How do I track my Sentry spend in one place?

Connect Sentry to CostCompass and it turns your error and span volume into a running month-to-date cost, a per-day trend, and a next-month forecast — Sentry sitting on the same dashboard as every other provider you pay for. You paste one token, pick your plan, and click Refresh; from there CostCompass reads your daily error and span counts, prices them against your plan, and keeps a running total you can read the moment you look, instead of opening Sentry's own console to find one number that stops at Sentry.

Which Sentry token do I need, and why won't my normal auth token work?

CostCompass needs an Internal Integration token with the org:read permission, created under Settings then Developer Settings then Custom Integrations. An ordinary Organization Auth Token (the kind that starts with sntrys_) won't work — those have a fixed set of CI-oriented permissions that can't be edited to add org:read, while an Internal Integration lets you grant exactly that one read scope and nothing else. So create the Internal Integration, give it org:read, and paste its token. That's the credential that can read your organization's usage.

Where do I find my Sentry organization slug?

It's the short name in your Sentry URL — the part after organizations in a link like sentry.io/organizations/your-org/. You enter that slug when you connect so CostCompass reads the right organization; it's validated when you connect, so a wrong one is caught right away rather than showing an empty dashboard. Your Sentry region is worked out for you, so the slug and your plan are the only two details you supply alongside the token.

Why did my Sentry span bill jump when my usage didn't change?

Usually because your plan's included span allowance changed, not your traffic. Sentry halved the span allowance on the Team and Business plans in 2025 — from 10 million to 5 million a month — so a project sending 8 million spans went from fully covered to paying overage on 3 million, with no change on its side at all. Each plan carries its own allowance and its own overage rate, so the same volume can cost differently as those numbers move. That's exactly the kind of shift a running per-day trend surfaces early, while there's still month left to react.

How close is the Sentry cost CostCompass shows?

It's a close estimate, and CostCompass marks it as one. Your error counts are exact, because Sentry never samples errors. Your span counts reflect the spans Sentry stored, which on a project using trace sampling can sit below the volume Sentry actually bills — so spans can read a little low. The estimate also can't see a reserved-volume discount you prepaid or a pay-as-you-go budget cap, both of which move your real rate. Within those bounds it's a working number you can act on as the month runs, not a reconstruction of your settled invoice.

Which parts of my Sentry bill does CostCompass track?

It tracks the categories that drive most developers' Sentry cost — errors and spans (Sentry's tracing metric) — plus attachments, logs, metrics, and profiling, each priced against your plan's included allowance and overage rate. Replays and cron and uptime monitors aren't tracked yet, so treat the figure as covering those priced categories rather than your entire Sentry bill. Everything it tracks, it measures the way Sentry's own tooling meters it, and it's honest about where that scope ends.

Does CostCompass store my Sentry token?

Not in any form it can read. Your token is encrypted in your browser with your vault password before it ever leaves your device, and what CostCompass keeps at rest is only the resulting ciphertext — an opaque blob it can't open, because your vault password never reaches it. When it's time to read your usage, the token is unsealed in your browser and forwarded to Sentry through a relay that holds it for the length of that one call and is built not to log it. So the plaintext passes through for a single request but is never written to the database or the logs.

Why use CostCompass instead of Sentry's own billing page?

Sentry's billing page shows Sentry's spend, for Sentry alone, in Sentry's console — useful, but it stops at Sentry and you read it apart from everything else you pay for. It won't tell you how Sentry sits next to Claude, your GPUs, or your hosting, or where your whole stack lands this month. CostCompass prices the same error and span usage and rolls Sentry up with every other provider into one running total and one forecast — current the moment you click Refresh, instead of a per-tool number you go read in each console.

Sentry cost tracking

See your Sentry spend as a running cost, next to everything else

Sentry meters your errors and spans and totals the cost in its own console, apart from the rest of your stack. Here's how CostCompass prices that same usage against your plan into a live running total and a next-month forecast — Sentry folded in with every other provider you run.

