Uptime and MTTR KPIs Worksheet: What to Write Down and Why
Every portfolio has that quarterly slide. It carries one number, produced by somebody's spreadsheet, describing how well the locks behaved, and in the meeting the number is either congratulated or attacked. Almost nobody in the room can answer the question that actually matters: which doors does it come from, what counted as a fault, when did each clock start and stop, and what was left out. The uptime and MTTR KPIs worksheet exists to make that question answerable before the slide is drawn, not after. It is a small grid of columns — a register of doors, a report line, an operable-again line, a few fields about what was done — and its purpose is narrow: to fix down, in advance, what you will write when something goes wrong at a door, and to tell you what each column will silently swallow.
Direct answer. To put it plainly: uptime and mttr kpis worksheet has to hold for uptime and mttr kpis worksheet. uptime-and-mttr-kpis has to hold at the same time. The sections below describe the order uptime and mttr kpis worksheet is settled in. An uptime and mttr kpis worksheet sets out the maintenance record for a system in service. uptime and mttr kpis worksheet carries the early-warning signal for uptime and mttr kpis worksheet.
- Set the interval — define how often uptime and mttr kpis worksheet is inspected
- Set the trigger — name the condition that forces a change to uptime-and-mttr-kpis
- Track the number — keep one metric that shows column one is healthy
- For uptime and mttr kpis worksheet, the order below is fixed and uptime and mttr kpis is the reference point.
This is a narrative essay about that grid, following one facility manager, Priya, through the quarter in which she stopped reporting a number and started reporting a sheet. The building, the estate and the people are a composite illustration written for this article, not a case study of any real site. The definitions the worksheet rests on are the ones the operations world argues about; the numbers it produces belong entirely to the site that fills it in. Nothing here is a target, and nothing here is supplied as a fact about any product: the plan sets the targets, and the worksheet is how you supply the measurement.
Why the Blank Sheet Comes Before the Number
Priya inherited the problem in her second month. The portfolio ran maybe a hundred and fifty doors of smart hardware across three buildings — card readers on the hotel wing, keypads and fingerprint panels on the office side, cabinet locks in the spa and the back-of-house — and the previous operator had been reporting an uptime percentage every quarter. When she asked for the underlying records, she got the percentage and a shrug. There was no list of which doors had been counted, no definition of what made a door unavailable, no timestamps anyone could point at. The number was unfalsifiable, which meant it could be argued with but never corrected.
She could have bought that behaviour from a dashboard. Every management system produces some figure, and a figure with no definition behind it is the most expensive thing a facility manager can carry into a review meeting, because it trains the audience to distrust everything else she says. The alternative was not a better tool. It was a blank sheet: the uptime and MTTR KPIs worksheet, agreed line by line before the next fault, so that when the next fault came, writing it down was a routine and not a decision.
The order of work matters more than people expect. Most teams try to define the measure first — write down what availability means, what MTTR means, agree the formulas — and then go and collect. That order fails, because the definitions that survive contact with a real Tuesday afternoon are the ones constrained by what the technician can actually observe at a door. Start with the columns. Ask what a person standing at a misbehaving mortise lock can honestly write, in the ten minutes they have before they must make a decision, and let the measures be built out of those honest entries. A definition that requires information no one can record is not a strict definition. It is a fictional one. The product family this applies to is the smart door locks.
So the worksheet is written before the first entry, and it is written in a specific sequence: population first, then the report line, then the two timestamps, then the evidence fields. Each of those is a column group, and the rest of this essay walks through them one by one — what each column answers, and what it quietly swallows. The omissions are the reason the article exists. A KPI without a stated blind spot is how portfolios end up congratulating themselves on a number that only counts the faults someone remembered to report.
Column One: The Population of Doors
The first column group is the least glamorous and the one with the largest consequences: which doors are in scope. Availability is operable time divided by scheduled time over a stated population and period, and the phrase "stated population" is doing most of the work in that definition. Priya's first worksheet draft covered all three buildings; her second, after a month of argument, separated them, because the hotel wing runs a different scheduled day than the offices and mixing them produced a figure that described neither.
