Mixpanel Pricing 2026: The Real Cost Per Event
Mixpanel pricing in 2026: 1M events free, then $0.00028 per event. What Growth costs at real volumes, the add-on surcharges, and a cheaper option.
Wilmer
Co-founder & CEO

Mixpanel is one of the few analytics vendors that still publishes a rate. One million events a month are free, and above that you pay $0.00028 per event, which is 28 cents per thousand. That transparency is unusual in a category where most competitors have retreated behind a quote. It also makes Mixpanel easy to model and easy to get wrong, because the rate is only half the story: add-ons carry percentage surcharges on your overage, you cannot set a billing limit, and events are counted when they arrive rather than when they happened. This guide covers the published prices as of 16 September 2026 and what they mean for an online store.
The short answer: Mixpanel for product teams, Arktis for stores
Mixpanel is the better choice for a product team that wants event analytics at a published rate and does not need attribution. Its free tier is large, its interface is the easiest in the category, and it charges the same for an anonymous event as an identified one, which most rivals do not. For an e-commerce store, Arktis is the clear winner. Event pricing means the bill rises with traffic you have not converted yet, and Mixpanel cannot tell you which campaign produced an order. Arktis charges on tracked visitors at $49, $149 and $349 a month, connects each order to the ad that earned it, and plays the buyer's session recording next to the number.
| Mixpanel | Arktis | |
|---|---|---|
| Priced on | Events, $0.00028 each above 1 million | Monthly tracked visitors |
| Free tier | 1 million events, 10,000 replays | None, 7-day Growth trial |
| Cost at 100,000 visitors | Up to $420 a month at 2.5 million events | $149 with recordings |
| Billing cap | Not possible, alerts only | Fixed tier price |
| Ad attribution | None | fbclid and gclid, CAC and ROAS per campaign |
| Built for | Product and growth teams | Online stores |
Key takeaways
Mixpanel has three plans. Free covers 1 million events a month with 10,000 session replays, unlimited seats and five saved reports per seat. Growth starts at zero, keeps the first million free, charges $0.00028 per event after that and includes 20,000 replays. Enterprise is quoted, and Mixpanel's docs state that its tracked-user Enterprise plan starts at $20,000 a year. The published calculator shows 1.5 million events a month at $140 on monthly billing and $120 on annual, which matches the list rate exactly. Add-ons do not just cost money, they multiply the overage: Group Analytics adds 40 percent of the calculated event overage and Data Pipelines adds 20 percent. There is no way to cap spend, only alerts at 85 percent and above.
What you get free, and what Growth adds
| Free | Growth | Enterprise | |
|---|---|---|---|
| Monthly events | Up to 1 million | First million free, metered to 20 million | Up to 1 trillion |
| Price | $0 | Starts at $0, then $0.00028 per event | Quoted; tracked-user plan from $20,000 a year |
| Session replays | 10,000 a month | 20,000 a month | 20,000, customisable |
| Replay retention | 30 days | 30 days | 7 to 360 days with the add-on |
| Saved reports | 5 per seat | Unlimited | Unlimited |
| Experiments and flags | 1,000 experiment users, 10 flags | 5,000 experiment users, 50 flags | Custom |
Two changes in 2026 make the free and Growth tiers better than most guides describe. In February, single sign-on, data views and sensitive data classification moved down from Enterprise into every Growth plan. In September, experiments and feature flags arrived on Free and Growth, which had previously been an enterprise conversation. If you priced Mixpanel a year ago, the package you were quoted is not the package on sale now.
One widespread error is worth correcting. Several pricing sites say Mixpanel switched to event-based pricing in February 2026. Mixpanel's own documentation says Growth plans purchased after February 2025 include the first million events free. The change is a year older than the internet thinks.
What it costs at real volumes
The published rate makes the arithmetic straightforward, with one caveat: volume discounts exist but are not published, so straight multiplication gives a ceiling rather than a quote.
| Monthly events | Billable events | List-rate cost | Roughly |
|---|---|---|---|
| 1 million | 0 | $0 | 40,000 store sessions |
| 1.5 million | 500,000 | $140, or $120 annual | 60,000 sessions |
| 2.5 million | 1.5 million | Up to $420 | 100,000 sessions |
| 5 million | 4 million | Up to $1,120 | 200,000 sessions |
| 12 million | 11 million | Up to $3,080 | 500,000 sessions |
The session estimates assume about 25 events per session, typical for a store with product views, cart actions and a checkout flow. The $140 at 1.5 million events is not arithmetic, it is what Mixpanel's own calculator displays, and it matches the list rate to the cent. Everything above it is the undiscounted ceiling.
For a store the row that matters is the third. A shop with 100,000 monthly visitors generating 2.5 million events pays up to $420 a month for product analytics that still cannot attribute a sale to a campaign. The same traffic costs $149 on Arktis with session recordings and attribution included.
The parts of the bill that are not the rate
Three mechanics catch teams out, and all three are documented.
Add-ons apply a surcharge to your overage rather than a flat fee. Mixpanel's own example: send 1.1 million events and you are charged for the 100,000 over the plan at the additional data rate, plus 40 percent of that calculated overage for the Group Analytics add-on. Data Pipelines works the same way at 20 percent. The base prices for those add-ons, and for Metric Trees, Warehouse Connectors and paid session replay, are not published at all.
