In short
Mixpanel or Amplitude if you used the product analytics. Matomo if you want a complete privacy-analytics suite. Umami or Plausible if PostHog was more product than you needed. mrkr if you used only the analytics and the replay and want them cookieless by default, in one tool. Nothing here replaces feature flags or experiments.
Start by listing which PostHog products you opened last month. That list, not a feature comparison, decides this. Teams who name flags and experiments should stay, and this page will not argue otherwise.
The mrkr overview is the whole product: visitors, sources, pages, goals, and revenue over one range, with nothing to enable first. That is a much smaller surface than PostHog, which is either the point or a dealbreaker.
Nothing on this list has them. No amount of analytics depth compensates for that.
A large allowance across products. Nothing else on this list matches it.
Matomo, Umami, and Plausible keep it. mrkr is proprietary and hosted only.
The picks
Best for deep product analysis
If what you used PostHog for was product analytics, Mixpanel does that half better and does nothing else. Segmentation across arbitrary event properties, funnels, retention, and a mature analysis interface. No feature flags, no experiments, no warehouse, and it is cookie-based.
Best for enterprise behavioural cohorting
The heavier product-analytics option, built for organisations with an analytics function and governance requirements. Deeper cohorting than anything else here and correspondingly more to operate. Like Mixpanel, it replaces one part of PostHog rather than the platform.
Best complete privacy-analytics suite
The other genuinely complete option: analytics, funnels, session recording, heatmaps, form analytics, tag management, and ecommerce. No feature flags and no experiments, and cookies are on by default. Pick it when open source and breadth matter more than the developer tooling PostHog wraps around them.
Best for the bare essentials
Small and cookieless. It does pageviews, sources, events, and funnels, and nothing else. If PostHog was overkill, this is the smallest honest answer.
Best simple hosted option
Cookieless, EU-owned, and deliberately minimal. Traffic, sources, pages, goals, funnels. No replay, no cohorts, no per-visitor detail, no flags. If your PostHog project is mostly a traffic dashboard with nine unused products attached, this is the honest destination.
Best if you used only the analytics and the replay
Our product, so read the gaps first: no feature flags, no experiments, no surveys, no warehouse, and no heatmaps. PostHog beats it on every one of those and on its free tier. What mrkr offers is the analytics-and-replay slice as a single tool that is cookieless by default, with traffic sources, funnels, and revenue attribution in one dataset and under 6 kB on the page.
Still right if you use the breadth
Worth stating clearly: if you ship behind feature flags, run experiments, and query the warehouse, nothing on this list replaces PostHog and moving would be a downgrade. The reasons to leave are the operational surface, the bill once the free allowance is exhausted, and the page weight when several modules are enabled.
Compared
| Tool | From | Banner? | Replay | Funnels |
|---|---|---|---|---|
| PostHog | Usage-based | Depends on configuration | Yes | Yes |
| Mixpanel | Event-volume pricing | Yes | Yes | Yes |
| Amplitude | Event-volume pricing | Yes | Yes | Yes |
| Matomo Cloud | ~EUR 19/mo | Depends on configuration | Yes | Yes |
| Umami | Varies | No | No | Yes |
| Plausible | $9/mo | No | No | Yes |
| mrkr | $1 for 14 days, then from $5/mo | No, in the default cookieless mode | Yes | Yes |
Every vendor in this table details checked against the vendor's own pricing and feature pages on 2026-08-24. Where a vendor does not publish a fixed list price, this table says how they charge rather than guessing a number. Vendors change packaging often, so confirm before you buy.
Live demo
FAQ
Most often because they only ever used two of its products and the rest was surface area nobody owned. After that: page weight once replay and flags are both enabled, and the bill once the free allowance runs out on a busy site.
Matomo and mrkr do. Mixpanel and Amplitude ship replay alongside their analytics as well. Umami and Plausible have none at any tier.
No. Nothing on this list has feature flags or experiments. If those are part of your workflow, you either keep PostHog or add a dedicated flags product next to whatever analytics you choose.
Umami, Plausible, and mrkr are cookieless by default. Matomo can be configured that way. Mixpanel and Amplitude are cookie-based and need consent in the EU, UK, and California.
Plausible, Umami, and mrkr are all in the low single-digit kilobytes. A PostHog bundle grows with the modules you enable, and replay plus flags plus surveys is a meaningful amount of JavaScript on a site measured on Core Web Vitals.
Start with one line of HTML.Your first visitors show up within a minute.