Write down, for each door: its object identifier from the door or unit schedule — not the room number, because room numbers get reassigned during refurbishments and the schedule survives — the model and variant as taken from the label or the catalogue row, the mount date if it is known, and the permission group the door belongs to. Model and variant details for any specific unit should be confirmed on the dated product page for the specific SKU rather than recalled from a purchase order, because the same family ships in more than one power arrangement and more than one reader layout, and those differences change what a technician can observe at the door.
There is a reason the register is per door and not per lock. A fault can live in the battery or cell compartment, in the motor and gearbox, in the latch or bolt drive, in the control PCB, in the reader, in the credential, or in the door itself — a closer set too strong, a rebate that has moved, a latch preloaded against the strike. The worksheet tracks doors because doors are what the building notices; the hardware inventory is the second sheet, not the first.
What this column swallows: any door you did not list. A population defined once in a spreadsheet and never reconciled against the estate drifts. New installations appear in the capital plan but not in the worksheet; a block taken out of service stays in the worksheet and stops having faults worth reporting, which — and this is the trap — improves the average while the building gets worse. Put a reconciliation date on the population column and hold to it, the same way you would hold a key-issuing list.
A practical test for whether the register is honest: pick any fault from the last month and ask whether the door was in the population on the day the fault happened, and whether it is still in the population today. If the answer takes more than a few seconds to look up, column one is doing its job badly, and everything downstream is carrying the damage.
Column Two: What Counts as a Reported Fault
The report line is where most worksheets quietly die, because it looks obvious. You write down when the fault was reported. But before the timestamp can mean anything, three prior decisions have to be written into the column's header: what counts as a fault, through which channel a report becomes a report, and what happens to reports that arrive anywhere else.
Priya ran into all three inside six weeks. A guest at the hotel wing complained to the front desk on Monday that the card had to be presented three times; the front desk mentioned it to an engineer at lunch on Wednesday; the engineer looked at the reader on Thursday. Three different candidate start times, all defensible, and whichever one the team picked changed the resolution interval by days. Her fix was not cleverer clockwork. It was an agreed channel — the ticket queue, and only the ticket queue — with the front desk trained that a mention over lunch is not a report line, however useful the mention turns out to be.
Write the channel into the worksheet header and hold the discipline that a fault enters the sheet when it enters the channel. Everything else — corridor conversations, app notifications nobody raised a ticket for, a contractor noticing something while doing other work — is intelligence, and intelligence belongs in a different column, or nothing prevents the sheet becoming a rumour ledger. The measure built on the report line then has an honest name: it measures faults reported through the channel, not faults that happened.
That distinction is not pedantry; it is the largest single silent entry in any uptime and MTTR KPIs worksheet. The unreported fault does not just miss the mean time to repair. It removes that door's downtime from the availability numerator too. A portfolio where occupants have learned that complaining about a keypad accomplishes nothing will show, quarter after quarter, a beautifully healthy number. The sheet cannot fix that on its own, but a cheap reconciliation can: ask the card encoder and event log people how many doors show repeated failed presentations with no matching ticket, and treat the gap as a reporting-culture finding. Records of what the lock itself saw are covered in RFID Hotel Lock Audit Trail and Event Records.
One more entry decision belongs in this column: alerts versus faults. A low-battery alert is a threshold crossing, not a measurement of the cells, and treating every alert as a fault will inflate the sheet; treating none of them as faults guarantees a dead door later. The honest middle is to log alerts as their own row type with a different start rule — scheduled replacement versus unscheduled repair — and never let the two row types share an average.
Column Three: The Two Timestamps and the Word Operable
Every interval on the sheet is built from two stamps: the date and time the fault was reported through the agreed channel, and the date and time the door was operable again. The first timestamp inherits whatever discipline column two has. The second one hides a definition question that most teams never answer until an argument forces them to.
What does operable mean? Priya found out the hard way on a corridor door at the office block. The battery pack was flat, a technician arrived within the hour, and the door was opened with the mechanical override key. By the end of the visit the door physically worked — for anyone holding the override. The occupant did not hold the override. The replacement cells arrived with the next parts run, days later. So when was that door operable again: at the visit, or at the cell swap? Both answers are defensible, and the difference between them is exactly the sort of thing MTTR debates are really about.
