let the analyst investigate behavior before every question has an explicit event.
Developer-led release team
Engineers want analytics, replay, flags and experiments beside the code and may self-host.
Recommendation: PostHog. Start with PostHog because the suite connects those release workflows and offers cloud or open-source deployment choices.
12M-event B2B SaaS
A four-engineer, two-PM team has 8,000 monthly active users, 300,000 identified events and account reporting needs.
Recommendation: PostHog. Model PostHog’s event, person-profile, replay, flag and warehouse meters with billing limits, then compare that operating cost with Heap’s custom session proposal.
Strict enterprise procurement
Security needs regional storage, access controls, a named support model and a warehouse path.
Recommendation: Heap. Put Heap’s Premier scope and Contentsquare support terms beside PostHog’s platform packages and deployment responsibilities in the procurement review.
The strongest fit is the product whose ownership model matches the people who will maintain identities, definitions, flags, experiments, replay policy and warehouse reconciliation.
Final verdict
Choose Heap if
- You want autocapture to preserve discovery options before instrumentation is complete.
- Analysts and PMs should define events visually from existing history.
- Contentsquare enterprise support, account analytics or regional storage belong in the buying brief.
- You can request and evaluate custom session pricing.
Choose PostHog if
- Engineering wants analytics, replay, flags, experiments, surveys and warehouse features together.
- Open-source deployment, self-hosting or selectable cloud regions matter.
- The team can govern raw events, person profiles, group analytics and flag cleanup.
- You prefer usage-based billing with product-level limits over a custom session quote.
FAQ
Does Heap or PostHog have better autocapture?
Both offer autocapture. Heap makes it the center of its workflow and emphasizes visual, retroactive event definition. PostHog combines autocapture with explicit events and properties, which suits teams that want engineers to define a durable event contract. Test the same staging flow and compare the data you can turn into a trusted metric.
Which tool is better for feature flags and experiments?
PostHog. Feature flags are a core product and experiments are billed with feature flags. Heap can analyze exposure data from another system, but its published product details do not describe it as a complete flag-and-experiment replacement. Keep flag ownership and expiry in the rollout plan either way.
How much does Heap cost?
Heap Free is $0 monthly for up to 10,000 monthly sessions, one project and six months of history. Growth, Pro and Premier use custom session pricing or an estimate after the free snippet is installed, so the honest paid answer is pricing on request. Ask for session definition, replay, warehouse, history and support terms in writing.
How does PostHog pricing work?
PostHog has a $0 monthly allowance for each product, including examples such as 1M analytics events, 5K recordings, 1M flag requests, 1,500 survey responses and 1M warehouse rows. After the allowance it is pay-as-you-go. Published event rates include $0.0000500 per event in the 1–2M tier and $0.0000343 in the 2–15M tier; identified events can also add person-profile charges. Model each meter rather than multiplying one headline rate.
Can PostHog be self-hosted?
Yes. PostHog describes an open-source MIT-licensed product. Self-hosting means your team owns infrastructure, upgrades, monitoring, backups, scaling, deletion workflows and incident response. Cloud customers can select US (Virginia) or EU (Frankfurt), which may be a simpler route when the requirement is regional storage rather than operating the platform.
Which one should a B2B SaaS choose?
Choose Heap when analyst-led discovery, autocapture and a Contentsquare enterprise relationship matter most. Choose PostHog when developers need analytics beside replay, flags, experiments and a warehouse, or when open-source deployment is a requirement. For a 12M-event workload, ask both vendors to price the same identity, retention, replay, warehouse and support assumptions.