Proving you met the SLA, not just meeting it

A service level agreement usually sets out numbers — coverage percentage, response times, how quickly a vacant post gets filled. Most contractors genuinely meet them most months. The separate question, which only surfaces when a client asks, is whether you can prove it with something more specific than "yes, we did."

The difference between meeting and evidencing

Meeting an SLA happens operationally, shift by shift, largely invisibly when everything goes to plan. Evidencing it requires that each of those shifts left a record — who was rostered, who actually attended, what time cover was found when a gap occurred — that can be pulled together after the fact into something a client would accept as proof, not just an assurance.

What tends to be missing when it's asked for

• –Attendance recorded against the rostered shift, rather than just "the site was covered" as a general statement
• –A timestamp for how a cover gap was filled, not just that it eventually was
• –A clear distinction between a shift that ran as planned and one that ran with a substitute officer
• –A consistent method of counting, so this month's numbers are comparable with last month's

The last point matters more than it sounds. An SLA report that changes its own definition of "covered" from month to month is not really evidence of performance — it is a number that happens to look acceptable, produced by whoever assembled it that particular week.

Where this connects to reporting generally

SLA evidence is a narrower version of the same problem covered in client reporting: what to send, how often, and what it costs you — the report is only as trustworthy as the records underneath it, and rebuilding those records by hand every month is both expensive and the point at which small inconsistencies creep in.

If attendance and cover are recorded against the rostered shift as a matter of course, SLA evidence becomes a view of that data rather than a project someone has to run before a contract review.

Performance you cannot evidence is performance you cannot bill

Contractors routinely meet their obligations and are unable to demonstrate it. Patrols happened, response times were met, posts were covered — and the record is a mixture of paper logs, an officer's recollection and data that was never collected. At renewal, or in a dispute about a deduction, that is functionally the same as not having met them.

The distinction worth drawing is between data captured because it is operationally useful and data captured because the contract will be judged on it. They overlap but are not the same set, and the second is often the one nobody specified.

Reading the contract for what it will be measured on

Worth doing once, properly: go through the contract and list every commitment that could be measured, then check whether anything currently captures it. The gaps found in that exercise are almost always things everybody assumed were being recorded somewhere.

Common mistakes

• –Capturing operational data without checking it covers contractual commitments
• –Evidence in formats that cannot be aggregated or reported
• –Relying on recollection for anything not routinely logged
• –No timestamped record of patrols or response times
• –Discovering the gap during a dispute rather than at mobilisation
• –Never reconciling what is captured against what the contract measures

What separates evidence from recollection

Capturing the measurable commitments deliberately, because someone read the contract and listed them, rather than hoping operational data happens to cover them. Those two sets overlap and are not the same, and the gap between them is discovered during disputes.

Worth doing once at mobilisation and revisiting when a contract is varied — which is the point at which new commitments arrive and nobody thinks to check whether anything captures them.

One practical test: evidence last month against the contract

Take the contract, list every commitment that could be measured, and try to produce evidence for last month against each one.

The gaps found in that exercise are almost always things everybody assumed were being captured somewhere. It is worth doing once at mobilisation and again whenever a contract is varied, because variations add commitments and nobody checks whether anything records them.

• –List measurable commitments from the contract, not from operations
• –Check what currently captures each one
• –Repeat the exercise when a contract is varied
• –Capture in a form that can be aggregated and reported
• –Do not rely on recollection for anything not routinely logged