Pricing
Understand the cost before you send the data
Start free. Upgrade only when you need more than 5,000 stored events.
Spark
$0/ month
Free up to 5k events
- Point-and-click visualizations
- Organized event collections
- Customizable dashboards
Usage-based
Published rate$12/ month
per million stored events
- Everything in Spark
- Automated alerts
- Full ClickHouse SQL notebooks
- Configurable data retention
- Direct support
No setup fee
Create an account and start sending events.
No read meter
Normal dashboard, query, and embed usage is included.
Retention controls
Expire collection data you no longer need to store.
Compare
What each plan includes
| Feature | Spark | Usage-based |
|---|---|---|
| Stored events | Up to 5,000 | $12 per million / month |
| Point-and-click visualizations | Included | Included |
| Organized event collections | Included | Included |
| Customizable dashboards | Included | Included |
| Automated alerts | Not included | Included |
| Full ClickHouse SQL notebooks | Not included | Included |
| Configurable data retention | Not included | Included |
| Direct support | Not included | Included |
Need capacity planning or a custom arrangement? Talk to us
What am I paying for?
The paid plan includes the complete analytics workflow: visualizations, dashboards, alerts, SQL notebooks, embeds, and configurable retention. Your subscription also supports an independent product team, so support and product decisions stay close to the people using GraphJSON.
How is event pricing calculated?
Every 15 minutes, GraphJSON records a watermark from your total stored-event count—the same count
shown in the all_events collection. At the end of the month, billing uses the highest watermark
recorded during that month. Deleted or expired events are not included in later watermarks.
Paid usage is metered in increments of one million stored events.
Do you charge based on reads?
No separate read fee applies to normal product usage. A persistently high rate of unique or computationally expensive queries may be throttled to protect the service. If your workload is read-heavy by design, email hi@graphjson.com and we can plan capacity with you.
Embedded graphs are cached with a 15-minute TTL by default, so repeated views of the same graph do not produce a new database query every time.
What happens if I go over the free tier usage limits?
GraphJSON will ask you to upgrade. We do not deliberately break logging or existing embeds at the instant you cross the threshold, giving you time to review your data and choose a plan. Continued usage above the free limit requires a paid subscription.
Can I control how long data is stored?
Yes. Retention is configured per collection, so operational logs can expire quickly while business events remain available longer. Shorter retention can reduce stored-event volume. Export anything you must archive before shortening a retention window.
Do I own my data?
Yes. GraphJSON does not sell your event data. You can export supported Samples, table, and API results, configure collection retention, and delete collections or clear workspace events. GraphJSON does not currently document a self-service whole-workspace raw archive or row-level deletion API. Review the data portability guide before relying on a specific export or deletion workflow.
For invoice review, payment details, cancellation preparation, and the difference between current storage and a monthly high-water mark, use the billing and subscription guide.
Who do I contact if I have more questions?
Email hi@graphjson.com, use the chat button in the lower-right corner, or message @GraphJSON.