Co living data retention: how co-living operators decide what door records their smart locks keep, how long they keep them, who reads them, and who confirms.

Diperbarui pada: Mei 19, 2025

Data Retention and Privacy for Smart Locks in Co-Living Operators

Data Retention and Privacy for Smart Locks in Co-Living Operators

This piece looks at co living data retention at building level, because the specification follows the door, not the sector. A retained record usually arrives in the operations meeting as a file. In one illustrative co-living operation, a resident who had already left the building wrote to ask whether the building still held a record of every time their unit door had been opened, and who had read it. The property manager could say that the lock produced records. He could not say which records, where they lived, how long they stayed, or who else could see them. He could say that an assistant had exported a list onto a laptop "in case", and nobody knew where the laptop was.

Direct answer. To put it plainly: co living data retention has to hold for co living data retention. co-living has to hold at the same time. The sections below describe the order co living data retention is settled in. A co living data retention ties every door class to a requirement. co living data retention decides which class ends up in the co living data retention specification.

  • Sort the openings — sort the door population into classes that behave alike, starting from co living data retention
  • Classify each door — state what co-living needs before anything is quoted
  • Show the fit — record the measurement that closes co living data retention
  • For co living data retention, the order below is fixed and co living is the reference point.

That question is not answered by a better lock, and it is not answered in the software menu either. Co living data retention is the name for the housekeeping that keeps a door record purposeful, readable by a defined group, and disposable on a date rather than at the convenience of whoever opens the next file.

The working rule. One record, one stated purpose, one stated period, one named reader group. Where those four answers are not written down for a door before the lock is fitted, the record is habit rather than policy, and habit is what turns a resident's question into an operator's problem.

The order the operator meets this in matters: what a door event record contains and when it identifies a person, which doors should not be producing a record at all, who sets the period, what the supplier must evidence about storage and deletion, and what happens on the morning a request arrives from a resident.

The export nobody owns

Co-living operators who inherit a site rarely inherit a records policy. They inherit a system, a set of doors, and a drawer of exported files whose names are dates. The exports are usually made for defensible reasons: a resident dispute about a parcel, a bike that went missing, a manager leaving, an insurance question.

Once a copy exists, the period that applies to it is whatever the file says about itself. A spreadsheet named for a date in spring has no expiry field. It sits in a shared drive, in a mailbox, on a laptop that has since been replaced, or in the personal account of a manager who no longer works there. Those are copies of the same question with different answers, and none was written down. The range this procurement eventually buys from is the smart door locks.

That is why co living data retention problems are rarely found in the lock. They are found in the copy, in the shared drive, and in the unanswered question from a former resident.

The lock's own event store is the least of the three places: it is bounded and overwritten as new events arrive, so it ages out without anybody deciding anything. The management system holds the long copy, and that one needs a decision. The operator's exports are the copies with no owner, and they make the question hard.

Two doors, two records, two clocks

The co-living door population holds two kinds of door inside one building. There are private unit doors, used almost exclusively by one household and expected to be private. There are shared kitchens, shared workspaces, a bike store and parcel lockers, which carry the traffic of the whole day and are expected to be convenient, and a staff or service door with a scheduled access window. The expectation attached to each is opposite, and the record each produces is different too.

A unit door record answers a question about one identified person: who opened this door, when, with what credential. A shared kitchen record answers a question about a hundred identified people over a day. The first is a small record about one resident. The second is a large record about everybody, and it exists because the building team wanted to know who came and went — a question that the resident's privacy interest and the building's operational interest both touch.

The vertical's own first question is a privacy question before it is a security question: which doors need an audit trail and which only need to open. A building that logs everything gets a record set that is large, expensive to hold, uninteresting to read, and impossible to justify to a resident who asks why the building knows they were in the laundry room at twenty-one hundred. The lock families deployed here differ in exactly the way that changes the answer: a standalone lock with a PC-linked card encoder keeps its working records on a card encoder and a laptop, a networked model with an app keeps them on a server, and on a standalone site the operator's export file may be the only record that exists.

flowchart TD
  A[Event store inside the lock] --> B[Bounded and overwritten as new events arrive]
  C[Access management system] --> D[Long copy exportable on demand]
  E[Operator export or spreadsheet] --> F[Copy with no owner and no end date]
  B --> G[Retention decision]
  D --> G
  F --> G
  G --> H[Purpose period and reader written down per door]

The privacy consequence follows the architecture. Where the record lives on a server the operator controls, a period can be enforced and evidenced. Where it lives in a file on a laptop, a period is a promise unless somebody owns the file.

