Is Claude Max worth it? Price your own usage.
Claude Pro is $20/month, Max is $100 or $200. Whether that is a bargain or a waste depends on what you actually run. Paste your own ccusage output below and see your real Claude Code usage valued at list API rates, compared to each plan. It runs entirely in your browser.
Step 1: get your usage
In your terminal, run the open-source ccusage reader. It reads the logs Claude Code already writes to local disk. No account, no API key.
Step 2: paste the output
How this is calculated
Priced at the rate on the day
Each day of usage is valued at the input, output, and cache rates that were live on that day, from the open AI Price Index, not a flat guess. Cache reads are 0.1x input; cache writes 1.25x. It matches how Goei prices your history.
Your usage never leaves your browser
The parser and the price dataset are bundled into this page. What you paste is read, priced, and shown locally. No upload, no account, no API key.
Same engine as the app
This uses the exact open-source parser and pricing engine that runs inside Goei, so the number here matches what you see on your first import. No surprises.
Reading the verdict honestly
A subscription is worth it when the usage it unlocks would cost more at list API rates than the flat monthly fee. A multiple above 1x means the plan is cheaper than paying per token for the same work. A high multiple, say 10x or 20x, means the subscription is a clear bargain for how heavily you use it.
A multiple below 1x is not a failure, it is information: this period you used less than the plan costs at list rates, so a smaller plan (or pay-as-you-go) might fit better. Usage is lumpy, so check a few months before switching. If you paste a full history, the calculator compares your total value against the plan price times the number of months it spans.
Two numbers, never one: what you pay is a real charge; your usage value is what the same work would cost at list API rates, an estimate labeled LOCAL EST. They are compared as a ratio and never added together. That is the same discipline Goei uses everywhere.
Track it across a team
To track Claude Code costs across a team, each developer runs the open-source budgetclaw CLI, which reads the session logs Claude Code already writes and pushes daily dollar-and-token rollups to Goei, a hosted dashboard that dedupes across machines and teammates: cost per project, per developer, per git branch. No API keys, no prompts, nothing in the request path.
See the Goei overview, how it compares to ccusage, or the re-priced history behind the numbers.
Frequently asked questions
Is Claude Max worth it? +
It depends on how much you actually use it. Claude Max costs $100/month (5x) or $200/month (20x); Claude Pro is $20/month. This calculator values your real Claude Code usage at current list API rates from the open AI Price Index, then compares that value to the plan price. If your usage is worth several times the plan, the flat subscription is saving you money versus paying per token; if it is worth less than the plan, you may be on a larger plan than you need. Paste your own ccusage --json above to see your number.
How do I find out what my Claude Code usage is worth? +
Run `npx ccusage@latest --json` in your terminal. ccusage reads the session logs Claude Code already writes to local disk, with no account and no API key, and prints your daily usage as JSON. Paste that output above and this page prices every day at the input, output, and cache rates that were live on the day it ran, using the open AI Price Index, then sums it.
Does this calculator upload my usage data? +
No. The parsing and pricing run entirely in your browser, against a price dataset bundled into the page. Nothing you paste is sent anywhere. The same open-source parser runs inside the Goei app, so the number you see here matches what Goei shows on your first import.
What is the difference between what I pay and my usage value? +
They are two different measures and are never added together. What you pay is the flat subscription charge, $20, $100, or $200 a month. Your usage value, labeled LOCAL EST, is what that same usage would cost at list API rates. The verdict is the ratio between them, not a sum. A subscription that costs less than your usage is worth is a good deal.
Can I track this across my whole team? +
Yes. This calculator is a one-machine snapshot. To keep it live across every machine and teammate, each developer pushes their usage to Goei, a hosted dashboard that dedupes across machines and people. The quickest way is npx goei-sync (zero install, daily usage by day and model); the open-source budgetclaw CLI is the always-on version that adds per-project and per-git-branch attribution and hard spend caps. No API keys, no prompts, nothing in the request path.