Write the answer into the worksheet before you need it, and write it as the building's answer, not the supplier's. A door is operable when it can be used normally, by the people who normally use it, through the normal entry paths — the card, the code, the fingerprint, the app, whatever that door's population actually uses — not when it can be forced open with a tool held by three staff. If you want a partial-credit state, give it its own column: door openable but on reduced entry paths, or openable only by override. Collapsing those into one binary is what produces the quarterly slide nobody trusts.
There is a related subtlety on the report side too. When a fault is found without being reported — a technician spots a cracked escutcheon during another visit, a reader fails its self-test at commissioning — what is the start stamp? The clean rule is the discovery timestamp, with the row flagged as found-not-reported, so the row can be excluded from the reported-fault measure without being excluded from the record. The sheet stays honest because the flag is visible, not because someone remembered at the quarter's end.
The other thing column three swallows: anything that happened before the channel opened. A door that has been intermittent for six weeks and finally dies on the day it is reported has one interval on the sheet and a much longer history in the building. That is acceptable — the sheet measures the reported and the recorded, not the true — but the quarter's narrative has to say so out loud, and the place where that admission lives is the next column group: the evidence fields. Keep the timestamps minimal and factual; keep the interpretation in the columns that follow them.
The Route a Fault Takes to Become a Cell
With the three column groups fixed, the worksheet's logic can be drawn as one path. The value of drawing it is that every branch point is a decision someone will otherwise make differently on a wet Tuesday, under pressure, with an occupant waiting.
flowchart TD
A["Something goes wrong at a door"] --> B{"Does it reach the agreed channel?"}
B -->|Yes| C["Report stamp written - date and time"]
B -->|No| U["Fault lives only in building memory"]
U --> X["Door is absent from both measures"]
C --> D["Attendant acts - checks logged in order"]
D --> E{"Entry paths usable normally again?"}
E -->|No, parts or unit needed| W["Reduced-service state flagged, clock keeps running"]
W --> D
E -->|Yes| F["Operable-again stamp written"]
F --> G["Interval computed for the MTTR column"]
F --> H["Outage hours computed for the availability column"]
X --> J["Mean shortens while the estate disagrees"]
G --> K["Quarter review reads the sheet, not the slide"]
H --> K
Read the diagram as the worksheet's constitution. The two exits on the left — the unreported fault and the reduced-service state — are where numbers go to be wrong in opposite directions. The unreported fault makes everything look better; the undefined operable state makes intervals look shorter than the building experienced them. A team that has genuinely agreed on this path can answer, in a review, not just what the number is but which doors generated it and which doors were never asked to generate anything.
After the diagram, the arithmetic itself is the least interesting part — and the most common place for quiet errors — so it deserves its own short treatment.
The Arithmetic, Column by Column
MTTR is the mean of the resolution intervals across a defined period and a defined population. Availability is operable time divided by scheduled time over the same two definitions. Both sentences are short; the discipline is in the words mean, period and population, and in the subtraction. Each interval is the report stamp subtracted from the operable stamp, in the same clock, and the sheet should compute intervals from the raw stamps rather than from someone's mental note of roughly how long it took.
Here is the arithmetic illustrated on a fragment of Priya's sheet — a composite example, no real building, substitute your own records:
| Door ID | Reported | Operable again | Interval | Row type | Notes column, abbreviated |
|---|---|---|---|---|---|
| B2-014 | Tue 09:40 | Tue 14:15 | 4 h 35 m | reported | cells swapped, reader fine |
| B1-207 | Wed 21:05 | Sun 11:30 | 2 d 14 h | reported | waiting on retained-unit check |
| C3-002 | Fri 07:50 | Fri 08:20 | 0 h 30 m | found-not-reported | flagged at commissioning walk |
| B2-119 | Mon 06:15 | Thu 19:00 | 3 d 12 h | reported | drive part not in the local pool |
Four rows, two different means depending on which rows you let in. Including the found row (C3-002) lowers the mean; excluding the long waits on parts (B1-207, B2-119) lowers it much further and hides the portfolio's real problem, which is not diagnosis speed but holding stock. The availability side works the same way: scheduled time is what the building expects the door to serve — 24 hours for the hotel wing, weekday business hours for a back-of-house store room, whatever your rule is, written down — and the outage hours come from the same intervals, only converted into door-hours rather than fault-counts.
