Am I too slow? The published office rates disagree by 7×

Published production rates for the same kind of office run from 2,500 to 18,000 square feet per labour hour. No benchmark that wide can tell you whether your crew is slow. Here is what can.

It is one of the most common questions an owner-operator asks, usually after a night that ran long: am I too slow? Sometimes it is asked about a crew, sometimes quietly about themselves.

The instinct is to look up a production rate and compare. That instinct fails, because the published rates for one space type — ordinary office — do not agree with each other. They are not a few percent apart. They span a factor of seven.

What is actually published for office work

These are figures we read on pages we fetched, quoted as published, for general or open office cleaning. They are not a ranking and not a standard; they are a snapshot of what an operator finds when they go looking.

Set the last row aside for a moment and the rest still run 2,500 to 5,000 — a two-fold spread for the same task in the same kind of room. Put the last row back and the published answer to “how fast should an office clean?” runs from 2,500 to 18,000, which is 7.2 times.

A benchmark that wide is not a benchmark. Whatever your crew did last night, some published figure says they were fast and some other published figure says they were slow.

What the spread does to one building

Take a single 10,000 sq ft office, cleaned three nights a week, eight restroom fixtures, 0.4 hours of travel and lock-up per visit. Nothing varies except the production rate you believe. Months are 13 visits at three a week (52 weeks ÷ 12).

Every row adds 0.44 hours of fixture time and 0.40 hours of travel to the floor time, so a reader with a calculator gets the same numbers. The range is 18.20 to 62.92 labour hours a month on one building — 3.5 times — and $491.40 to $1,698.84 of labour to buy. That is a gap of $1,207.44 a month, or $14,489.28 across a twelve-month contract, created by nothing but which page you happened to read.

Now price it. Say the contract pays $1,800 a month:

One of those is a thin job you should reprice. One of those is an excellent account. They are the same contract. The only thing that changed is an assumption you borrowed from someone else’s page.

So measure your own. It takes one account and ten minutes

You do not need a table. You already run buildings, and a building you run is a measurement nobody can argue with. The method is division.

1. Pick one account you know well and whose hours you can state honestly.
2. Get its cleanable square footage — the area the crew actually touches, not the lease number.
3. Take the hours the crew really spends per visit. Not the bid hours. The real ones.
4. Back out the time that does not scale with floor area: fixtures and travel.
5. Divide area by what is left.

Worked through, on an account of 14,000 cleanable square feet where the crew spends four and a half hours a visit and there are seven restroom fixtures:

That is a real number about real people in a real building. Do it across four or five accounts of the same type and take the middle one, and you have a rate card calibrated to your crew, your equipment and your buildings. From then on the published figures are a curiosity, not an input.

“Too slow” is a money question, not a speed question

Here is the part the benchmark hunt hides. Suppose that 14,000 sq ft account pays $2,100 a month and the crew works 4.50 hours a visit, three nights a week.

Whether 3,769 sq ft an hour is “slow” is now answerable, because it has stopped being a comparison with strangers:

At a $24 loaded cost that crew is not slow, they are profitable. At $30 they are still profitable, with less room for a bad month. If the loaded cost were $36 they would be underwater at any speed, and no production-rate table would ever have told you, because the table does not know what an hour costs you.

Your crew is not slow or fast against the industry. They are slow or fast against your own break-even, and that is a number only you can compute.

And if the answer is yes, you are slow

Then speed has a price tag, which makes it a decision rather than a worry. On the same account, getting 15% faster — 4.50 hours a visit down to 3.91 — takes the month from 58.50 labour hours to 50.83. At a $27 loaded cost that is $207.09 a month and $2,485.08 a year, on one building.

That is worth real effort: better equipment, a route map so nobody crosses the building twice, fixing the vacuum that lost suction, training a new hire properly before putting them on a tight account. It is also worth knowing when it is not worth it — if the account yields $8,353.80 a year as it stands, chasing another $2,485.08 may matter less than repricing the account next door that is yielding nothing.

What you cannot do is answer the question by looking it up. The published figures for one kind of office disagree by seven times, and the gap between the slowest and the fastest of them is larger than the gap between a good contract and a loss.

Method and sources

Production-rate figures are quoted as published on the pages named below, read on 19 October 2026 except the association estimating figure, which was read on 12 October 2026. The 2,500, 3,000, 3,500 and 5,000 figures all come from a single operator’s published post, which states the first three as its own and attributes the “up to 5,000” figure to two industry bodies; we quote that attribution as the operator makes it and have not independently verified it. The 3,000–4,000 band is the range commonly published for basic office work across several sources, listed in our rate benchmark audit. The 18,000 figure is the arithmetic conversion of one association’s stated estimating example of 1 minute per 300 sq ft for standard office space; it is quoted for commentary and criticism, and nothing here is derived from any association’s proprietary time standards.

All worked figures use 0.055 hours per restroom fixture and 0.40 hours of travel and lock-up per visit, and months of 13 visits at three nights a week (52 ÷ 12 weeks, rounded to two decimals at each displayed step so that the arithmetic shown reconciles). The self-measurement example and the $2,100 contract are illustrative, not drawn from a real client.

• My Clean Pivot — why cleaning production rate matters
• ISSA — how to calculate commercial cleaning rates
• ProveBid — the published rate benchmarks don’t reconcile

On this analysis. We sell a bidding and contract-tracking tool, so we are not a disinterested party — which is why every assumption above is stated and the arithmetic is written out for you to disagree with. If we have misread a source we will correct it in public: support@provebid.com. This is a planning aid, not tax, accounting or legal advice, and it is not affiliated with or endorsed by any industry association.