How to claim an SLA credit after an outage
How credit structures are usually built, what to record while an outage is running, and how to file a claim inside the window.
Does the credit arrive on its own?
Usually not. Published business service level agreements commonly put the burden on the customer: you have to claim, in writing, within a set number of days of the incident, and normally referencing the provider's own trouble ticket. Some state outright that the right to a credit is waived if the claim is late. Miss the window and an otherwise valid credit is not paid.
How SLA credits normally work
| Element | Typical structure | What to check in your SLA |
|---|---|---|
| Trigger | Availability falling below a stated figure, or a repair time being exceeded | Whether the measurement is monthly and how downtime is counted |
| Credit size | A percentage of the monthly charge, rising with the length of the outage | Whether it applies to the affected service only or the whole bill |
| Cap | Always capped, and often far lower than people expect — some published broadband SLAs cap the total monthly credit at a single day of the monthly charge, while dedicated services are commonly more generous | The cap itself, and whether multiple incidents in a month stack or share one cap |
| Claim window | Around 30 days is a common figure, running from the incident rather than from the invoice | The exact number of days, whether they are calendar or business days, and what date they run from |
| Exclusions | Scheduled maintenance, power failures, customer equipment, and events outside anyone's control | How much notice counts as scheduled maintenance |
| Form of credit | A credit on a future invoice, not a refund | Whether it survives if you leave before the next invoice |
What to do during the outage
Open a trouble ticket immediately
Most SLAs measure downtime from the moment a ticket is opened, not from when the service actually failed. Minutes recorded now are money later.
Record the ticket number and the timestamp
Both are what the credit calculation will be built on.
Keep your own timeline
When it failed, when you reported it, each update you received, when service was restored, and who confirmed it.
Capture evidence as you go
Monitoring graphs, alert emails, screenshots of the portal status. Retrieving them weeks later is harder than it sounds.
Ask for the closure time in writing
The restoration timestamp on the ticket is what determines which credit band applies.
What to do after service is restored
Read the SLA before you write
Find the availability figure, the credit table, the claim window and the exclusions. Claim against the terms, not against how bad the day was.
Send the claim in writing within the window
Include the account, the affected service, the ticket number, the start and end timestamps, the total downtime, and the credit you believe is due.
Do the arithmetic yourself
State the percentage and the dollar figure. Claims with a number attached are processed; claims without one are discussed.
Confirm which invoice it lands on
Get it in writing, then check that invoice when it arrives. Approved credits do go missing.
Log repeat incidents
A pattern of outages is a stronger basis for renegotiation, or for release, than any single credit.
What a credit is not
- It is not compensation for lost business — most SLAs exclude consequential loss explicitly
- It is not automatic, and a provider is generally not obliged to tell you that you qualify
- It is rarely cash; it is usually a credit against a future invoice
- It does not usually cover time the provider classifies as scheduled maintenance
- It does not cover failures of your own equipment or power
If you are not sure what your agreement actually guarantees, what an SLA guarantees covers how to read one. If the credit is agreed and then never appears, treat it as a billing error and follow disputing a charge.
Service level terms, credit bands and claim deadlines differ between providers and between contracts. Your own SLA governs.
Last reviewed September 14, 2026