Two habits keep the arithmetic honest. First, compute both measures over the same population and period, printed side by side; a portfolio whose availability is counted across all doors while MTTR averages only the ticketed doors is telling two different stories with the same breath. Second, keep the row-type column. The moment anyone asks why the number moved, the answer should be countable — three long parts-wait rows left the population this quarter — not atmospheric.
What the Mean Hides: the Tail, the Silence, the Selection
MTTR is defined as a mean over a period and a population, and it tells you nothing about the tail and nothing about the doors that were never reported. That is not a flaw in the measure; it is the measure's boundary, and a worksheet that ignores the boundary ends up optimising the wrong thing. Priya's quarter with the corridor door in B2 is the illustration. Three faults, average interval respectable, and one of them was a door unusable for the better part of a weekend because the removed unit had been retained pending inspection and nobody had written a rule about what a retained unit does to a clock.
The tail is handled by refusing to let the mean stand alone in the review pack. A median companion column costs one extra computation and tells you immediately when the mean is being dragged by outliers; a max row, or a count of intervals longer than whatever threshold the building's own plan treats as severe, keeps the tail visible. The threshold is the plan's to set — the worksheet supplies the measurement, not the target — but the column that shows how many rows exceeded it is the sheet's to carry.
Silence is the harder one. A quiet door is indistinguishable, in the data, from a healthy door; the event log can distinguish them — repeated failed presentations, or a credential path that errors without ever generating a ticket — and so can a simple occupancy-side check: which doors in the population have produced no rows this quarter, and is that plausible? Zero is a result that deserves an explanation, not a tick.
Selection effects creep in through the population and the channel both. When the help desk is reorganised and the ticket queue changes character, the report line stays the same words over a different reality; when a block is taken out of service, its future silence is recorded as improvement. Any measure kept long enough will be gamed, but most games are played accidentally, by teams that never wrote down what changing conditions would make the numbers no longer comparable. A named line on the sheet — comparability notes, one per quarter — is cheap armour.
The question the tail column keeps asking, quarter after quarter, is whether the building would notice if the number improved. If the answer is uncertain, the sheet is measuring the reporting system rather than the estate, and the reporting system is what needs the corrective work. And when the improvement looks too easy to believe, the two companion sheets worth reading are Common Mistakes in Uptime and MTTR KPIs, which catalogues the ways a filled grid still gets the quarter wrong, and Uptime and MTTR KPIs Documentation Set, which is the paper — headers, versions, sign-off — that keeps the definitions from drifting unnoticed.
First-Time Fix and the Row That Looks Clean
Add a flag for whether the visit ended with the door operable and no repeat visit for the same symptom within the window the plan defines. That is the first-time-fix measure, and of everything on the sheet it is the most seductive and the most treacherous, because a unit swap is usually a first-time fix in the letter of that definition and almost never one in the spirit.
The trap is specific: a technician at a door whose motor does not drive can, in the time available, either work the tree of checks — is the handle free, does the override turn, what do the indicator and the app say, is there power, is the recorded state the physical state — or replace the unit and move on. Replacement restores the door fast. It scores perfectly. And if the real cause was a configuration drift or a batch problem rather than the unit, the next door of the same model will produce the same row, scored as another clean first-time fix, while the cause compounds quietly across the estate. The measure cannot see this, because the measure was defined over symptoms, not causes. Nothing tells the truth about the root cause except the evidence columns.
The related discipline lives in The Field Failure Triage Site Record Sheet, whose sequence of checks is what a first-time-fix row should be able to point back to, and in Root Cause Analysis 8d Worksheet: What to Write Down and Why, which is the sheet you reach for when the same symptom keeps arriving from different doors. On the uptime and MTTR KPIs worksheet itself, the honest version of the flag is a paired one: fixed on this visit, and separately, symptom reappeared within the window, yes or no. The gap between the two columns is the size of the swapping culture, and it is one of the few numbers on the sheet that changes how people work rather than only how the quarter reads.
