GraphJSON vs PostHog

Developers comparing analytics tools often land on the same shortlist, and PostHog is usually on it. In this post we compare GraphJSON and PostHog - two tools that both store events but aim at different sweet spots. (Also see our takes on Google Analytics, Mixpanel, and Datadog.)

Introduction to GraphJSON#

GraphJSON is an independently built event analytics product. You log JSON events, and it provides graphs, dashboards, alerts, SQL notebooks, and embeds. Events are stored in ClickHouse, so there is no analytics infrastructure for your team to run.

Introduction to PostHog#

PostHog is a popular open-source product analytics platform. It bundles a broad suite - product analytics, session replay, feature flags, A/B testing, and more - and can be used as a hosted cloud service or self-hosted on your own infrastructure.

GraphJSON vs PostHog: Key Differences#

1. Simplicity and Focus GraphJSON has near-zero setup: logging is a single JSON POST to our API, so any language that can send an HTTP request works - no SDK to install, no suite to learn. You can go from zero to a real graph in minutes.

   PostHog is an all-in-one suite. That's powerful, but breadth comes with surface area: more concepts to learn, more UI to navigate, and more decisions before you see your first insight.

2. Data Storage and Flexibility GraphJSON stores your events as schemaless JSON in ClickHouse. Log whatever payload you have today, add fields whenever you want, and use typeahead suggestions to discover your field names and values while exploring.

   PostHog captures structured events through its SDKs and autocapture. It works well for standard product analytics, but working with arbitrary custom JSON structures is not its focus.

3. Custom Querying and Data Analysis GraphJSON exposes full ClickHouse SQL notebooks. Joins, window functions, complex aggregations - you can query your events directly and visualize the results, no data export required.

   PostHog lets you build insights through its UI and offers SQL access via HogQL on paid tiers. If raw, unrestricted SQL against your events is your happy place, GraphJSON leans into it fully.

4. Embedding and Sharing GraphJSON graphs can be embedded anywhere as iframes, cached with a 15 minute TTL so millions of views are no problem. The dynamic Embed API lets you change filters per request - perfect for personalized per-user dashboards or public /open metrics pages.

   PostHog's insights and dashboards are mainly designed for internal use by your team rather than embedding into your own product.

5. Pricing and Operations GraphJSON's pricing is public: free up to 5,000 stored events, then $12 per million stored events per month. See the pricing page for current metering details. There is no analytics infrastructure for your team to operate.

   PostHog's cloud pricing is usage-based across its many products, and the self-hosted route trades money for operational burden - you're running a real analytics stack yourself.

Where PostHog Shines#

Credit where it's due: if you want one open-source platform that covers product analytics plus feature flags, experiments, and session replay, PostHog is genuinely impressive. The self-host option also matters to teams with strict data residency requirements.

How to choose#

Choose PostHog when you want a broad, integrated product stack—analytics, replay, feature flags, and experimentation—with an open-source deployment option. Choose GraphJSON when you want a focused managed event store, direct ClickHouse SQL, simple HTTP ingestion, and customer-facing embeds with less surface area.

Both are developer-friendly products with different boundaries. Evaluate the operational model and the workflows you will use every week, not the longest feature list.