See your own Sentry spend — start free

By Joubert Berger Published July 14, 2026

An antique almanac engraving: a watchman's tally desk where two streams of small marks — one of jagged fault-symbols, one of fine thread-like traces — pour in from the left and pass under a plated rate-comb of stepped teeth; below the comb a single copper dial gathers both streams into one running figure. — Two streams — faults and traces — pass under the stepped rate-comb of your plan, and the copper dial below gathers them into the one running figure.

Sentry watches your software for a living: every error it catches, every trace it records, tells you something about what you shipped. It bills for that the way it should — by how much you send it. What it doesn't give you is the same thing every other tool in your stack withholds: a single, current number for what all of it costs together, this month, while you can still do something about it. Sentry prices Sentry, in its own console. The rest of your bill lands somewhere else.

Sentry already keeps the counts underneath — errors and spans, day by day. The work left is pricing them into a running cost and a forecast, and setting Sentry next to everything else you run.

What does CostCompass show you for Sentry?

Connect Sentry and it becomes a live line in your spend, not a number you go fetch. CostCompass reads your daily error and span counts, prices them against your plan, and gives you three things Sentry’s own console keeps in three different places or not at all: a running month-to-date cost for Sentry, a per-day trend so a climb is visible while there’s still month left to act, and a next-month forecast from your recent burn rate. All of it sits on one dashboard beside every other provider you’ve connected, with Sentry folded into the whole-stack total.

The CostCompass dashboard showing month-to-date spend across providers with a forecast and burn rate, Sentry among them. — Month-to-date spend across every connected provider, with a forecast and burn rate, rolled up in one view — Sentry folded in.

The number is current the moment you click Refresh, not on some schedule behind your back. You look, CostCompass reads your latest counts and reprices them, and the running total moves to match.

How is that number built?

Sentry meters usage rather than handing back a cost, so CostCompass prices the usage. It counts your errors and spans the way Sentry’s own Issues and Discover views count them, then applies your plan’s schedule: the monthly allowance each category includes for free, and the overage rate past it. Count the units, price them by plan, sum to a running total. That’s the whole model.

A flow diagram: two labeled streams, errors and spans, entering a stepped schedule marked free allowance then overage, converging into a running-total meter with a forecast arrow, and a faint boxed region to the side labeled outside view holding a reserved-volume discount and a pay-as-you-go cap. — Error and span counts pass through your plan's allowance-and-overage schedule into one running cost and a forecast; the reserved discount and pay-as-you-go cap sit outside what the estimate can see.

Because the figure is built from your own plan’s allowances and rates, it reflects your Sentry, not a flat per-unit guess.

Connecting Sentry: the token to create

Sentry keeps usage behind a credential you create once. The working setup is an Internal Integration token with a single read permission:

In Sentry, go to Settings → Developer Settings → Custom Integrations and create a new Internal Integration.
Give it the org:read permission — that one read scope, nothing more — and copy the token it generates.
In CostCompass, paste the token, enter your organization slug (the name after organizations/ in your Sentry URL), and pick your plan.

One practical warning saves a detour: the ordinary Organization Auth Token (the kind that starts with sntrys_) won’t work here. The Internal Integration exists precisely to grant a custom scope like org:read, which is why it’s the one to create. Your Sentry region is discovered for you, so the token, the slug, and the plan are all you supply.

Why your plan changes the cost of the same usage

On Sentry, identical usage can bill differently depending on which plan you’re on, because each plan carries its own included allowance and its own overage rate. The paid plans include a monthly allowance of errors and spans — on the order of tens of thousands of errors and a few million spans — and meter anything past it; the free Developer plan is capped rather than billed, so it prices at zero. That’s why CostCompass asks for your plan: it prices your counts against your plan’s quotas, and the free-allowance breakdown shows only the allowances that are actually yours.