What the first-time-fix column swallows: whether the root cause was found, whether the retained unit was ever examined, and whether the fix was the repair or the paperwork. The paired flag covers the first; the evidence columns cover the rest, which is where we go next.
The Three Clocks, the Contract's Vocabulary
Reviews generate questions that the two-stamp worksheet cannot answer: not how long a fault took, but how long the response took. That vocabulary belongs to the plan, not the sheet, but the sheet's stamps can serve both if the definitions are carried over faithfully. Response time is the interval between the fault arriving through the agreed channel and a person or system acknowledging it; on-site attendance is acknowledgement to a body at the door; resolution is report to operable again. Resolution time is the only one of the three that the worksheet computes directly, because the sheet holds both of its stamps. The other two need the acknowledgement and arrival stamps, which are exactly the timestamps a dispatch system already makes and a ticket record should never lose.
| Measure | Interval between | Whose records create it | What it never tells you |
|---|---|---|---|
| Response time | report through the channel and an acknowledgement | the receiving desk or system | whether anyone was dispatched |
| On-site attendance | acknowledgement and a technician at the door | dispatch | whether anything was fixed |
| Resolution time | report and the door operable again | the worksheet | why it took that long |
| MTTR | the mean of resolution intervals, defined period and population | the worksheet | the tail, or the unreported doors |
| Availability | operable time over scheduled time, stated population | the worksheet | whether that population is the one that matters |
| First-time fix | visits ending operable with no repeat of the symptom | worksheet plus flag | whether the cause was found |
This table is the service level vocabulary of the portfolio, and the column that earns its keep is the last one. When somebody in the review says response times are fine and asks why the slide looks bad, the answer is visible right there: response measures the desk, resolution measures the door, and a plan that manages the first while the building experiences the second will keep producing that argument. Severity classification and the escalation routes hang off the same stamps; the sheet for that conversation is Escalation and Severity Levels Worksheet You Can Fill in on Site, and the wider multi-site version of the whole comparison is Running Uptime and MTTR KPIs Across Several Sites.
One warning about the vocabulary: these words mean different things in different contracts. The worksheet should carry its own definitions visibly, in the column headers themselves, and when a supplier's figures are compared against the sheet, the first question is always which interval their words denote. Which clocks a supplier's own support process can even see is a per-supplier question, answered in writing before the first quarter is compiled, not during it.
The Evidence Columns: What Was Ruled
Everything above produces the two headline figures. This column group produces the thing that makes them defensible when challenged: what happened at the door, in order, and what each step eliminated.
The minimum record, repeated across the whole lifecycle discipline, is a short list: the door identifier, the model and variant from the label or catalogue row, the two stamps, the symptom in the technician's own words plus the measured observation, the checks run in the order they were run and what each one ruled out, the parts fitted or the unit swapped with the old unit retained, the credential state before and after where a credential was involved, and the photo or reading that supports the conclusion. On the uptime and MTTR KPIs worksheet this appears compressed — a ruled-out column, a parts column, a credential flag, a media reference — but compressed to fit the grid, not skipped.
The ruled-out column is the one people cut first, because it is the slowest to fill and the least visible in the moment. It is also the only reason the record is evidence rather than a list of actions. "Checked battery, checked drive, replaced unit" describes activity. "Cells measured within range, rules out power; override turns freely, rules out seized gearbox; drive commanded without movement, points at the drive path; unit replaced, old unit retained as B2-014-OLD" describes a conclusion, and can be audited, disputed or confirmed. A month later, when the same symptom arrives at B2-031, the second description is what tells you whether you are dealing with a repaired door or an unanswered question.
What the evidence columns swallow if kept badly: the retained unit's whereabouts. A record that says old unit retained but lives in no shelf register is a claim, not a fact, and the claim collapses at exactly the moment it matters — when a warranty conversation needs the failed unit itself, which is the territory of Warranty Claim Evidence: A Field Worksheet to Complete. The same discipline at the other end of the supply chain — what gets ordered from where, before a door can be restored at all — is covered by Spare Unit Stock Policy: A Field Worksheet to Complete and, on the ordering mechanics, Smart Lock Spare Parts Ordering.
