A signed cleaning contract sits in your file. Buried on page seven is the SLA: response times, frequency commitments, quality thresholds. Six months later, the renewal meeting starts. The client asks: "How are we tracking against the SLA?"
What data do you have? For most contractors, the honest answer is "the client hasn't complained much." That is not an SLA report. That is the absence of escalation.
Key Takeaways
- A measurable SLA names the zone, the metric, the threshold, and the proof. Anything vaguer is aspirational, not enforceable.
- Four metrics actually hold up at renewal: response time, completion frequency, photo coverage, and resolution rate.
- The proof has to come from the work itself. Reports assembled after the fact look defensive — even when they're accurate.
- A contractor who can pull SLA performance per zone, per shift, per month, wins the renewal conversation before price comes up.
What Makes a Cleaning SLA Measurable
Most cleaning SLAs are written like marketing copy. "Maintain a clean and welcoming environment." "Respond to issues in a timely manner." "Ensure all areas meet high standards of cleanliness."
These sentences feel like commitments and behave like nothing. There is no zone, no number, no proof requirement. At the renewal meeting, the client says the environment hasn't felt clean lately. You say it has. Neither of you has the data to settle it.
A measurable SLA has four parts:
The zone. "All restrooms on floors 2-5" is measurable. "The facility" is not. The SLA needs to point at a specific area you can verify against.
The metric. "Cleaned three times per shift" is a metric. "Cleaned regularly" is not. The metric is what the data will say.
The threshold. "Response within 30 minutes of report" is a threshold. "Prompt response" is not. The threshold is the number you either hit or miss.
The proof. "Logged in the system with a timestamp and resolution photo" is proof. "On the supervisor's checklist" is not. The proof is what the client can review without calling you.
A contract clause that includes all four parts becomes defensible. A clause missing any one of them becomes a debate.
The Four Metrics That Hold Up
The cleaning industry has tried to measure quality in many ways. Most of them collapse under scrutiny — too subjective, too easy to game, too expensive to verify. Four metrics survive because the data is naturally event-based: each one creates a record when it happens, not after.
1. Response time on reported issues. When an issue gets reported — by a member, a porter, a manager — the clock starts. When the issue is resolved, the clock stops. The SLA says "30 minutes for safety issues, 4 hours for restroom complaints, 24 hours for general." The data either supports it or it doesn't.
This metric works because the timestamps are automatic. The reporter doesn't have to remember when they reported. The resolver doesn't have to estimate. The system records both.
2. Completion frequency by zone. The contract says "lobby cleaned three times daily." Each cleaning visit creates a scan log with a timestamp. At the end of the month, you can show: lobby cleaned 84 times across 28 days, average 3.0 visits per day. The SLA is met.
This metric works because frequency is countable. Either the scans exist or they don't.
3. Photo coverage on problem zones. Some zones get complaints more than others — usually restrooms, sometimes break rooms, occasionally specific equipment areas. The SLA can specify "photo verification at end of shift for [zones X, Y, Z]." Then you count: did the photo set exist for each shift?
This metric works because photo presence is binary. The photos are either there or they're missing.
4. Resolution rate on escalated issues. Of the issues escalated as urgent in a given month, what percentage were resolved within the SLA window? This metric matters because it captures the failure mode that hurts most — the urgent issue that sits open until the client follows up.
This metric works because the escalation flag and the resolution flag are both system events. No interpretation needed.
These four cover most of what an FM actually cares about. For more on what FMs prioritize when they evaluate contractors, see what facility managers look for in a cleaning company.
Capturing the Data Without Adding Paperwork
The trap with SLA tracking is asking the crew to fill out reports. Reports get filled out partially, late, or inconsistently — and the data quality is poor enough that the reports become argument fuel rather than argument settlers.
The fix is to make the data a byproduct of the work, not an add-on to it.
Scan-based check-ins record when a zone was entered and exited. No separate timesheet. The scan is the timesheet.
Photo capture at completion records the state of the zone. No separate report. The photo is the report.
Issue resolution stamps record when an issue was opened, who took it, what was done, and when it closed. No separate log. The resolution is the log.
The crew does the work the same way they always have. The system captures the data because the data is the work, not paperwork about the work. This is the difference between proof of work and worker surveillance — covered in detail in proof of work vs. surveillance for cleaning crews.
What Defending the SLA Looks Like
Six months in, the renewal meeting is scheduled. Two weeks before, you pull the SLA dashboard for the account.
The contract specified three things: lobby cleaned three times daily, restroom response within 4 hours, and photo verification for problem zones at end of shift.
Your data shows: lobby visits averaged 3.1 per day across the period. Restroom response averaged 2.7 hours, with 94% under 4 hours and the four exceptions tied to a single Tuesday in March where the supervisor was out. Photo coverage on the three flagged zones hit 96% — the four missing shifts are flagged with notes.
You walk into the meeting with a one-page summary and a link to the full dashboard. The client can pull the underlying scans themselves.
The conversation that follows is different from the conversation a contractor without data has. The client is not asking whether the work happened — that question is already answered. They are now asking what to do about the 6% miss on photo coverage. That is a much better question to be answering.
For the full version of this conversation — how documentation changes the renewal dynamic — see how to prove your cleaning work and win contract renewals.
The Renewal Math
The renewal conversation has three possible structures.
Conversation A: No data. The client decides based on whether they remember any complaints. If the relationship feels neutral or worse, they invite competing bids. Price wins.
Conversation B: Reports assembled after the fact. The data exists but looks defensive. The client sees the effort and may extend the contract on goodwill — but the next renewal cycle, the same question comes back harder.
Conversation C: Live SLA data the client can pull themselves. The conversation is not about whether you delivered. It is about how to evolve the SLA for the next cycle. The renewal is rarely in doubt.
The difference between A and C is not service quality. It is documentation infrastructure.
The Bottom Line
An SLA without data is a wish list. An SLA with data is a contract you can defend.
A system like CleanScan captures the SLA inputs as a byproduct of the work itself — scan-based check-ins, photo proof, issue timestamps. The reporting layer rolls those up into per-zone, per-shift, per-month views your client can pull anytime. By the time renewal comes around, the SLA report writes itself.
Elijah Weske is the founder of CleanScan, a platform for scan-based work records and client-facing service reports.