This isn’t hypothetical. Sentry halved the span allowance on the Team and Business plans in 2025 — from 10 million to 5 million a month — so a project sending 8 million spans went from fully covered to paying overage, with nothing changed on its side. Allowances move, and when they do the same volume costs more. A running per-day trend is what catches that early.

A horizontal bar chart of month-to-date Sentry spend split into two priced lines, errors and spans, each turned into money from its daily count, the whole marked estimated. — Month-to-date Sentry spend split by category — errors and spans, each priced into money from the counts Sentry reports per day.

Sentry is a subscription, not prepaid credit, so there’s no balance to show — just spend. On the paid plans, the flat monthly plan fee is part of what you owe, and CostCompass folds it into your running total alongside the metered usage; the free Developer plan has no fee.

How close is the number?

Close, and marked as an estimate rather than passed off as your invoice. Two things are worth knowing so you can trust it for what it is:

Errors are exact. Sentry never samples errors, so the error count — and the cost built from it — matches what Sentry counts.
Spans can read a little low. The span figure reflects the spans Sentry stored, and a project using trace sampling stores fewer spans than it’s billed for, so on a sampled project spans can sit under the billed volume.

Beyond the counts, the estimate can’t see two things that move your real rate: a reserved-volume discount you prepaid, and a pay-as-you-go budget cap. Both change the marginal price Sentry charges, and neither is exposed to read. Within those bounds the number is a working estimate you can act on as the month runs — which is exactly what a forecast needs.

How do you forecast next month’s Sentry bill?

Forecasting usage-based spend needs nothing exotic, just a burn rate: take your cost over the last several days, average it per day, and project it across a full month. CostCompass scales your trailing seven-day burn rate to the length of next month, giving one forward number for what Sentry costs if the current pace holds. Seven days is long enough that one noisy day doesn’t skew it, short enough to pick up a recent change like a release that started emitting far more spans. Because the same engine runs across every provider you’ve connected, Sentry’s forecast is part of one whole-stack projection.

How does CostCompass track your Sentry costs?

CostCompass reads your organization’s daily error and span counts through the credential you connected, prices them against your plan’s allowances and overage rates, and records the result as an estimate. On the first connect it pulls a recent window of usage rather than your whole history, and each refresh brings the running total current — the moment you click Refresh, not on some schedule.

Two things make that practical for a solo developer. First, your token is encrypted in your browser before anything is stored. The token you paste is sealed with your vault password on your own device, and only the sealed version is ever saved — a blob CostCompass can’t open, because the vault password stays with you. When it’s time to read your usage, your browser unseals the token and passes it to a relay that uses it for that one call to Sentry and is built not to log or keep it. What’s held at rest is locked ciphertext; the credential your organization runs on never reaches us in usable form.

Second, Sentry doesn’t sit alone. The same dashboard rolls its running cost up with every other AI and compute provider you’ve connected — Claude, your GPUs, your hosting, other usage-metered tools like PostHog and ElevenLabs — into one figure with a forecast. How you get to that cross-provider number — by hand, through tooling, or read straight from each usage API — is its own decision; the ways to track AI costs across providers lay the options out side by side.

	Sentry’s billing page	CostCompass
Scope	Sentry spend alone	Sentry beside every other provider
Forecast	Projects Sentry’s spend only	One forecast across your whole stack
Lives in	Sentry’s own console	One dashboard with all your spend
Currency	Updates on Sentry’s billing cycle	Current the moment you click Refresh

Getting started takes three steps:

In Sentry, create an Internal Integration (Settings → Developer Settings → Custom Integrations) with the org:read permission, and copy its token.
Paste it into CostCompass with your organization slug and plan. It’s encrypted in your browser before it’s stored, so the server only ever holds ciphertext.
Click Refresh. CostCompass reads your daily error and span counts, prices them against your plan, and from there your running cost, forecast, and category split roll up — Sentry folded into the whole-stack number.

