Cleaning Benchmarking and Competitive Bids: Using Comparisons Without Racing to the Bottom
How to benchmark cleaning costs and use competitive bid comparisons honestly, so a low number is tested against scope instead of matched on price alone.
Every competitive cleaning bid arrives with a number attached, and the number is the least informative part of the document. Two contractors can quote the same building at rates that differ by a third and both be correct, because they are describing different services with the same words.
Cleaning benchmarking and competitive bids only work together when the comparison is built from scope, not from price. A benchmark tells you whether a number is plausible. It never tells you whether a number is right.
Key takeaways
- Benchmark cost per unit of work, not headline price: per fixture, per work station, per visit and per square foot of cleanable area
- Normalise every bid to the same area basis, same frequency and same task list before you compare a single figure
- A bid that is far below the field usually has missing scope, an unrealistic labour estimate or a wage-rate problem, and all three surface later
- The most useful internal benchmark is your own delivered history on buildings of the same type and density
- Score quality separately from price using a weighted evaluation matrix so the cheapest number cannot win by default
- Re-measure the benchmark at every renewal, because wage rates, frequency and fit-out all drift
Why a benchmark is not a price
A benchmark is a range that tells you whether a quote is in the realm of the possible. A price is what you charge for a defined scope on defined terms. Clients use a benchmark to test whether a bid looks reasonable; contractors use one to test whether their own estimate is sane. Neither use is a substitute for measuring the building, estimating labour task by task, and pricing what the schedule actually says. The working method for that is set out in how to bid janitorial contracts.
The rule is simple: build your number from labour first, then compare. A number that starts from a competitor's rate is not a bid. It is a guess with a deadline.
The units worth benchmarking
Headline price per square foot is the number everyone quotes and the number that misleads most often, because floor area is a weak predictor of cleaning labour. These units carry more information:
| Benchmark unit | What it captures | Best used for |
|---|---|---|
| Cost per cleanable square foot | Broad size effect | Portfolio and budget sanity checks |
| Cost per visit | Frequency effect | Comparing a three-visit and a five-visit week |
| Cost per restroom fixture | The heaviest daily labour load in most buildings | Offices, schools, clinics |
| Cost per work station | Furniture and desk density | Open-plan and call-centre environments |
| Cost per hard floor square foot | Floor care and periodic restoration | Buildings with large finished floor areas |
| Cost per entrance or glass panel | Weather, traffic and finish expectations | Lobbies and retail frontage |
A bid that looks cheap per square foot can be expensive per fixture, and per-fixture is usually closer to the truth. Restrooms, kitchenettes and entrances consume labour in a way that floor area does not record.
Normalising a set of bids
Before you compare anything, force every submission onto the same footing. Skipping this is how a comparison produces a confident wrong answer.
- Convert to a common area basis. Rentable, gross and cleanable areas are different denominators
- Convert to a common period. Monthly, annual and per-visit figures are not interchangeable
- Equalise the frequency. Three visits a week and five are different services
- Equalise the task list. Compare task by task, not section by section
- Equalise the inclusions. Consumables, liner and paper supply, periodic floor care and daytime porter cover are the usual gaps
- Rebuild the unit rate from the normalised totals on one cleanable unit
When the numbers are normalised, the spread usually narrows sharply. Bids that looked 30 percent apart often land within a few percent once one has consumables included and the other does not.
Testing a low bid
An unusually low bid is not automatically a bad bid. Sometimes it reflects genuine productivity advantages, a building simpler than the specification suggests, or a route that passes the door anyway. More often it reflects one of a small set of predictable problems.
| What the low number usually means | How to test it | What it costs later |
|---|---|---|
| Scope items missing from the response | Line-by-line comparison against the specification | Add-backs and variations in year one |
| Labour hours under-estimated | Ask for hours per task and per area | Chronic misses on the task list |
| Wage rate below the local market | Ask about loaded rates and relief cover | Turnover, vacancies and failed visits |
| No supervision or inspection time | Ask who inspects, how often and with what record | Quality drifts with no early warning |
| No floor care allowance | Ask what happens in month nine | A large unplanned restoration bill |
| Consumables excluded | Ask who supplies liners, paper and soap | Recurring monthly disputes |
The professional move is to ask, not to assume. A contractor who can explain their hours, rates, supervision model and inclusions is a different proposition from one who cannot.
Building an evaluation matrix
Price should be a scored criterion rather than the deciding factor. A weighted matrix makes the trade-offs visible and defensible, and it removes the temptation to award on the lowest number because it looks like the only objective figure in the room.
| Criterion | Weight | What to score |
|---|---|---|
| Scope completeness against the specification | 25% | Task-by-task coverage, inclusions and exclusions |
| Labour realism | 20% | Hours per area, per fixture and per visit, with method |
| Supervision and quality assurance | 15% | Who inspects, how often, what record the client receives |
| Normalised price | 15% | Total cost on a common basis and period |
| Scheduling and coverage model | 10% | Start times, absence cover, response times |
| References and comparable sites | 10% | Sites of similar type, density and traffic |
| Transition and mobilisation plan | 5% | Staffing, training and a first-90-day plan |
Note where price sits. At 15 percent it still matters, but it cannot carry a bid on its own. The full method, including how to run the scoring session and document the award decision, is covered in cleaning tender evaluation.
