> ## Documentation Index > Fetch the complete documentation index at: https://docs.tetherdata.io/llms.txt > Use this file to discover all available pages before exploring further. # Coverage > How Tether measures and displays days of supply across the planning surfaces # Coverage Coverage — days of supply (DOS) — is how the planning surfaces express inventory health: how many days the current position lasts at the projected consumption rate. This page explains where coverage appears, how the before/after impact sections in drawers work, and the vocabulary behind the numbers. ## Coverage status Coverage status is computed server-side from your position versus the policies you author, and rendered directly — the client never re-derives it: | Status | Meaning | Grain | | ---------------- | ---------------------------------------------------------------------- | ------------- | | **No demand** | Nothing consumes this SKU here this horizon | Both | | **Out of stock** | Effectively empty, with live demand | Both | | **Low** | Below the requirement | Both | | **Healthy** | Within the requirement and the cap | Both | | **Constrained** | Above cap, forced by a downstream storage limit — carries its citation | Location only | | **Rebalance** | The network total is fine, but a location is short | Network only | | **Overstock** | Above the cap, with nothing forcing it | Both | Grids show these as tinted chips; drawers show them as colored bars. A location with **no authored policy** is not judged at all: it renders **Not set**, with no pill. A row with no status (a plan solved before the current model) reads **Awaiting re-solve** until the next solve. For what each status means, how the requirement and cap are computed from your authored layers, and a worked example across two grains, see [How runout and coverage status work](/inventory/coverage-and-runout). ## Coverage impact in drawers Recommendation drawers carry a **Coverage impact** section showing what the movement does to days of supply: * **Purchase** and **Production** recommendations show an **Across network** row on top, then per-node rows. * **Transfer** recommendations are headed **Coverage impact — source & destination**, and highlight the source and destination nodes; there is no network row. Each node row is a before/after bar: a muted segment for the current position and a delta segment for the change — green-tinted when the movement lifts coverage, red-tinted when it drops it. Up to three threshold ticks overlay the bar — **S** (Safety days), **T** (Target days), **M** (Max days) — rendered only where that layer is authored. ### The measurement date A caption under the heading names the date the before/after numbers are measured at — they are forward-looking projections at the moment the supply lands, not "today": * Purchases: **Coverage on PO Land Date** * Production: **Coverage on Run Completion Date** * Transfers: **Coverage on Arrival Date** ### What the rows do and don't show * Nodes with no coverage signal *and* no authored threshold (suppliers, off-SKU warehouses the plan touches but never stocks) are suppressed — they'd tell you nothing. * A zero-DOS node **with** an authored target is kept and rendered as a red bar: that's a managed stocking location that is out of stock. * When a recommendation bundles several SKUs, transfer and production drawers show one SKU's walk at a time — the heading says "Showing this SKU only (N SKUs on the rec)" and you click another row to switch. Purchase impact already aggregates across the PO's items. ### Special outcomes Two outcomes deliberately render without a number: * **Beyond the horizon** — the position outlives the planning horizon at its (very slow) depletion rate. Rendered as a fully-covered bar, not "0d". * **N/A — no demand signal** — nothing consumes this SKU from this point forward in this scenario, so coverage is *unmeasurable* — not zero, and not the same as overstocked. ## Coverage per lens Coverage is lens-scoped. Each [scenario lens](/supply-planning/overview) is walked end to end — supply *and* demand: "Do nothing" and "Follow my plan" drop the supply their decision set excludes and re-derive the consumption that supply would have driven. They are true counterfactuals, not the full-plan demand shape with supply removed. Consequently the same cell can legitimately show different DOS values — and different chip tints — under different lenses, and a chip never falls back to another lens's numbers. If a value can't be measured under the active lens, it says so rather than borrowing one. ## Coverage glossary The same word — "coverage" — is rendered at more than one grain, and the grains are not interchangeable: | Term | Definition | | ----------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | **Node coverage** | One location's sellable on-hand runway for the current period. Excludes goods in transit and in production. | | **Network coverage** | The whole network's sellable position, including inventory in transit between your own nodes, sampled per month. | | **Worst status across nodes** | A status aggregation, not a coverage value — a period is shaded by the worst coverage status across the SKU's nodes, so one out-of-stock node shows red even when the others are healthy. | Network coverage is not the sum of node coverage, and a status aggregation is never a DOS value. When two surfaces disagree on "coverage" for the same SKU, check which grain each one labels — the difference is usually by construction, not an error. ## Best practices The before/after delta segment is the decision signal: it shows the direction and magnitude of a movement's effect regardless of where the absolute numbers sit on the scale. "Coverage on PO Land Date" measured months out can look healthy while the intervening weeks stock out. Pair the drawer's impact section with the Worksheet's period-by-period view for the full path. An item with stock but nothing consuming it reports N/A, not a runway. Investigate why demand is absent (channel gating, forecast gaps) before writing the stock off — see [Data issues](/supply-planning/data-issues). Threshold ticks and status tints only appear where safety, target, or overstock thresholds are authored. Nodes without thresholds fall back to defaults and get suppressed when idle — set thresholds at every location you actively manage. ## Troubleshooting | Issue | Cause | Solution | | ------------------------------------------------ | ----------------------------------------------------------------------------------------- | ------------------------------------------------------------------ | | A drawer shows no coverage section | No coverage impact was authored for that movement (common for synthesized committed rows) | Check the SKU's coverage on the grid or Worksheet instead | | An expected node is missing from the impact rows | It has no coverage signal and no authored threshold | Author a coverage threshold at that node if you manage stock there | | Two pages show different DOS for one SKU | Different grains (node vs network) or different lenses | Check the grain label and the active lens on each page | | A stocked node reads "no demand signal" | Nothing consumes the SKU from that point forward under the active lens | Verify the forecast and channel setup for that SKU/location | | Chips are gray with no tint | The plan predates status computation, or no plan is solved | Run a solve; statuses populate from the new plan | ## Next Steps The scenario lenses that scope every coverage number The drawers where coverage impact drives decisions Current-period inventory health by location The consumption projections coverage is measured against