Mixpanel Pricing Explained: Events, Plans and Surprises

Mixpanel bills on usage. You send events, and what you pay follows how many you send, along with the plan you pick. That model is fair in principle, since heavy users pay more. In practice it makes the bill hard to predict, because event volume is decided by engineering choices as much as by business growth.

This post explains how usage-based analytics pricing behaves, what to check on Mixpanel's own pricing page, and how a flat per-site price like OnyxMetric's changes the math. We sell OnyxMetric, so treat our comparison as one side of the story. We avoid quoting exact Mixpanel rates because they change; use their current page for numbers.

The shape of the pricing

At time of writing, Mixpanel offers a free tier with a monthly event allowance and paid plans above it. Paid plans are priced by the number of events you track, with per-event rates that drop as volume rises. Higher plans add features such as advanced governance, data pipelines and longer support options. Details, thresholds and names shift over time, so verify before you plan a budget.

The takeaway is structural: the unit of billing is the event, not the site or the seat.

Why event billing surprises people

An event is one tracked action: a click, a view, a form submit. Sounds simple. The surprises come from how many events a normal product generates.

Instrumentation decisions. Tracking every button click yields far more events than tracking five key actions. Two teams with identical traffic can have bills several times apart.

Growth compounds. Double your users and you roughly double events. If you also add instrumentation as features ship, volume rises faster than users.

Noise counts too. Bot traffic, retries and duplicated calls can all become billed events unless filtered.

Forecasting is hard. To estimate next quarter's bill you need a model of user growth times events per user, and the second number moves whenever someone ships new tracking.

None of this is unique to Mixpanel. It applies to any tool that meters events. It is why finance teams often prefer predictable pricing, even at a higher average cost.

A worked example

Picture a SaaS app with 20,000 monthly active users. If each user triggers 40 tracked events a month, that is 800,000 events. Add a redesign that instruments every menu and toggle, and per-user events can jump to 120, which is 2.4 million. Nothing changed about the business, yet the metered volume tripled. A team that planned its budget from the first number is now explaining the second to finance.

The same app on a flat plan pays the same on both days. That is the whole argument for flat pricing, and also its limit: if your real volume is small, you overpay for predictability you did not need.

What the free tier is good for

A free allowance is useful for early-stage products, internal tools and evaluation. You can learn the tool, build funnels, and decide if it fits. It becomes uncomfortable when you cross the limit mid-month, or when you must prune tracking to stay free rather than tracking what you need.

A reasonable rule: use the free tier to prove value, but model what the paid price looks like at twice and five times your current volume before you build dashboards the whole company depends on.

Questions to ask before committing

1. What does the bill look like at 2x and 5x current events?
2. Which features do I need that sit on higher plans?
3. Are replay, experiments or other add-ons billed separately?
4. How do overages work: automatic upgrade, throttling or blocked data?
5. Who owns tracking hygiene so volume does not balloon?

If you cannot answer the fourth one from public documentation, ask sales in writing.

The flat alternative

OnyxMetric prices per site per month instead of per event: Base is $99, Plus is $199, Pro is $299, with yearly billing at ten times the monthly price. Sending more events does not change the number on the invoice.

What the tiers include, per the product today:

• Pageview, session and real-time reporting, plus custom events and goals.
• Funnels and retention.
• User profiles and sessions, with session replay on Pro.
• Web Vitals, error tracking, feature flags and experiments.
• Cookieless tracking.

Flat pricing has a cost of its own. For a tiny product with a few thousand events, a free tier will beat $99 per month. Flat pricing also charges per site, so a company with many small sites pays per site. It wins when volume is high, when you instrument liberally, or when a predictable invoice matters more than a bargain at low volume.

Where Mixpanel is stronger

Be honest about fit. Mixpanel has years of depth in cohort analysis, flows and user-property segmentation for logged-in app behavior, and a large community of analysts who know it. If your core questions are about in-app behavior across many custom events, it remains a serious choice. OnyxMetric is stronger when you also want website traffic, sources, Web Vitals and errors in the same tool.

How to compare with your own data

Take one month of real event counts from your current tool. Price it on Mixpanel's calculator, then compare with the OnyxMetric tier you would need. If you are not tracking yet, install the OnyxMetric script for a trial period and count what a normal month produces.

Our Mixpanel comparison page lays out the feature differences side by side, and the alternatives guide covers other tools worth testing.