> ## Documentation Index > Fetch the complete documentation index at: https://docs.cloudthinker.io/llms.txt > Use this file to discover all available pages before exploring further. # Cost Health > Read the Cost Health score, see which FinOps checks fail on an account, and find the fixes that raise the score most Cost Health is a daily score out of 100 for one cloud account. CloudThinker checks the account against five FinOps capabilities and shows the evidence behind every point. ## Why Cost Health * **One number per account.** The score, its stage, and its trend tell you whether cost practice is getting better. * **Every point has evidence.** Each check lists the resources and spend lines that pass or fail, so you can verify the score yourself. * **Fixes come ranked.** **What raises your score** lists the changes worth the most points first. * **Unchecked spend never looks clean.** Spend CloudThinker couldn't check lowers coverage instead of raising the score. ## Open Cost Health In the sidebar, select **Optimize โ†’ Cost Health**, then pick the cloud account next to the page title. The **Cost Health** card on **Overview** links to the same page. ## The five capabilities The capabilities and their checks follow the FinOps Framework. Each check has a weight, and the weights in a capability add up to 100. | Capability | Checks (weight) | Money it flags | | - | - | - | | **Rate optimization** | Commitment coverage (40), Commitment utilization (40), Commitment expiry (20) | Committable On-Demand spend | | **Workload optimization** | Idle compute (30), Unattached storage (25), Stale snapshot (15), Previous-generation instance (15), Idle database (15) | Spend in idle or stale resources | | **Allocation** | Tagged spend share (50), Untagged resource (30), Shared cost rule (20) | Untagged spend, which isn't money lost | | **Anomaly management** | Anomaly review time (40), Explained anomaly (30), Spend alert (30) | Spend in anomaly gaps | | **Forecasting** | Forecast error (50), Budget coverage (30), Forecast horizon (20) | Spend that is forecast badly | Checks about spend lines and commitments weigh each row by its last 30 days of spend. Checks about resources, anomalies, or the account count each row once. ## How the score is set CloudThinker measures and the scoring rules score. CloudThinker gives each resource or spend line a verdict, and the score comes from those verdicts with fixed rules. CloudThinker never picks the number itself. 1. **Coverage.** CloudThinker counts how much of the in-scope spend it could check. Below 50% coverage, a capability shows **Not measured** and gets no score. 2. **Checks.** Each check scores the passing share of its measured rows. The capability score is the weighted average of its checks. 3. **Stage.** The score maps to a FinOps maturity stage. The page shows a plain label next to it. | Stage | Label | Meets | | - | - | - | | **Crawl** | **Needs work** | Measured, score under 50 | | **Walk** | **On track** | Score 50 or more | | **Run** | **Strong** | Score 80 or more, with 80% of spend measured | The top of the page shows the overall score, its stage, and **% of spend measured**. It also shows the change since the last score with the capabilities that moved it, for example **Rate optimization 49 โ†’ 61**. A first score reads **First score**, and the sparkline shows the last 30 days. ## What raises your score **What raises your score** lists the fixes worth the most points, three at a time. Select **Show n more moves** for the rest. Each card shows: * The points the overall score would gain if you fix that gap, worked out from the same rows. **Gains may overlap**, so two fixes together can be worth less than their sum. * The capability and how its score would change, for example **Rate optimization 0 โ†’ 40**. * The monthly spend involved, labelled by kind. * **Review finding** when a [finding](https://docs.cloudthinker.io/guide/cost-optimization/recommendations.md) already covers the gap, or **Ask CloudThinker** to plan the fix in a conversation. | Kind | Meaning | | - | - | | **committable** | Steady spend paid at the On-Demand rate that a commitment could cover | | **wasted** | Spend on idle, unattached, or old resources | | **no owner** | Spend nobody owns. Not money lost, but nobody can act on it | | **at risk** | Spend that is not watched, budgeted, or forecast well | Optimize never adds money across kinds, because a committable dollar and an untagged dollar mean different things. ## Read a capability Below the moves, one card per capability shows its score, its stage, and a short summary. Select a card to open its detail. Each check tile shows the points it earned out of its weight and one of four states: | State | Meaning | | - | - | | **Pass** | Every measured row passes | | **Fail** | At least one measured row fails the check | | **Unmeasured** | CloudThinker couldn't check the rows | | **Not scored** | The check has no rows on this account, with the reason | Select a tile to filter the evidence to that check. The evidence table has four tabs, 25 rows a page: | Tab | What it lists | | - | - | | **Gaps** | Rows that fail a check. This is the default | | **Passing** | Rows that pass | | **Not measured** | Rows CloudThinker couldn't check. They lower coverage, never the score | | **Out of scope** | Rows no check applies to, such as spend no commitment can cover. They count in neither the score nor coverage | Open a row to see its subject, the reason for its verdict, and its source. **Ask CloudThinker why** on the capability, or **Ask CloudThinker** on a row, opens a conversation about it. **Open finding** goes to the finding that covers it. The capability, check, tab, and page stay in the URL, so you can share the exact view. ## Before the first measurement The first full scan measures the account. Until then, the page shows the provider's own score, when the provider has one. ![Cost Health card reading No account score yet, 1 of 3 pillars measured, with Rate optimization and Resource efficiency Not set up and Cost allocation 36](https://mintcdn.com/cloudthinker/IOnNsZPtwT4C7tHY/images/cost-optimization/cost-health/01-before-measurement.webp?fit=max\&auto=format\&n=IOnNsZPtwT4C7tHY\&q=85\&s=e15039d3f452218d796019db3a98a803) Before CloudThinker measures an AWS account: the provider score, with one of three pillars measured What you may see before or between measurements: | What you see | Meaning | What to do | | - | - | - | | **No account score yet ยท 1 of 3 pillars measured** | A provider score from three pillars (**Rate optimization**, **Resource efficiency**, **Cost allocation**) while CloudThinker hasn't measured the account | Select **Run full scan**. Azure has no provider score | | A capability shows **Not measured** | Less than half of its spend could be checked | Read the reason on the capability. When a setting blocks it, a notice under the score names it, with **Open Settings** | | **n days old** on the score | The daily run hasn't refreshed the score since then. The line below says why when the last run stopped | Check [refresh health](https://docs.cloudthinker.io/guide/cost-optimization/settings.md#refresh-health) | A score that isn't ready never shows as zero. ## Download and share Select **Download PDF** in the page header to save Cost Health for the selected account as a PDF. Any workspace member can download it. To send Cost Health on a schedule, use [Optimize reports](https://docs.cloudthinker.io/guide/cost-optimization/reports.md). ## FAQ > **Why did my score drop when nothing changed?** > > The score is measured again each day. New spend, a new resource, or an expiring commitment can fail a check. The change line under the score names the capabilities that moved. > **Why is a capability Not measured on a large account?** > > CloudThinker checks the largest spend first, inside a time budget per run. When less than half of a capability's spend is checked, it gets no score. The next daily run starts where the last one stopped. > **Does fixing a gap raise the score at once?** > > No. The score changes after the next daily run measures the account again. ## Related > **[Optimize findings](https://docs.cloudthinker.io/guide/cost-optimization/recommendations.md)** > > Work the savings findings that close Cost Health gaps > **[Optimize settings](https://docs.cloudthinker.io/guide/cost-optimization/settings.md)** > > Fix the billing source or a paused refresh behind a missing score This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.