Spares, Freight and the Long Rows
Availability arithmetic has a physical underlayer that no sheet can abolish: an interval ends when the parts or the unit arrive. Whether a door is out for the length of a diagnosis or the length of a delivery depends on where the spares live, which variants they match, and whether the local technician can draw them without a signature chain. Those are holding questions, and the worksheet's job is to make the answers visible in the intervals rather than to promise them.
Priya's first quarter taught her this through the two long rows in the fragment above. A door down for days with a healthy response time and a dead resolution time is not a support failure; it is a holding policy showing up in the measure. The quarter's real decision came from reading the tail rows against the population register: which models, in which buildings, generated the waits, and whether a small local pool of the matching variants — matched at the variant level, confirmed on the dated product page for the specific SKU, since power and reader arrangements differ inside one family — would have converted either row into an afternoon.
What this column group swallows: the difference between waiting on a part and waiting on a decision. A door held open by an approval chain looks identical, in the interval, to a door held open by freight. The fix is a one-word reason code on the waiting stretch — part, approval, access, repeat-symptom — and a standing instruction that the code is filled from the notes column, not from memory at the quarter's end. Four codes, applied honestly, turn the tail from an embarrassment into a to-do list.
The planning loop that the reason codes feed is the maintenance calendar rather than the incident sheet — the preventive-versus-corrective split is exactly what those codes measure in aggregate — and the calendar's own grid lives in The Maintenance Calendar Design Site Record Sheet. The two sheets share a population register; keeping that register single and reconciled is the difference between a portfolio that learns and two spreadsheets that drift.
Who Writes in the Sheet
A worksheet is a procedure, and procedures have authors. The first-line technician owns the evidence columns and the operable stamp — those can only be written by the person at the door, in the moment, and an evidence column reconstructed later from a van dashboard has the shape of fiction. The dispatch or help desk owns the report stamp, because the report stamp is the channel's own record. The portfolio manager owns the population register, the comparability notes and the row-type decisions, because those are quarterly judgements rather than door-level observations. Nobody owns the arithmetic, which is the point: the figures are computed, never keyed.
The training question is real and has a published answer in this library's own terms — the competence line between a first-line technician who can run the checks and one who can authorise a swap or touch the configuration is exactly the technician skill-matrix question that Smart Lock Installer Training Guide and What Information Should a Smart Lock Supplier Provide? set up. The worksheet should not outrun the training: if the sheet asks for a ruled-out note but nobody has been taught what a check rules out, the column will contain noise, and noise in the evidence columns eventually poisons the arithmetic above it.
Timing rules matter more than format. The report stamp is written by the system when the ticket lands; the rest of the row is closed out at the door or within the shift, and the sheet should say which. A worksheet that tolerates next-week filling trains amnesia, and within a quarter the notes column becomes a genre of creative writing. Priya's version of the rule was small and social rather than technical: a row stays open on the technician's list, visibly, until the ruled-out column is filled — an empty grid cell, on screen, in front of everyone, which turned out to be a stronger nudge than any policy paragraph.
What the authorship column swallows: the handover gap. A fault reported at 21:00 and attended at 06:00 has an overnight stretch in its interval that is nobody's laziness and everybody's design choice. Keep it, don't apportion it — the building experienced it — but flag night-and-weekend starts in the row-type column so the review can tell a resourcing story from a competence story instead of mixing them into one mean.
Carrying the Filled Sheet into Supplier Conversations
The filled worksheet earns its keep in conversations, and the conversations come in two kinds. The first is the portfolio's own review, where the sheet replaces the unfalsifiable slide: figures computed on a named population, a visible tail, reason codes on the long rows. The second is with the support side of the supply chain, and here the sheet's role changes — it becomes evidence, and evidence has format requirements that the supplier, not the buyer, sets.
What the support conversation should extract, and extract in writing rather than in a forum post: what evidence a return or replacement is authorised against and in what form it must arrive; whether the removed unit is wanted back and in what condition; what states the controller holds for a door and how a technician reads the recorded state at the door — since that comparison is a branch no site can run without it; and what the supported way is to compare one installed unit's configuration against another's. Each of those is a CONFIRM-PER-SUPPLIER question — answered per product, per variant, per agreement, never assumed from a datasheet — and the answers belong pinned to the worksheet's header so the next person at the door inherits them. The warranty-handling mechanics on both sides of that exchange are laid out in Private Label Smart Lock Warranty and Smart Lock Distributor Warranty Management.
