An SLA percentage is only as honest as the priorities and clock pauses behind it. Check the data before you measure it, and the number you send the client will survive their questions.
Every PSA will tell you what percentage of last month’s tickets met their SLA. Few MSPs check whether that number is right. A P3 that should have been a P1, a ticket left in “waiting on client” for a week while nobody chased, or a server outage excluded as “planned maintenance” can each move the figure by several points, in either direction. This free SLA performance review checklist gives service managers and service-desk leads a monthly routine for getting the number right and acting on it. It exports the month’s tickets, validates priorities, pauses and exclusions, measures response and resolution by priority, and lists every breach. In a month with breaches, a conditional phase finds the root cause of each one, calculates any service credit the agreement requires and routes it for approval before it reaches the client or the invoice. The month closes with a client SLA summary, improvement actions and a line for the next QBR.
The Agreement Sets the Targets. The Monthly Review Proves Whether You Met Them.
Three processes touch an MSP’s service levels, and they are easy to blur. The agreement defines them: priorities, targets, service hours, clock rules and the credit schedule. Our IT Support Agreement Checklist covers drafting and reviewing those terms. The MSP QBR Checklist sits at the other end, where a quarter of results gets one slide and a conversation about the client’s business. This checklist is the work in between. It runs every month, for every client with SLA commitments, and its job is narrow: produce a figure you can defend and settle any credits it triggers.
The usual failure is not missing targets. It is reporting a number nobody has checked. When the client’s office manager remembers a two-day outage and your report says 98% compliance, the conversation stops being about service and starts being about whether your reporting can be trusted. A short validation step before measurement prevents that, and it also stops you from crediting money you did not owe.
PSA SLA report
Generated, unchecked
Source: whatever priorities and statuses technicians set.
Exclusions: applied by status, whether or not they were justified.
Breaches: counted, not explained.
Credits: left for someone to notice at renewal.
Monthly SLA performance review
Validated, explained, settled
Source: the same export, with priorities and pauses checked.
Exclusions: each one evidenced against the agreement.
Breaches: root cause recorded and an action raised.
Credits: calculated, approved and passed to billing the same month.
What the SLA Performance Review Checklist Covers
Six phases take each client’s month from ticket export to QBR input. Phase 4 appears only when a target was missed.
Phase 1
Phase 1: Open the Review Period
Run it in the first week of the month, before the monthly service report is written, so the report uses validated figures.
Open the review for the client and month — confirm the agreement, the SLA schedule that applies and the measurement period
Pull the SLA terms from the agreement — targets by priority, service hours, clock-pause rules, exclusions and the service credit schedule
Export the month’s tickets from the PSA — every ticket opened or resolved in the period, with priority, timestamps, status history and paused time
Carry forward last month’s improvement actions — each with its owner, due date and current status
Phase 2
Phase 2: Validate the Data
Every change made here is recorded with a reason. The client should be able to see what was adjusted and why.
Check priorities against the matrix — review every P1 and P2 and a sample of the rest, and correct any priority that doesn’t match impact and urgency
Check clock pauses — every pause for the client has a matching request for information, and no ticket sat paused without a follow-up
Apply agreed exclusions — planned maintenance, third-party outages and client-caused incidents, each with evidence and the clause that allows it
Remove tickets that don’t belong — duplicates, merged tickets, spam and internal work logged against the client
Record every adjustment — the ticket, what changed, why and who approved it
Phase 3
Phase 3: Measure Performance
Calculate response compliance by priority — the share of tickets that met the first-response target, counting a human response, not an automatic acknowledgement
Calculate resolution compliance by priority — the share resolved within target, with paused time and valid exclusions removed
Measure availability commitments — uptime for any managed service the agreement covers, using the measurement method it defines
Compare with the previous three months — the trend by priority, so one bad week doesn’t hide a slow decline or the reverse
List every breach — ticket number, priority, target, actual time and how far it was missed by
Phase 4 — Breaches Only
Phase 4: Breach Review & Service Credits
Shown only when at least one target was missed. In a clean month the checklist goes straight to the client summary.
Find the root cause of each breach — dispatch delay, staffing, an escalation that stalled, a vendor, or a priority set wrongly at intake
Check whether the result triggers a credit — some agreements credit per breached ticket, others only when monthly compliance falls below a threshold
Calculate the service credit — with the formula, the fee it applies to and the cap set in the agreement, and keep the workings
Approve the service credits — the service manager answers Approved or Not approved before anything reaches the client or the invoice
Pass the approved credit to billing — the amount, the clause it comes from and the invoice it should appear on
Raise a problem record for repeated causes — when the same cause produced more than one breach
Phase 5
Phase 5: Client SLA Summary
Write the SLA summary — compliance by priority against target, the three-month trend and a plain explanation of each breach
Show the adjustments — excluded tickets and paused time in an appendix, so the client can check the arithmetic
Peer review the figures — a second person checks the summary against the export and the adjustment log
Deliver the summary — inside the monthly service report or on its own, as the agreement specifies, with any credit shown
Phase 6
Phase 6: Improve & Feed the QBR
Agree improvement actions — for each weak priority or queue, one change with an owner and a date
Flag targets that no longer fit — targets missed or beaten by a wide margin for three months go to the agreement review
Add the month to the QBR pack — compliance trend, credits issued and the actions taken
Close the review — attach the export, the adjustment log and the summary as sent
Seven Measurement Rules to Check Before You Trust the Number
Most arguments about SLA performance are really arguments about definitions. The agreement should settle each of the rules below. When it doesn’t, agree the interpretation with the client once, write it down and apply it the same way every month. The middle column shows common practice, not a standard. Your agreement takes precedence.