What a door event record actually holds

A door event record is a row, and the row has fields. The fields below decide whether a record can be read without re-identification, and therefore how sensitive the retention decision is.

FieldWhat it carries in a co-living buildingWhat it changes in a retention decision
Credential or user referenceA card, code, fingerprint template or app account tied to a personDetermines whether the row identifies someone directly
Door identifierUnit number, shared door code, locker cellDetermines which period applies and who reads it
TimestampThe moment the credential was presentedDetermines identifiability and how long the row stays useful
Result or reason codeGranted, refused, expired window, forced, overrideSeparates routine traffic from events worth keeping

The catalogue record for a keypad and card lock built for hotels and apartments lists guest card issuance, floor and room permissions, check-in validity windows and unlock records, on a standalone lock with a PC-linked card encoder and no in-room network required. A fingerprint and keypad apartment lock in the same catalogue lists remote unlock, temporary and periodic passcodes, user management and access records. Confirm on the dated product page for the specific SKU before any of these appear in a retention schedule, because the store inside the door and the store on the server age differently.

What matters for privacy is the threshold, and the threshold is not technical. A door event log becomes personal data the moment a row can be tied to a named resident, and a row carrying only an aggregate count does not. Keeping a row longer than the door's operating purpose turns it into personal data whatever anybody has called it in a policy document, which is why the period belongs with the person who answers for data protection. A smart lock data retention policy is therefore a document about doors, not about the product.

The retention matrix that has to exist

Most co-living operators have a door schedule and a credential list. Very few have the third document, the one that answers the question a resident actually asks: for this door, this record, this period, this reader. Access log retention for co-living buildings starts here.

DoorWhat the record is forReader groupPeriod to be written downWho confirms the period
Private unit doorAnswering a dispute about entryProperty manager, escalating to the reviewerShort, with a documented incident extensionData protection reviewer
Shared kitchen and workspaceAggregate footfall for cleaningBuilding team, on a rolled-up recordRolling, short, never person-linkedProperty manager, reviewed by the reviewer
Bike storeInvestigating a missing bicycleBuilding team, then escalationIncident window onlyProperty manager
Parcel lockerProving a parcel was delivered or collectedBuilding team, with a short exception pathShort, scoped to the locker cellProperty manager
Staff and service doorConfirming that a scheduled window was usedOperations, for window complianceMatching the window lengthProperty manager
Plant and staff-only roomsSecuring a restricted spaceOperations onlyShort, and unrelated to resident recordsProperty manager

Two features of the matrix do more work than the rest of it. The reader group is named per door rather than per building, because the person who needs a bike store record on a Wednesday is not the person who needs a unit door record. And a door appears in the matrix even when the answer is that it carries no resident record: writing down "this door logs a count and nothing else" is the cheapest privacy control available, and it is available before any lock is bought.

The five questions the reviewer will ask

The period is not a decision the property manager makes alone. It is decided by the operator together with the person in the organisation who answers for data protection, and the property manager's job is to bring the door schedule, the credential design and the export habits to that conversation.

The smart lock privacy review questions are short. Who owns the access records at this site: the building owner, the operating company, or a group entity that holds the contract? On a portfolio with mixed ownership this is unsettled, and the answer decides who may even make the retention decision. What does the residency agreement say about access records, because silence leaves the operator writing the rule alone? Is the access-system provider a processor for the operator, and is that arrangement written down? Where does the hosting sit? And what is the deletion route: expiry automatic in the system, a scheduled job the operator runs, or a person with a spreadsheet?

CONFIRM-WITH-LOCAL-DATA-PRIVACY-REVIEWER — which data protection regime and guidance apply to the access records this operator holds at this site, whether the access-system provider is acting as a processor for the operator, what the reviewer requires the retention schedule to contain, and who inside the operator signs the schedule off.

Note what the question is not. It is not whether a particular lock "is compliant". Compliance is not a property of a device; it is the state of an arrangement between an operator, a system provider, a set of residents and the authority that governs them. A lock can be fitted to a door that produces records while the operator still has no documented basis for keeping them.

Setting the period: what the decision depends

The period is not chosen from a menu of sensible durations. It falls out of a small number of dependencies, and an operator can work through them before the question is put to anybody.

