PostHog Alternatives 2026: Analytics and Replay Tools

PostHog is hard to beat on breadth, so most teams that look for an alternative want less of it, not more: a smaller bill past the free tier, analytics they can run on their own server without a hobby-scale caveat, or a tool that only does the one job they use PostHog for. PostHog Cloud is free up to 1 million events and 5,000 session recordings a month and metered after that, and its self-hosted edition is a single-machine Docker Compose deployment that PostHog itself does not support. The alternatives below fall into three groups: self-hosted tools that cover analytics and replay (Temps, Umami, Matomo, OpenReplay), a focused web analytics tool (Plausible), and hosted product analytics suites with free tiers of their own (Mixpanel, Amplitude, Microsoft Clarity).

TL;DR: Pick Temps if you want web analytics, funnels and session replay included in the self-hosted platform that deploys your app, next to error tracking and uptime monitoring. Pick Umami for cookie-free web analytics that now records replays and heatmaps when self-hosted. Pick OpenReplay for debugging-grade replay on your own servers. Stay on PostHog, or move to Mixpanel or Amplitude, if cohorts, retention analysis and experiments are the reason you use it.

PostHog Alternatives at a Glance

Prices verified on October 10, 2026 on each vendor's pricing page; check them before you budget.

Why Do Teams Look for PostHog Alternatives?

PostHog's pricing is transparent and its free tier is generous, so the reasons people leave are rarely "it's too expensive at small scale". They are about what happens past the free tier, what self-hosting really means, and how much of the product a team actually uses.

Every product is metered past the free tier

PostHog's free plan resets monthly: 1 million analytics events, 5,000 session recordings, 1 million feature flag requests, 100,000 error-tracking exceptions, 1,500 survey responses and 1 million data warehouse rows. On pay-as-you-go there is no base fee, and each product is billed per unit above its allowance: $0.00005 per event from 1 to 2 million, and $0.005 per recording from 5,001 to 15,000. Rates fall as volume grows, and you can set a billing limit per product.

A worked example from those published rates: 2 million events and 15,000 recordings a month cost $50 for the second million events plus $50 for the 10,000 extra recordings, $100 in total. If those events carry person profiles, PostHog bills them as identified events, an add-on at $0.000198 per event between 1 and 2 million, which adds another $198 at that volume. None of this is unreasonable, but it grows with traffic, which is what pushes some teams to look at tools that cost the same whatever the volume.

The free plan also has limits beyond volume: one project, one year of data retention, and recordings kept for one month. Pay-as-you-go raises that to six projects, seven years of data retention and 90 days for recordings. SSO enforcement, SAML and RBAC sit in platform packages: Boost at $250/month, Scale at $750/month and Enterprise at $2,000/month.

Self-hosting PostHog is a hobby deployment

PostHog is open source, but its own docs describe the self-hosted edition narrowly. More on that in the next section.

Most teams use a slice of it

PostHog now spans product analytics, web analytics, session replay, feature flags, experiments, surveys, error tracking, a data warehouse, data pipelines, logs and AI observability. A team that only looks at traffic and watches the occasional replay is carrying a large product to get there. A smaller tool can be easier to run, easier to explain to a privacy reviewer and cheaper at volume.

It sets a cookie by default

PostHog's JavaScript SDK defaults to localStorage+cookie persistence and sets a first-party cookie named ph_<project_token>_posthog that expires after 365 days. PostHog documents a cookieless mode and consent controls, so this is configurable, but with the default setup EU sites generally need consent before it runs. That pushes some teams toward cookie-free tools such as Umami or Plausible.

Can You Just Self-Host PostHog?

You can, with clear limits. PostHog offers "a free Docker Compose deployment under an MIT license", installed with its deploy-hobby script. What the docs say about it:

• It is unsupported. "Self-hosted deployments are officially unsupported", PostHog offers no paid support plans for them, and it states it cannot help debug or recover from data loss.
• It runs on one machine. The open-source disclaimer says "everything runs on a single machine, and thus is unlikely to scale past a couple 100ks events without significant effort in scaling." The decision flowchart asks whether you expect more than 300k events, 1k recordings or 300k /flags calls before recommending Cloud.
• It needs a large box. The requirement is "something equivalent to a Hetzner VM with 4 vCPU, 16GB RAM, and more than 30GB storage."
• Paid features stay on Cloud. "All paid-plan features are Cloud-only." The disclaimer lists what self-hosters lose, including group analytics, data pipelines, one-year event retention, downloading recordings, multi-environment feature flags and multi-question surveys.
• There are no versions. PostHog doesn't tag self-hosted releases or publish CVEs; every change ships continuously from master, and it recommends always running the latest image. New Kubernetes deployments are no longer supported.