Frequently asked questions

How do I track my Sentry spend in one place?: Connect Sentry to CostCompass and it turns your error and span volume into a running month-to-date cost, a per-day trend, and a next-month forecast — Sentry sitting on the same dashboard as every other provider you pay for. You paste one token, pick your plan, and click Refresh; from there CostCompass reads your daily error and span counts, prices them against your plan, and keeps a running total you can read the moment you look, instead of opening Sentry's own console to find one number that stops at Sentry.
Which Sentry token do I need, and why won't my normal auth token work?: CostCompass needs an Internal Integration token with the org:read permission, created under Settings then Developer Settings then Custom Integrations. An ordinary Organization Auth Token (the kind that starts with sntrys_) won't work — those have a fixed set of CI-oriented permissions that can't be edited to add org:read, while an Internal Integration lets you grant exactly that one read scope and nothing else. So create the Internal Integration, give it org:read, and paste its token. That's the credential that can read your organization's usage.
Where do I find my Sentry organization slug?: It's the short name in your Sentry URL — the part after organizations in a link like sentry.io/organizations/your-org/. You enter that slug when you connect so CostCompass reads the right organization; it's validated when you connect, so a wrong one is caught right away rather than showing an empty dashboard. Your Sentry region is worked out for you, so the slug and your plan are the only two details you supply alongside the token.
Why did my Sentry span bill jump when my usage didn't change?: Usually because your plan's included span allowance changed, not your traffic. Sentry halved the span allowance on the Team and Business plans in 2025 — from 10 million to 5 million a month — so a project sending 8 million spans went from fully covered to paying overage on 3 million, with no change on its side at all. Each plan carries its own allowance and its own overage rate, so the same volume can cost differently as those numbers move. That's exactly the kind of shift a running per-day trend surfaces early, while there's still month left to react.
How close is the Sentry cost CostCompass shows?: It's a close estimate, and CostCompass marks it as one. Your error counts are exact, because Sentry never samples errors. Your span counts reflect the spans Sentry stored, which on a project using trace sampling can sit below the volume Sentry actually bills — so spans can read a little low. The estimate also can't see a reserved-volume discount you prepaid or a pay-as-you-go budget cap, both of which move your real rate. Within those bounds it's a working number you can act on as the month runs, not a reconstruction of your settled invoice.
Which parts of my Sentry bill does CostCompass track?: It tracks the categories that drive most developers' Sentry cost — errors and spans (Sentry's tracing metric) — plus attachments, logs, metrics, and profiling, each priced against your plan's included allowance and overage rate. Replays and cron and uptime monitors aren't tracked yet, so treat the figure as covering those priced categories rather than your entire Sentry bill. Everything it tracks, it measures the way Sentry's own tooling meters it, and it's honest about where that scope ends.
Does CostCompass store my Sentry token?: Not in any form it can read. Your token is encrypted in your browser with your vault password before it ever leaves your device, and what CostCompass keeps at rest is only the resulting ciphertext — an opaque blob it can't open, because your vault password never reaches it. When it's time to read your usage, the token is unsealed in your browser and forwarded to Sentry through a relay that holds it for the length of that one call and is built not to log it. So the plaintext passes through for a single request but is never written to the database or the logs.
Why use CostCompass instead of Sentry's own billing page?: Sentry's billing page shows Sentry's spend, for Sentry alone, in Sentry's console — useful, but it stops at Sentry and you read it apart from everything else you pay for. It won't tell you how Sentry sits next to Claude, your GPUs, or your hosting, or where your whole stack lands this month. CostCompass prices the same error and span usage and rolls Sentry up with every other provider into one running total and one forecast — current the moment you click Refresh, instead of a per-tool number you go read in each console.

About the author

Joubert Berger builds CostCompass, a spend-intelligence dashboard that pulls usage from AI and compute providers into one month-to-date total, a forecast, and a per-provider breakdown. This guide reflects how CostCompass reads each provider's own usage API — see the security model for how your keys are handled.

Stop opening the Sentry console to find one number

Connect Sentry once and turn its error and span volume into a live running cost, a per-day trend, and a forecast — rolled up with every other provider, in one click.

Start free