DependencyWhat it changesEvidence the operator should hold
Whether an incident window must stay openWhether a short base period is defensible at allThe incident procedure and who can extend a period
What residents were toldWhether the period can be shortened later without a new noticeThe notice text and the date it was issued
Identifiability of the rowWhether a person can be singled out from the recordThe field list for the door
Whether expiry is automaticWhether the period is enforced or merely intendedThe configuration and the deletion job output
Whether the supplier supports itWhether the period can be set in the system or only in practiceWritten confirmation on the dated product page

The order of these dependencies matters. Whether a record can identify somebody comes before how long it is kept, because a period chosen for an unidentifiable count can be generous while a period chosen for a person-linked row cannot. And the last row often returns an uncomfortable answer: some configurations cannot delete on a schedule, in which case the period is a human practice, and a human practice needs a named person and a monthly check.

There is a fallback that deserves to be said out loud in the room. Where the operator cannot say what a record on a given door would ever be used for, and cannot name a reader who needs it, the answer is not a long period. It is to stop producing the record on that door. Co living data retention is not a euphemism for keeping everything for a long time.

When a resident asks what is on their record

The request arrives on a Wednesday afternoon and it is ambiguous in the way requests are. "What do you have about me" can mean show me my own record, correct something in it, or delete it. Each leads somewhere different, and none can be answered by whoever happens to be on the desk that week unless the groundwork was laid in advance.

The first hour of handling a request is spent identifying scope rather than answering it. Which doors are covered. Which period is being asked about. Which of the three places the record might live holds it: the lock, the management system, or a copy somebody exported. The third breaks the routine, because an export is not indexed in any system and lives wherever the person who made it put it. Deletion request handling for co-living sites lives or dies on that third place.

Once scope is known, the answer is written by the reviewer rather than by the community manager. The manager contributes the factual part: what the site actually holds, which doors log, and where the copies went. That has to be written down in advance, because the natural human answer to "where else might this be?" is a guess, and a guess in a written response to a resident is a bad place to be.

stateDiagram-v2
  [*] --> Held
  Held : stored under a written period
  Held --> Read : a named reader opens it
  Read --> Held : no change and no finding
  Read --> Amended : the resident says it is wrong
  Amended --> Held : correction recorded
  Read --> Deleted : the period ends or a request is actioned
  Deleted : nothing further is retained about it
  Deleted --> [*]

The lifecycle above is deliberately dull. Most records should never leave the first box, and the number that matters is not how many records were read but how many were created for no reason at all — a number usually driven by the shared doors rather than the private ones.

Architecture decides which kind of site this is. Where the locks are standalone and the encoder is a laptop, the answer to a question about the co-living unit door entry record may be a file on that laptop, and the deletion question becomes a file question. Where the site is networked, the answer is a database query and the deletion question is a configuration question. That choice is covered in Choosing Between Standalone and Networked Smart Locks for Co-Living Operators, and it is a retention decision before it is a hardware decision.

What to ask the supplier and the access-system provider

The operator can write a retention policy, but on many sites the operator cannot implement it without the supplier's cooperation. These are the questions to put into the request for quotation and the contract conversation, and the evidence to demand with each one.

Question to the supplier or providerWhy it bites in this verticalEvidence to demand
What does a single event row containDetermines whether the record identifies a resident directlyA written field list for the specific model
Where is the event data hosted and does a new site change itDetermines who can be asked about a recordA named hosting location and a change-of-site procedure
Are sub-processors used for any part of the serviceDetermines the number of organisations holding the recordA list, with the notice period for changes
What export formats exist and can an export be deleted at the provider sideDetermines whether an operator export can ever be withdrawnA format list and a deletion route in writing
Can a retention period be configured per doorDetermines whether the matrix above is enforceableA configuration reference for the specific SKU
Who can read the whole event database at the operator's siteDetermines the reader group the policy must nameA role and permission description

Two of these are worth raising early. The export route decides whether a site can honour a deletion request, and it is most often answered with a format list rather than a process. The administrator log most often comes back empty, and an empty administrator log means the reader group named in the policy cannot be enforced or evidenced.

The published material in the library covers the technical side of the same ground: Smart Lock Data Retention for the storage question, and Smart Lock Data Synchronization for what happens when records move between a lock and a platform. The procurement-side conversation sits in Questions to Ask a Supplier Before Choosing Locks for Co-Living Operators, and the credential design that decides which rows identify a person is in Smart Locks for Co-Living Operators: Credential Model that Survives Rapid Onboarding and Offboarding.