So self-hosted PostHog is a reasonable way to try the product or run a small site. It is not the production self-hosting path, and PostHog says so itself. If keeping data on your own servers is the goal, the tools below are built to be run that way.

---

1. Temps: Analytics and Replay Built Into Your Deploy Platform

Temps is a self-hosted deployment platform with web analytics, funnels and session replay built in. It is one binary you install on your own server, and the same install deploys your apps from Git, tracks errors through a Sentry-compatible DSN, runs uptime checks and collects OpenTelemetry traces, metrics and logs. It replaces seven SaaS tools: Vercel (git-push deployments), PostHog (web analytics and session replay), Sentry (error tracking), Pingdom (uptime monitoring), Datadog (OpenTelemetry traces, metrics and logs), Resend (transactional email) and E2B (AI agent sandboxes).

What the analytics cover: visitors, pageviews, referrers, UTM and device breakdowns, custom events with properties, funnels, Core Web Vitals and AI crawler traffic, stored in TimescaleDB on your own server. Session replay is built on rrweb, off by default, and turned on per project; it records the DOM and user input, with input values masked by default (password and email fields always). It does not capture console logs or network requests. Replays are kept for 15 days. See the analytics docs and session replay docs for setup.

Cookies: the Temps proxy sets two first-party, HttpOnly cookies, _temps_visitor_id (365 days) and _temps_sid (30 minutes). That is a different trade-off from PostHog's single cookie plus localStorage, not an absence of cookies, so EU sites generally still need consent. The React provider can stay disabled until a visitor accepts.

Pricing: free to self-host under MIT / Apache 2.0, with unlimited projects and users. Events and replays aren't metered; the limit is your server's disk. A Premium licence ($399 per organization per month) adds SAML SSO, SCIM, custom roles and approval gates for teams that need them. Temps Cloud, a hosted control plane for servers you own, is coming soon from $29/month.

Pros

• Analytics, funnels and replay in the same install as deployments, error tracking and uptime monitoring
• No per-event or per-recording metering
• Self-hosting is the primary way Temps runs, not a hobby edition
• Data stays on the server you choose
• Free under MIT / Apache 2.0

Cons

• Far less analytical depth than PostHog: no cohorts, retention analysis, experiments, surveys or SQL access to your data
• Feature flags are per-environment values with a kill switch; there is no targeting or percentage rollout
• Funnels are basic, and there are no heatmaps
• Replay has no console or network capture, and its 15-day retention isn't configurable
• Two first-party cookies, so a consent banner is still part of the setup in the EU
• You run the server; anonymous telemetry is on by default (TEMPS_TELEMETRY=0 turns it off)
• Temps is pre-1.0 and has a smaller community than PostHog, which is also why it was built with analytics and replay inside the platform from the start rather than as a separate product

Best for: teams that deploy their own apps and want traffic, funnels, replays and errors in one self-hosted tool instead of a separate analytics subscription. If you mostly need product analytics depth, PostHog is still the stronger product. The head-to-head is at Temps vs PostHog.

---

2. Umami: Cookie-Free Web Analytics, Now With Replay

Umami is an MIT-licensed web analytics tool known for a clean dashboard and for working without cookies; its site says it "works without cookies and never collects personal data". It has also grown past pageviews. Version 3.1.0 (April 2026) added session replay built on rrweb, with configurable masking levels, and version 3.2.0 (June 2026) added click and scroll heatmaps with "self-hosted heatmap recording and storage support".

Pricing: self-hosting is free, and the open-source build includes replay and heatmaps. Umami Cloud's Hobby plan is free up to 100K events a month, Pro is $20/month for 1 million events, and Business is $200/month for 10 million events, 5,000 session replays ($0.005 per extra replay) and heatmaps. Replays and heatmaps aren't on Hobby or Pro. Umami notes that some Cloud features, such as email reports and the streaming API, aren't in the self-hosted version.

Pros

• Cookie-free by design, which simplifies consent
• Replay and heatmaps included free when self-hosted
• Simple, fast dashboard; MIT licence

Cons

• Web analytics rather than product analytics: no feature flags, experiments or surveys
• On Umami Cloud, replay starts at the $200/month Business plan
• No error tracking, uptime monitoring or deployments

Best for: teams that use PostHog mainly for traffic stats and the occasional replay, and want a lighter self-hosted tool without cookies. Our Plausible alternatives roundup compares Umami with other web analytics tools.

---

3. Matomo: The Google Analytics-Style Option

Matomo is a long-running web analytics platform released under GPL v3, written in PHP with MySQL or MariaDB. It covers what Google Analytics users expect: acquisition reports, goals, ecommerce, segments, custom dimensions and a tag manager. Cloud data is stored in Frankfurt, Germany, and self-hosted data stays wherever you run it.

