Usage Based Pricing

Seat Based Pricing is Dead

Coding agents are transforming coding output, causing "seats" to no longer be a useful proxy for how much code we process and review. Across our customers in the last three months, each dev "seat" on average produced 1.8x more commits, 1.9x more code reviews, and average reviews that are 1.7x larger. This compounds into much higher costs per seat while our price per seat has remained flat.

In order to better align our model, we're shifting to usage-based pricing:

Code Review

Catch bugs before you ship

$0.05 per KB reviewed

10KB min per review

Best-in-class bug detection
Auto-fixes issues
Auto-approves safe PRs
Custom review rules

Status

Understand what's changing

$0.05 per commit

Commit summaries
Sprint reports and weekly digests
Project classification
Productivity stats for devs and agents

Agent

Ask questions and take action

Included with Status

We plan to eventually charge for Agent above a certain threshold once usage patterns stabilize. We'll give advance notice before any change.

Writes code and ships PRs
Understands your codebase
Available via Slack, GitHub, API
Connects to your tools (e.g. PostHog, GCP, MCP)

Why usage-based pricing

We believe usage-based pricing is the most fair, aligned, and future-proof model for our customers:

Fair: you pay only for what you use and control.
Aligned: costs scale with work we do (process and review the code you push).
Future-proof: this model works for both agent-driven and human-driven code.

Billing and spend controls

You prepay via credits, which are consumed as usage occurs and auto-replenished. New workspaces receive a $100 free credit to get started.

You can configure Spend Controls to set a maximum budget:

Per-review spend cap: $10 default
Per-PR spend cap: $50 default
Monthly spend limit: set by you

You can also restrict usage via exclusions:

Only review certain repos
Only review certain authors
Only review when manually invoked
Skip certain file types (e.g., lock files, generated code)

What to expect

Based on recent historical data across our customers, with this pricing model:

Average review cost would have been: $0.95
Over 50% of reviews would have cost $0.50

By comparison, Claude Code Review costs $15–$25 on average per review.

Note: teams with heavy use of AI coding agents may see higher bills than the legacy seat-based plan without spend controls configured. To estimate what your team's usage would look like, use the below estimator:

Usage Cost Estimator

Avg review ≈ $0.95 · Median $0.50

Code Review$0.05/KB

Reviews / moStatus$0.05/commit

Commits / mo

AgentIncluded

Estimated total~$39.25/mo

For more details, see our Pricing page and Docs.

For our current customers

You'll remain on the legacy seat-based plan until April 27, after which you'll automatically transition to usage-based pricing. If you have a longer term enterprise agreement with us, your pricing is unaffected by this announcement. We hope this gives you time to observe your team's usage patterns and configure spending controls as you see fit.

If your team is interested in an annual commitment at a discounted usage rate, we'd be happy to discuss. Reach out to us at support@macroscope.com.