Comparing cleaning bids on cost per square foot
One calculation that makes three incomparable proposals line up, and what the resulting numbers actually mean.
Three bids, three different scopes, three different monthly figures. The usual response is to pick the middle one. There is a better move, and it takes about two minutes.
The calculation
Convert every bid to cents per cleanable square foot per month, then normalise it to a common frequency.
monthly figure
÷ cleanable square feet
× 100
= cents per square foot per month
Cleanable square footage is the number you should already have asked every bidder to state. If two of them used different figures, that alone explains part of the spread.
Then divide by the frequency factor so that a five-night bid and a three-night bid can be read on the same scale. Frequency does not scale linearly, because travel, setup, supervision and supplies are close to fixed per night. As a rough working set against a five-night baseline: three nights is about 0.7, twice a week about 0.5, once a week about 0.3.
What the numbers tend to mean
These are the bands we see in our own market, expressed per square foot per month at five nights a week. Yours will differ by region, by labour cost and by how much of your floor is carpet.
| Building type | Typical band |
|---|---|
| Warehouse and industrial | 9 to 15 cents |
| Office | 16 to 26 cents |
| Retail | 18 to 28 cents |
| Schools | 21 to 32 cents |
| Gyms and studios | 25 to 38 cents |
| Medical and dental | 28 to 42 cents |
A bid meaningfully below the band for your building type is not necessarily bad. It is a question. Ask it, using the six checks in the scope document, and the answer is usually one of: fewer hours, thinner frequencies, subcontracted labour, or extras that will be invoiced later.
Where the calculation stops working
It compares like with like, so it fails when the scopes are genuinely different. If one bid includes a day porter and two do not, strip the porter out before comparing. Same for consumables, floor programmes and window work.
This is the real reason to insist on separately priced lines: not because bundling is dishonest, but because a bundled bid cannot be compared with anything.
The second calculation
Once you have the cents figure, do one more: the annual all-in cost, including every extra at its stated frequency. Monthly figures hide an annual carpet extraction and a twice-yearly strip and refinish rather effectively.
A bid that wins on monthly and loses on annual is common enough to be worth ten minutes of checking.