Pricing: the on-premise Community edition is free with unlimited users and hits. The features PostHog users tend to want, such as funnels, user flows, form analytics and Heatmap & Session Recording, are premium plugins. On-premise they come in bundles: Team is €275/month (€230/month billed annually) for up to 4 users and 5M hits a month. Matomo Cloud lists heatmaps and session recordings in its feature table, priced by monthly hits.

Pros

• Mature, full-featured web analytics with a long track record
• Free self-hosted core under GPL v3
• Heatmaps and session recordings available from the same vendor

Cons

• Replay, heatmaps and funnels are paid add-ons when self-hosted
• A PHP and MySQL stack to run and upgrade
• Marketing analytics first; no feature flags, experiments or error tracking

Best for: organizations replacing Google Analytics that also want replay from the same vendor and are fine paying for plugins.

---

4. OpenReplay: Self-Hosted Replay for Debugging

OpenReplay is an open-source session replay platform built for engineers. Next to the replay it captures network activity, console logs, JavaScript errors, store state and performance metrics, which is the part of PostHog's replay that Temps and Umami don't match. It also has product analytics (funnels, trends, journeys, heatmaps) tied back to replays, co-browsing with live cursor control and calls, and native mobile replay for iOS and Android.

Pricing and licence: the open-source edition is free to self-host and community-supported. OpenReplay's pricing page marks it for "low traffic". Its Dedicated plan starts at $199/month, billed hourly ($0.276/hour), and runs in OpenReplay's cloud or yours; there is also a usage-based Serverless plan. The repository is mostly AGPLv3, with some MIT directories and an ee/ directory under a commercial licence. The README asks for a single VM with 2 vCPUs, 8 GB of RAM and 50 GB of storage as a baseline.

Pros

• DevTools-grade replay: network, console and errors next to the recording
• Self-hostable, with guides for the major clouds, Kubernetes and Docker
• Co-browsing and mobile replay

Cons

• A separate multi-service stack to run beside your analytics
• The free edition is positioned for low traffic; scale and support are paid
• No feature flags or experiments; product analytics is lighter than PostHog's

Best for: engineering teams that use PostHog replay mainly to reproduce bugs. The self-hosted session replay roundup compares it with the other replay tools in more detail.

---

5. Plausible: Lightweight Web Analytics Only

Plausible is the option for teams that realize they only ever opened PostHog's web analytics dashboard. It shows traffic, sources, pages, goals and countries on one page, and its site says "No cookies, no persistent identifiers."

Pricing: there is no free plan, only a 30-day trial. Starter is $9/month at 10k pageviews, Growth $14 and Business $19 at the same volume, and prices rise with pageviews. Plausible Community Edition is free to self-host under AGPLv3, but it is a long-term release published twice a year, community-supported only, and it leaves out marketing funnels, ecommerce revenue goals, SSO and the sites API.

Pros

• Very simple, cookie-free and lightweight
• Predictable pricing by pageviews
• Self-hostable Community Edition

Cons

• No session replay, heatmaps, feature flags or experiments
• The self-hosted edition lacks funnels and updates twice a year
• No free cloud plan

Best for: content sites and marketing pages that need privacy-friendly traffic numbers and nothing else. For tools that add replay to this kind of analytics, see privacy-first analytics and session replay platforms.

---

6. Mixpanel: Hosted Product Analytics

Mixpanel is the closest hosted match for PostHog's product analytics: event-based funnels, retention, flows and cohorts, with session replay alongside. It is SaaS-only.

Pricing: the Free plan includes up to 1M events a month, 10K session replays a month and unlimited seats. Growth starts at $0 and is priced by usage up to 20M events, with up to 500K session replays a month; Mixpanel publishes the rates through a calculator rather than a rate table. Enterprise is quote-only.

Pros

• Deep, mature product analytics
• Free plan with twice PostHog's monthly replay allowance
• Unlimited seats on the Free plan

Cons

• No self-hosting; data lives in Mixpanel's cloud
• Usage pricing past the free plan, shown through a calculator
• Enterprise pricing is quote-only

Best for: product teams that want hosted product analytics and don't need to self-host.

---

7. Amplitude: Product Analytics for Larger Teams

Amplitude is another hosted product analytics suite, with behavioral cohorts, heatmaps and session replay. It offers an EU data center option on all plans, but no self-hosted version.

Pricing: the Free plan includes 2M events a month, 10K monthly session replays, one year of data access and one month of replay retention. Plus starts at $0 with the first 2M events free and adds heatmaps, cohorts and two-year retention. Growth and Enterprise are custom-priced and add more replays, SSO and permissions.

Pros

• 2M free events a month, twice PostHog's allowance
• Strong behavioral analysis and cohorts
• EU data center option

Cons

• No self-hosting
• Paid tiers above Plus are custom-priced
• Short replay retention on the Free plan

Best for: larger product organizations that want hosted product analytics with SSO and permissions on the higher tiers.