Two boundaries keep the conversation productive. The first is that the buyer's numbers do not become the supplier's promises. Presenting a quarter's intervals as a demand that future response or resolution be equal to the past invites a contractual number nobody should want; the honest use of the sheet is to ask the supplier's support to work against the same definitions — which interval, which stamps, whose clock — so that both sides' figures become comparable rather than competitive. The second boundary is the cost frame. Every long row is money in the building's own terms, but which rows and how much is a site arithmetic that Smart Lock Maintenance Cost sets up; the supplier's quote conversation starts where that internal arithmetic ends, at Request a Quote rather than at the review meeting.
What this section swallows: the fact that a worksheet is only as credible as its rarest honest column. The comparability note nobody reads is the one that, in the quarter the ticket channel changed, prevents the sheet from being accused of inventing the improvement. Keep it filled, keep it boring.
Local Rules Around the Sheet
The worksheet records events at doors, and doors sit inside buildings that sit inside jurisdictions, so a few questions arrive from outside the sheet and must be written down rather than intuited. None of them has a portable answer, which is why this section exists in every lifecycle document of this cluster: the questions are real, the answers are per-site, and getting them from the people who actually hold them — the owner, the insurer, the appointed consultant, the local authority — is the task.
Where event logs feed the sheet's verification — the repeated-failed-presentation check, the credential state before and after — the retention of card, code and biometric-linked records against occupant privacy law is not a spreadsheet decision. CONFIRM-WITH-LOCAL-LEGAL — retention, access and lawful basis for any lock event or audit records used as worksheet evidence in this jurisdiction. What a photo attached to a row may show, when doors sit in occupied corridors or behind guest-room entries, is the same class of question. CONFIRM-WITH-LOCAL-LEGAL — photography and recording practice in occupied areas for fault evidence, per site rules and local law. Where a building's own life-safety regime governs what any door may be left in while awaiting parts — held open, held closed, signed out of service — that regime outranks the arithmetic. CONFIRM-WITH-LOCAL-AUTHORITY — which local rules govern a door left non-operable pending repair, and who signs off the interim measure. And where an insurer or a landlord has stated expectations about how outages are recorded, the sheet's own definitions should be reconciled to theirs rather than defended against them. CONFIRM-WITH-LOCAL-INSURER — any required incident-record format, channel or attestation for access faults at this property.
Nothing in a worksheet answers these, and a KPI grid should never imply otherwise. Their proper place on the sheet is a reference line in the header — a dated note of whose answers are being relied on — so that a reader three years from now can see which local judgement the quarter was computed under.
The Cell to Write First
If this essay has a single argument, it is that order: the sheet before the number, the population before the sheet, the channel before the population, and one honest word — operable — defined before anything else depends on it. Priya's version ended up taking two pages: the grid itself, header-loaded with its definitions, and a half-page of per-quarter notes carrying the population reconciliation, the comparability statement and the local-rule references. The number she now sends to the review arrives attached to those pages, and the questions in the meeting have changed shape — they are about doors and holds and channels, which are answerable, instead of about a percentage nobody could defend, which never was.
A blank worksheet is not a report waiting to happen. It is the portfolio's written answer to what it will notice, and the notice decides everything the measures can later say. Start with the population register. Add the report rule and the channel. Define the word operable in the building's own terms, with a flag for reduced service. Then, and only then, compute.
None of it is hard, and all of it is discipline; nothing on the sheet asserts that any lock will or will not fail, and no figure published here is anyone's target — the numbers belong to the site that fills in the grid, and the vocabulary belongs to the plan the site writes around it. Where teams want the paper template, the header definitions and the sign-off blocks pre-drafted, the buyer resources library carries the companion formats, the operational answers to routine support questions sit on the faqs and service pages, and the warranty process itself is laid out at support and warranty. When the sheet has told you which holds, spares or variants your portfolio actually needs a decision on, the commercial conversation is shorter — and it belongs at contact or request a quote.