Benchmarking from your own history
External benchmarks are published as wide ranges for good reason: buildings differ enormously. Your own records are a tighter and more honest benchmark, provided you keep them in a form you can query.
What to record for every contract you deliver:
- Cleanable area, with the basis and the date measured
- Frequency and the number of visits per week
- Fixture and work station counts per zone
- Budgeted and actual labour hours per period
- Consumable and equipment costs per period
- Inspection scores over time, by area
- Variations and add-backs requested and approved
With four or five comparable buildings on file, you have a benchmark that reflects your own productivity, wage market and supervision model. That is worth more than any published range, and it becomes the evidence you cite when a client challenges a rate. The measurement discipline that feeds it is described in building a cleaning bid walkthrough.
Property management and industry bodies also publish useful high-level data, and BOMA Canada is a reasonable reference point for measurement conventions in commercial property. Treat those figures as a sanity check only, never as a quotation source.
Where benchmarking goes wrong in a tender
- Awarding on normalised price alone. The cheapest normalised bid can still be the weakest on supervision and records
- Comparing across verticals. An office rate applied to a clinic, a kitchen or a warehouse is a category error, not a benchmark
- Ignoring condition. A neglected hard floor costs more per square foot to bring back, and that is periodic work
- Freezing the benchmark. A three-year-old number has absorbed three years of wage and chemical inflation
- Assuming the incumbent's price is the market. It may sit well above or well below, and neither tells you what the scope is worth
The last point matters most at renewal. An incumbent rate is a historical artefact: it reflects the scope and wage market at signature, plus whatever escalations were negotiated since. The mechanics of fixing that are covered in cleaning contract retention, and the cost-side arithmetic sits in commercial cleaning prices per square foot.
Where comparison stops being about cleaning
A benchmark assumes a scope made of cleaning tasks. Where a tender includes work outside that boundary, the comparison stops being meaningful and the item needs to be evaluated on its own terms.
Scope note: Confined space entry, clinical or sharps waste handling, pest treatment, pool water chemistry, machinery servicing, electrical work and blood-borne pathogen remediation are normally outside a cleaning contractor's scope. Confirm in writing what is included, and report rather than act on anything outside it.
Two rules follow, and they apply to both sides of the table:
- Never blend an out-of-scope task into a per-square-foot rate to make a bid look competitive. Price it as its own line or exclude it — critical
- Never accept a bid that carries a specialist task without evidence of training, equipment and disposal arrangements. If it is in the contract, it will be asked for — critical
A comparison that quietly averages a regulated task into a daily cleaning rate is not a cheaper tender. It is an unpriced risk that gets settled in month four, usually at short notice.
Keeping a benchmark alive across a portfolio
For a facility or property manager running several buildings, the practical version of benchmarking is a small, consistent dataset rather than a large, inconsistent one.
Use the same units everywhere. Per cleanable square foot, per fixture, per visit and per month, recorded identically for every site. Consistency beats precision.
Review the outlier, not the average. The site that is well off the portfolio pattern has a scope problem, a staffing problem, or a specification nobody has read in years.
Tie the benchmark to delivered results. A cost benchmark describes what was bought; a scored inspection record describes what arrived. Keeping both in the same system turns a price comparison into a performance comparison, and the measurement method is set out in how to measure cleaning quality.
NeatScore lets you build area-specific checklists with critical items flagged, record completion with photos, and keep a dated report your client can trust. Try it free for 7 days, or see pricing.
Where this fits in the quality system
This sits under our winning commercial cleaning work, which sets out the standard these checks are measured against.
Frequently asked questions
What is cleaning benchmarking?
Cleaning benchmarking is the practice of comparing cleaning cost against a consistent unit of work, such as cleanable square foot, restroom fixture, work station or visit. It is a plausibility test for a price rather than a method for setting one, and it only produces useful answers when area basis, frequency and task list are normalised first.
Why are cleaning bids so far apart for the same building?
Usually because the bids describe different services. One may exclude consumables, periodic floor care or supervision time; another may estimate labour from floor area alone. Once inclusions, frequency and area basis are normalised, the spread between competent bids typically narrows sharply.
Should I always take the lowest cleaning bid?
No. Award on a weighted evaluation that scores scope completeness, labour realism, supervision and quality assurance alongside normalised price. A bid well below the field usually signals missing scope or an unsustainable wage rate, and both become the client's problem within the first year.
How do I benchmark cleaning costs without published data?
Use your own delivered contracts as the dataset. Record cleanable area with its basis and measurement date, frequency, fixture and work station counts, budgeted versus actual labour hours, consumables, and inspection scores. Four or five comparable buildings give you a benchmark grounded in your own productivity and wage market.
How does cleaning benchmarking relate to a competitive bid?
Benchmarking tells you whether a number is plausible; a bid commits you to delivering a defined scope at that number. Use the benchmark to test the bid after you have priced the work from measured labour, never as the starting point for the price itself.