Events are billed at ingestion, not by timestamp. Send an event on 10 January dated 10 December and it counts against January. Backfills count in the month you run them, which makes a historical import a budget event.
You cannot set a billing limit. Mixpanel's documentation says so plainly and offers alerts instead, at 85, 100, 110, 120, 200, 300 and 400 percent of plan volume. On the free plan, going over blocks your reports until the month resets rather than charging you, which is the safer failure mode.
One thing Mixpanel does better than its rivals: it does not distinguish identified from anonymous events for billing, and the identity calls themselves are free. Identify, alias, merge, opt-in and profile updates are all explicitly non-billable. PostHog charges more for events that carry a person profile, so a site with heavy anonymous traffic is cheaper to run here.
What reviewers say
Mixpanel holds 4.5 out of 5 on G2 from 1,372 reviews and 4.5 out of 5 on Capterra from 146, where its lowest sub-score is ease of use at 4.1. G2 simultaneously ranks it first for mid-market usability in product analytics, so the picture is a product people like once it is set up correctly.
The complaints cluster in three places. Cost at scale is the loudest, and its sharpest form is self-censorship: one reviewer describes the event-based model forcing the team to think about which events to fire at all, which is the opposite of what an analytics tool should encourage. The second is implementation burden, where a rushed tracking plan produces messy data that is expensive to fix later. The third is consistency between reports, with reviewers describing funnels that do not match Insights for the same properties.
Where Arktis fits, and where it does not
Arktis is the product we build, so weigh this accordingly. It is not a product analytics tool. There are no cohort explorations across billions of events, no experiments and no feature flags. If your job is understanding how people use a software product, Mixpanel is the right tool and it is not a close call.
For a store the calculus is different. Arktis prices on monthly tracked visitors, so the bill does not move when a campaign spikes your traffic and it cannot be inflated by a chatty tracking plan. It captures fbclid and gclid on landing, imports Shopify orders, matches Stripe customers, and reports customer acquisition cost and return on ad spend per campaign under five switchable attribution models. Session recordings are included from $149 a month, gated behind explicit consent, with capture buffered in the browser and discarded unless consent is given, and hosted in the EU by default.
Our Arktis vs Mixpanel comparison covers the feature detail, and ecommerce attribution software covers the wider field. To see attribution on your own traffic, start the 7-day Growth trial, or read the tiers on the pricing page.
How to keep a Mixpanel bill down
Audit what you send before tuning anything else, because the rate is fixed and volume is the only lever you control. Strip autocaptured events nobody queries, and be careful with high-frequency events such as scroll depth. Set the volume alert at 85 percent rather than waiting for the 100 percent notice, since there is no cap to save you. Plan backfills deliberately, as they land in the month you run them. And if you are approaching 20 million events a month, ask for the volume discount explicitly, because the published rate is the undiscounted one.
Sources
Mixpanel pricing page, accessed 16 September 2026; rate, plan cards and calculator
Mixpanel docs: pricing and billing, overage mechanics, add-on surcharges and the Enterprise tracked-user plan
Mixpanel docs: session replay, allowances and retention
Mixpanel changelog, the February and August 2026 packaging changes
G2: Mixpanel reviews, 4.5 out of 5 from 1,372 reviews on 16 September 2026
Capterra: Mixpanel reviews, 4.5 out of 5 from 146 reviews
Frequently Asked Questions
How much does Mixpanel cost per event?
Above the free million events a month, Mixpanel charges $0.00028 per event, which is 28 cents per thousand, with unpublished volume discounts at higher tiers. Its own calculator shows 1.5 million events a month at $140 on monthly billing or $120 on annual, matching the list rate exactly. A store generating 2.5 million events a month would pay up to $420.
Is Mixpanel free?
Yes, up to 1 million events a month, with 10,000 session replays, unlimited seats and five saved reports per seat. Since September 2026 the free plan also includes experiments and feature flags. Exceeding the free limit does not create a charge; it blocks your reports until the next month or until you upgrade.
Does Mixpanel charge differently for identified and anonymous events?
No. Mixpanel bills raw event volume regardless of whether the user is identified, and the identity calls themselves are free: identify, alias, merge, opt-in and profile updates are all explicitly non-billable. That makes it cheaper than tools priced per tracked user for sites with a lot of anonymous traffic.
Can you cap spending in Mixpanel?
No. Mixpanel's documentation states that setting a billing limit is not currently possible. It offers usage alerts instead, at 85, 100, 110, 120, 200, 300 and 400 percent of your plan volume. Events are billed when they are ingested rather than by their timestamp, so a historical backfill counts against the month you run it.
What is a cheaper alternative to Mixpanel for e-commerce?
For an online store, Arktis costs $49 to $349 a month on published, visitor-based tiers with a 7-day free trial on Growth, and it answers a question Mixpanel does not: which campaign produced the order. It captures click identifiers, matches Shopify and Stripe orders, and reports CAC and ROAS per campaign with consent-gated session recordings. For product analytics specifically, PostHog and Amplitude are the closer comparisons.
Free tools for this topic
From the blog
Try Arktis free for 7 days
Attribution tracking, session recordings, and AI automation in one platform. Set up in 2 minutes, free for 7 days.
Start 7-day free trialWritten by
Wilmer
Co-founder & CEO
Wilmer leads product strategy at Arktis, focusing on privacy-first analytics and attribution tracking for e-commerce brands.