The doors that should carry no record at all

The characteristic failure of this vertical is putting an audit trail on every door and paying for the friction. Privacy carries the mirror image of that failure: the cost of an unnecessary record is not the storage, it is the request it generates.

A shared kitchen door opened by dozens of people a day does not need a person-linked trail for any reason a resident would accept. What the building team wants from that door is a signal about how busy the kitchen is and when it needs cleaning, and that signal can be produced from a count, a schedule or a sensor that never needed to know who walked past. The moment the building keeps a person-linked trail on that door, it has created a record of residents' daily rhythms that nobody asked for, and an obligation to explain it.

The same reasoning applies with more force to doors where a record could never be read by anybody. A door that opens on a proximity sense in a corridor open during the hours it is open produces rows that no party would ever use, and every one of them is liability with no offsetting benefit. Resident privacy and door access records get harder to defend as the population of useless rows grows.

The opposite error is equally expensive and less visible. Under-logging the doors that matter — a unit door in a dispute, a bike store, a plant room — leaves the operator unable to answer the one question a resident or an insurer will ask. Decide per door which ones need a trail and which only need to open, and write that decision down before the hardware arrives.

What the resident is told

A retention policy that residents never see is one that will be discovered by a resident rather than by the operator. The notice does not need to be long, and it does not need to be legal in tone. It needs to say which doors are covered, what is recorded, who inside the operation may read it, how long it is kept, and how a resident asks a question or makes a request.

Timing matters more than length. The notice that works best is the one issued when the resident joins, because it sets the expectation before the first question is asked. A site that fits locks to unit doors after residents have already moved in has to treat the notice as part of the handover.

Two practical points. A notice supplied by the lock manufacturer describes the device and not the operator's arrangement, so it cannot be the resident-facing document. And a notice is only as good as the practice behind it: if the notice says a period and the practice keeps a copy indefinitely, the notice is worse than nothing, because it is now the evidence of the gap. Where practice and notice disagree, one of them is fixed before the next resident joins.

CONFIRM-WITH-LOCAL-DATA-PRIVACY-ADVISER — what a resident-facing notice must contain for this operator's access records, which parts of the arrangement are described to residents and which are kept internal, and what wording the adviser will accept on the retention period and on the route for a resident request.

The one-page review sheet

The review is not a large exercise. It happens once per site, it takes one sheet, and it is repeated whenever the door schedule changes or the access-system provider changes.

Question on the sheetEvidence the operator holdsOwnerReviewed when
Which doors log a person-linked recordThe door schedule with a mark on each doorProperty managerAt fitment and at every schedule change
What each record is forThe purpose column of the retention matrixProperty managerOnce, then on any new door
The period per doorThe matrix, signedData protection reviewerAnnually
The reader group per doorNamed roles, not job titlesProperty managerAnnually
The export inventoryA list of every export made, with a named ownerOperations managerQuarterly
The deletion routeConfiguration evidence or a written process with a named personOperations managerQuarterly
The notice residents receivedThe text and the date it was issuedProperty managerAt each tenancy year

The export inventory is the item most likely to be blank on the first pass, and the one that does the most work afterwards. Every entry in it is a file somebody created, and naming its owner is the first step towards being able to delete it.

The four answers a property manager can produce

The co-living operator who has done this work ends up with four answers per door rather than a policy document nobody reads: what the record is for, how long it stays, who may open it, and where the copies are. That is a smaller thing than a data protection programme, and it is the thing that sits between a resident's question and an operator's problem.

The related work in this cluster fills in the parts around it. Which records are worth keeping at all is the subject of Audit Trail Requirements for Co-Living Operators: What Records Matter. What the resident experiences when the record exists is in Guest, Resident and Staff Experience of Smart Locks in Co-Living Operators. Who the community manager is allowed to be is set out in Access Policy for Co-Living Operators: Community Managers Permissions Design, and what happens when a contractor needs to be inside the building for a week is covered in Contractor and Third-Party Access for Co-Living Operators.

It is worth pairing this with sl 6301 for wooden doors. That is the same question co living data retention has to answer here.The commercial step is short. Send the door schedule, the retention matrix and the standing request procedure as one document through the request a quote page, pull a physical sample to fit to the worst door in the population through order samples, and take the current model lines through request catalog so the review sheet is written against the dated record of the hardware that will actually be on the door.

Keranjang Anda
Pengalaman Anda di situs ini akan meningkatkan dengan mengizinkan cookie Kebijakan Cookie
Chat on WhatsApp