Rule
Common practice
What to check each month
When the clock starts
When the ticket is logged, by email, portal or phone
Phone calls logged late, so the ticket looks younger than the problem
What counts as a response
A technician’s first contact, not an automatic acknowledgement
Auto-replies or bulk status updates counted as responses
What counts as resolved
Service restored, sometimes by a workaround the client accepts
Tickets closed then reopened within days, counted as two successes
Service hours
Business-hours clock for P3 and P4, often round the clock for P1
Out-of-hours P1s measured on the wrong calendar, or bank holidays missing
Clock pauses
Paused while waiting for the client, with the request recorded
Pauses with no request, or tickets parked to protect the figure
Exclusions with no evidence, or outages labelled as maintenance after the event
Priority changes
Measured against the corrected priority, with the change logged
Downgrades made after a breach to make it disappear
A worked service credit example
Credit schedules vary widely, so treat this as an illustration of the arithmetic, not a recommendation. Suppose the agreement measures P1 and P2 resolution compliance against a 90% target each month and sets three bands: 85% to 89.9% earns a credit of 5% of that month’s managed service fee, 80% to 84.9% earns 10%, and anything lower earns 15%, which is also the cap.
The client’s monthly fee is £4,000. In September, 26 P1 and P2 tickets were raised and the PSA shows 22 resolved within target: 84.6% compliance, a 10% credit of £400.
Validation finds one P2 that should have been paused for four hours while the client sourced a replacement part, which the agreement allows. The technician forgot to set the status.
Recalculated, 23 of 26 met target: 88.5%, a 5% credit of £200. The adjustment log shows the client which ticket changed and why.
Validation cuts both ways. It removes credits you don’t owe, and it also finds the outage someone excluded as maintenance after the event. Either way the summary can show the client exactly which tickets missed, why, and what is being done, and the credit appears on the next invoice without anyone having to ask for it. Many agreements also make credits the client’s sole financial remedy for missed targets and require them to be claimed within a set period. Read your own terms before you promise anything, and note the claim period in the review.
Why Run Your SLA Reviews in CheckFlow?
1
The same checks, every client, every month
A recurring schedule creates the review for each client on the first working day of the month, with the export and validation assigned to the service-desk lead. Priority checks and pause checks happen because they are tasks, not because someone remembered them.
2
No credit leaves without approval
The breach phase appears only in months that need it, and the credit task asks the service manager for an Approved or Not approved answer before billing sees a figure. The approval is recorded with the name, the date and the workings attached.
3
A record that answers the hard questions
Every adjustment, exclusion and breach cause is kept against the client and the month. When a client disputes a figure at renewal, you can show the export, what was changed and why, rather than rebuilding a month from memory.
SLA reviews are one of the monthly processes the MSP process management guide recommends standardising first. CheckFlow for MSPs runs them alongside onboarding, reporting and QBRs across every client.
It is a monthly check of how a managed services provider performed against the service levels in a client’s agreement. The review exports the month’s tickets, checks that priorities, clock pauses and exclusions were applied correctly, measures response and resolution compliance by priority and explains every breach. Where the agreement provides for service credits, the review calculates them and passes approved credits to billing.
How do you calculate SLA compliance?
+
For each priority, divide the number of tickets that met the target by the number of tickets measured, after removing valid exclusions. Calculate response and resolution separately. Avoid reporting a single blended figure across all priorities: thirty quick P4s can hide two missed P1s, and the P1s are the ones the client remembers. Agree with the client whether a ticket is counted in the month it was raised or the month it was resolved, and keep to it.
Should time waiting for the client count against the SLA?
+
Usually not, if the agreement allows the clock to pause and the pause is genuine. That means a recorded request for information or access, and a follow-up if the client doesn’t reply. Pauses with no request behind them are the most common way SLA figures become unreliable. The validation phase checks every paused ticket for exactly this reason.
How are MSP service credits usually calculated?
+
It depends entirely on the agreement. Common structures include a fixed credit per breached P1 or P2 ticket, a percentage of the monthly fee when compliance falls below a threshold, or bands that increase the percentage as performance drops. Most schedules cap the total credit in a month. Calculate exactly as the agreement says, keep the workings with the review and have the service manager approve the figure before it reaches an invoice.
Should we apply a service credit if the client hasn’t asked for it?
+
If the agreement says the credit is due, applying it unprompted is usually the better commercial decision, even when the contract requires the client to claim. A credit the client discovers for themselves at renewal costs far more in trust than it saves in cash. Check your own terms first. Some agreements require a claim within a fixed period, and the review should record when that period ends.
What should we do when we keep missing the same target?
+
Find out which it is: a capacity problem, a process problem or a target that was never realistic. Root causes from Phase 4 usually make that clear within two or three months. Fix process problems with a problem record and a change. If the target itself is wrong for the service the client pays for, take it to the agreement review rather than letting compliance quietly erode.
Is CheckFlow free for this template?
+
14-day free trial, no card required. The Business plan is $10 per user per month after the trial. Full details at checkflow.io/pricing.
Report SLA Figures You Can Defend, Every Month
Free trial — no credit card required.
Do you like cookies? 🍪 We use cookies to ensure you get the best experience on our website. Learn more