Kora is a chain-of-custody ledger. Every handover is counted and signed by both parties — so a shortfall stops being a mystery and becomes a line item with a date, a leg, and two names on it.
A few percent of shrinkage a month is survivable as a cost of business. What isn't survivable is that it can't be priced, can't be recovered under contract, and can't be reduced — because nobody can say which hop it happened on. Kora doesn't stop linen going missing. It makes every disappearance addressable.
Everything else — dashboards, statements, vendor scorecards — is derived from that one repeated act. Stock levels are a view, not a number anyone types in.
Soiled in, clean out. Steppers on a phone, under thirty seconds, works with no signal at all.
The receiver confirms with a one-time code on their own phone, a badge tap, or a PIN. Each method is recorded with its own strength.
Append-only. A wrong entry is cancelled by a reversing entry that points at the original. Both survive forever.
A reconciliation statement per site, built only from attested movements, ready for both parties to countersign.
Ward to depot, depot to laundry, laundry back. Each is its own leg with its own counts, so a shortfall lands somewhere specific instead of in a single column called "missing".
Linen returned still stained goes to rewash, never into clean stock. Infected linen is segregated with its own rate. Your depot never starves while a scorecard looks perfect.
Consignments age. An unreturned lot at eight days is visible, valued, and chaseable — long before month-end makes it somebody's word against somebody else's.
Declare a surge and attestation relaxes to sign-later, reserve stock unlocks, and every movement in the window is flagged. None of it closes until the signatures catch up.
See every site by days of cover rather than raw counts. When one runs short, Kora ranks your options — including the one people forget, which is that the wash is already done.
Issued, returned, condemned, unaccounted — per site, per month, from signed movements only. The document that turns a dispute into a debit note.
The depot supervisor's app does the work. Everyone else either countersigns or reads. Every extra feature on a housekeeper's phone is a reason to uninstall it.
A code sent to the receiver's own phone proves possession of a device you don't control. A PIN on your supervisor's handset proves only knowledge. Both are allowed — but the record says which one you got, and so does the monthly statement.
Phone left in the locker, dead battery, no signal, new starter. The handover still happens and still gets recorded — with a reason code attached. Blocking it just teaches people to work around the app.
The fallback path needs a reason before it proceeds. If it were quicker than the strong path, everyone would take it and the distinction would be decorative.
Your statement reads "July: 94% strong, 6% adequate, 0 provisional." That's a number a site can be held to and improved — without ever blocking a single handover.
No blame field, no automatic penalties, no recommended action. Kora produces evidence; you and your client settle disputes under your contract. The moment a tool assigns blame, both sides start gaming it.
There is no "adjust quantity" button anywhere. A balance is the sum of signed movements. Corrections happen through a counted stocktake with a reason and an approver — never a silent fix.
Your laundry sees the scorecard you see. A floor sees its own position. No party gets a report the other can't — asymmetry is what turns a shared record into somebody's weapon.
No targets, no scores, no rankings, no location tracking on housekeeping staff. Kora tracks assets. Point it at people and you'll lose the site's cooperation, and with it your data.
Nobody approves their own condemnation, at any site, in any configuration, during any emergency. Unwatched write-off is the drain that hides everything else.
Sites see stockouts, consumption per bed-day, and their own exposure. If the app only serves you, the people doing the counting have no reason to count honestly.
Nobody abandons the book on day one, and they shouldn't. The gap between the two is the proof — and it's usually the most interesting number a client has seen in years.
Opening balances are counted together and signed by both parties. Most sites have never had an accurate number — this alone tends to land.
Ward exchanges and depot movements recorded in both. Divergence is measured daily rather than argued about monthly.
Dispatch and receipt at the dock, partial returns, rewash, discrepancies raised to both sides at once.
Each site is configured, not built — floor stores or counter-only, hotel or hospital, whichever modules apply.
That's the real risk, and we treat it as one. The exchange screen is built to finish in under thirty seconds and to work with no signal — if it's slower than shouting at the supervisor, it gets abandoned, and no amount of dashboard quality saves it. We test that screen with actual staff before anything else gets built. The staff-facing app also has no targets, no scores and no comparisons, deliberately, because a tool that feels like surveillance gets fed dishonest counts.
They get a signed link that does the three things that matter at the dock — confirm pickup, confirm delivery, dispute a count — with no account and no installation. If they'd rather have the app, it's two taps and works offline, which is better at 7am in a basement bay. Either way, the vendor also gets their own scorecard: the same numbers you see, at the same time. That reciprocity is usually what gets them to participate.
They will, within about six weeks. Kora states facts — a floor is thirty-one pieces net negative, a vendor returns 96.8% against a 99.4% benchmark — and stops there. It doesn't accuse, recommend or penalise. We'd strongly suggest agreeing the response protocol with your client before go-live, because the first uncomfortable finding is where systems like this either get acted on or get dismissed as broken.
No. Kora tracks item type and count, which is what the paper register already does — it just makes each handover attested. Per-item tagging is a later phase and slots into a field that's already in the data model, so counting becomes scanning without changing a single screen or report. The usual trigger for it isn't technology, it's a laundry serving two clients whose loads get mixed.
Depends entirely on how your site actually runs, and it's the first thing we ask. If the depot is unstaffed overnight, there are workable answers — a night custodian with their own credential, sign-later with a morning queue, or a small buffer held on the ward. What doesn't work is people sharing one code, because a shared credential makes every signature worthless. So we settle this before the pilot, not after.
Set out in writing before rollout: who owns consumption data, whether you can benchmark it across clients, and what happens on termination. Staff personal data is handled under the DPDP Act with consent language, a stated retention period, and a deletion route that removes the person while preserving the attestation — the record has to stay intact or the whole thing is pointless.
A walkthrough takes forty minutes and covers your actual sites — how many depots, who staffs them, what your laundry contract says. You'll leave knowing whether this fits before anyone builds anything.