PostHog vs Amplitude 2026: Cost and Fit

PostHog and Amplitude are aimed at the same job from opposite directions. Amplitude is an enterprise analytics platform sold through a sales team; PostHog is an open source developer tool that publishes every rate and expects you to self-serve. Both now bundle analytics, session replay and experimentation, so the choice comes down to who on your team will own it and how much predictability you need from the bill.

The short answer: PostHog if engineers own it, Amplitude if the enterprise does, Arktis if you sell things

Pick PostHog when your engineers want analytics, replay, feature flags and experiments in one stack, are comfortable in a query editor, and want published rates they can multiply out. Pick Amplitude when analytics is owned by a product or data organisation rather than by engineering, when you need enterprise governance and warehouse-native deployment, and when a procurement cycle is normal rather than an obstacle.

If you run an online store, Arktis is the clear winner over both. PostHog and Amplitude both tell you what a visitor did in your product. Neither tells you which ad produced the order, because neither ingests ad spend or matches ad click identifiers to purchases. Arktis does exactly that, reports customer acquisition cost and return on ad spend per campaign with five switchable attribution models, and plays the buyer's session recording next to the number, at $49, $149 and $349 a month with every tier published and a 7-day free trial on Growth.

Key takeaways

PostHog publishes a rate for every product it sells: one million events and 5,000 recordings a month free, then $0.000198 per identified event and $0.00005 per anonymous one, with replay, feature flags, surveys and experiments each metered on their own. Amplitude publishes a free tier of 2 million events and 10,000 replays and no rate above it; Vendr's marketplace data, updated February 2026, reports a median contract of $64,724 a year across 237 purchases. PostHog is open source and self-hostable, though it treats self-hosted deployments as officially unsupported and keeps paid features on Cloud. Neither does marketing attribution.

The pricing models are opposites

PostHog's approach is to publish everything and let you add up the parts. That is genuinely cheaper at small and mid volumes: the event rate is roughly a fifth of Mixpanel's list price, and the free allowances on each product are large enough that many small teams pay nothing for a long time. The catch is that four meters are four ways to be surprised. A team that turns on session replay across a high-traffic marketing site, or ships a feature flag evaluated on every page load, can move the bill without touching the event count they were watching.

Amplitude's approach is the opposite: one conversation, one number, no public rate. The pricing page carries an estimate link rather than a rate card, Plus is described only as growing with event volume up to 70 million events, and add-on expansions are sold as a percentage of the platform plan. Amplitude's own conference commentary puts the typical uplift at about 30 percent for each additional module. Several things teams assume are included, extra replays, extended replay retention, AI replay summaries, multi-armed bandits, holdout groups, approval workflows, are line items.

Where PostHog is genuinely ahead

It is open source, which means you can read the code, run it yourself, and keep your data inside your own infrastructure without asking permission. For a company with a hard data residency rule that is not a preference, it is the whole decision.

It is also one stack rather than a platform plus add-ons. Analytics, replay, feature flags, experiments, surveys and a data warehouse layer all ship together, and the free tier covers a slice of each, so an engineer can evaluate the full product on a Saturday without talking to anyone. And the published rate means a finance team can model a year from a spreadsheet.

Where Amplitude is genuinely ahead

Non-technical self-service is the honest gap. PostHog assumes fluency, and reviewers consistently describe it as a tool engineers love and marketers avoid. Amplitude is not simple either, but it is built so that a product manager can get to an answer without writing a query, and its behavioural depth at the top end, particularly pathfinding and predictive cohorts, is better.

Governance is the second gap: enterprise controls, approval workflows and role management are Amplitude's territory. Warehouse-native deployment on Snowflake is the third, running inside Snowflake's container service against your own tables with the first five seats included. Amplitude also holds 4.5 out of 5 on G2 from more than 2,700 reviews of its analytics product, so the criticisms here are about cost of ownership, not quality.

Why neither answers a store's question

Both are product analytics tools, and an online store is not a product in that sense. The number that decides whether the business works is the cost of acquiring the customers who converted, and reaching it means capturing fbclid and gclid when the visitor lands, holding those identifiers across sessions, matching a Shopify order or a Stripe customer back to the visitor, then dividing campaign spend by attributed conversions. PostHog and Amplitude do not do any of those steps, and event instrumentation cannot close the gap because the ad platform spend data is not in either tool.

Arktis is ours, so weigh this accordingly. It does those steps as the product rather than as an integration, offers first-touch, last-touch, linear, time-decay and position-based models so you can see how much the answer depends on the model you chose, and puts the session recording of each attributed buyer next to the revenue, gated behind explicit consent, with capture buffered in the browser and discarded unless consent is given, with EU hosting by default. Priced on tracked visitors, a campaign that triples traffic for a week does not produce an invoice surprise. It has no feature flags and no experimentation platform; if those matter, PostHog is the better tool and this is not a close call.

How to choose

Ask who will open the tool on a Tuesday. If the answer is an engineer, PostHog wins on cost and on breadth. If the answer is a product manager or an analyst supporting a wider organisation, Amplitude earns its price. If the answer is a founder deciding which campaigns to cut this week, neither tool has the data, and you want attribution instead.

To see attribution and recordings on your own traffic, start the 7-day Growth trial. The tiers are on the pricing page, Amplitude pricing covers what the company does and does not publish, and Amplitude alternatives compares eight options including both tools here.

Sources

PostHog pricing page, accessed 16 September 2026

PostHog source repository, open source licensing and self-hosting

Amplitude pricing page, accessed 16 September 2026

Amplitude: Snowflake-native Amplitude

Vendr: Amplitude contract data, median $64,724 based on 416 purchases and 237 deals handled, updated February 2026

G2: Amplitude Analytics reviews, 4.5 out of 5 from more than 2,700 reviews

G2: PostHog reviews