---

8. Microsoft Clarity: Free Replay and Heatmaps

Clarity is Microsoft's free behavior analytics tool: session recordings, heatmaps and ML-generated insights. Its FAQ says: "Clarity is a free service forever. You never encounter traffic limits or are compelled to upgrade to a paid version."

Pricing: free. Recordings are kept for 30 days; favorites and a random sample are kept for up to 9 months, and heatmaps are available for up to 9 months.

Pros

• Free with no traffic limits
• Recordings and heatmaps without any infrastructure

Cons

• Data is stored in Microsoft's Azure cloud; there is no self-hosted option
• Relies on cookies, and Microsoft says full session replays need consent in consent-required regions
• No product analytics depth, feature flags or experiments

Best for: teams that want free replay and heatmaps and are comfortable with Microsoft hosting the data.

---

How to Choose a PostHog Alternative

Start from the part of PostHog you actually use:

• Traffic stats and the occasional replay. Umami covers both without cookies, and Plausible covers the traffic side alone. Temps covers both inside the platform that deploys your app.
• Replay to debug problems. OpenReplay captures console and network alongside each session. Temps puts replays next to error tracking in one install, but records the DOM and input only.
• Cohorts, retention and experiments. Stay on PostHog, or move to Mixpanel or Amplitude. None of the self-hosted tools here match that depth.
• Data on your own servers. Temps, Umami, Matomo, OpenReplay and Plausible Community Edition all self-host. PostHog's own self-hosted edition is hobby-scale and unsupported.
• A bill that doesn't grow with traffic. Self-hosted tools cost the server they run on. On PostHog, Mixpanel or Amplitude, usage past the free tier is metered.
• Fewer tools overall. If analytics is one of several subscriptions (errors, uptime, deployments), Temps folds them into one install. Compare the details at Temps vs PostHog.

Moving Off PostHog

Whichever tool you pick, the steps are similar:

1. List what you use. Pageviews and custom events move easily. Feature flags, experiments, surveys and cohorts need a destination before you switch, or a decision to keep PostHog for them.
2. Swap the snippet. Remove the posthog-js initialization and add the new tool's script or SDK. For Temps, that is <TempsAnalyticsProvider> from @temps-sdk/react-analytics, or a script tag for other sites, as described in the analytics docs.
3. Keep event names. Send the same custom event names so dashboards and funnels can be rebuilt with familiar steps.
4. Run both for a while. A couple of weeks in parallel shows whether the numbers line up before you remove PostHog.
5. Export what you need. Historical events don't move into the new tool automatically, so export the reports or raw data you want to keep before closing the PostHog project.
6. Update your cookie notice. Cookie names differ between tools; Temps' two cookies, for example, are set by its proxy and belong in the notice.

Frequently Asked Questions

What is the best PostHog alternative in 2026?

It depends on which part of PostHog you use. For web analytics, funnels and session replay that come with your self-hosted deployment platform, Temps. For cookie-free web analytics with self-hosted replay and heatmaps, Umami. For debugging-focused replay, OpenReplay. For hosted product analytics with cohorts and retention, Mixpanel or Amplitude.

Is PostHog free?

PostHog Cloud has a free plan that resets monthly: 1 million analytics events, 5,000 session recordings, 1 million feature flag requests and 100,000 error-tracking exceptions, with one project. Usage stops at those limits unless you add a card and move to pay-as-you-go, which bills per unit above the same free allowance.

Can I self-host PostHog?

Yes, with a Docker Compose "hobby" deployment under the MIT licence. PostHog's docs say self-hosted deployments are officially unsupported, run on a single machine that is unlikely to scale past a few hundred thousand events, need roughly 4 vCPU and 16 GB of RAM, and don't include paid-plan features.

Which PostHog alternatives can I self-host?

Temps, Umami, Matomo, OpenReplay and Plausible Community Edition. Temps and Umami both cover web analytics and session replay in one install; Temps adds error tracking, uptime monitoring and deployments, while Umami is cookie-free and adds heatmaps. Mixpanel, Amplitude and Microsoft Clarity are hosted only. For Temps against PostHog feature by feature, see Temps vs PostHog.

Is there a free PostHog alternative with session replay?

Microsoft Clarity is free with no traffic limits, but it is hosted by Microsoft. For self-hosting at no licence cost, Temps, Umami and OpenReplay's open-source edition all record replays; you pay only for the server. Mixpanel and Amplitude include 10K replays a month on their free plans.

Is there a cookie-free PostHog alternative?

Umami and Plausible both work without cookies. PostHog can be configured for cookieless tracking, but sets a first-party cookie by default. Temps sets two first-party cookies, so it is not the pick if avoiding cookies is the requirement.