What it means
Service levels turn promises into measurable standards, and contracts can set response, delivery or availability targets. Exceptions arise when actual performance falls short or when data cannot be classified without review.
Keeping a record helps the team make an honest assessment instead of relying on a green or red dashboard alone. First identify the exact target and clock.
Does response time begin when a ticket arrives, when it is categorised, or during staffed hours only, and does the delivery promise end when the driver reaches the building or when the customer signs? A system may pause a clock while waiting for customer information, but only if the contract and operating procedure allow it, so use the agreed definition, not the most favourable interpretation after a miss.
Collect the evidence before assigning fault, because ticket timestamps, dispatch logs, call records, monitoring data and customer acknowledgements may disagree. Investigate missing data and manual edits, and label an exclusion only with a reason supported by the agreement and facts.
Notify the right owner early, since the frontline team may be able to restore service while an account manager handles the customer and a contract owner evaluates remedies. Critical outages or safety problems need immediate incident response, not only a monthly exception report.
Explain the problem and the next update time without promising a credit or admitting a contractual breach before the relevant terms and evidence are reviewed, and keep a record of customer communications. Resolve the operational issue, then examine why it happened, because a late delivery may follow a poor route plan, an unreliable supplier, a wrong address or an unrealistic sales promise.
Corrective actions should address the cause and have an owner and due date, and repeated exceptions of the same kind can mean the service model is not matched to demand. Review patterns by customer, site, priority and day, while respecting privacy and contractual confidentiality.
The commercial result may include a service credit, re-performance, an improvement plan or no contractual remedy, depending on the actual agreement, and a potential credit should not be netted against revenue without the accounting review. Check any claim procedure and notice period, and record the decision and link it to the underlying event and evidence.
For owners, exception management is a feedback loop: the aim is to see a broken promise clearly, repair the customer's service, apply the agreed terms and reduce recurrence.
In practice
Real-world examples.
Example
A help desk responds in three hours where the agreed target is two. It logs the missed response and investigates the queue rather than resetting the ticket timestamp.
Example
A delivery arrives after the stated window because the customer changed the address. The team records the change and checks whether the agreement treats it as an exclusion.
Example
A system outage lasts 40 minutes during a planned maintenance window. The provider checks whether the maintenance was properly notified before classifying its effect on availability.
Formula
Calculation
Service level exception rate = Valid exceptions in the measurement period / Eligible service events in the period x 100
Worked example. An invented courier tracks 1,000 deliveries under an agreed on-time rule. It validates 28 misses after checking timestamps and the contract's exclusions.
- Service level exception rate = 28 / 1,000 x 100 = 2.8%.
- A disputed delivery should be investigated before it is either counted as a valid exception or excluded.
Different contracts may measure compliance using averages, percentage targets or incident severity. Use the contract's actual formula for the formal result.Case study
Seen in the real world.
This illustrative and entirely fictional example follows Lumen Support, an invented IT service provider. Its monthly report showed 99% compliance with response targets, but a customer identified several critical tickets that waited too long. The system had paused those clocks automatically whenever an agent selected "awaiting customer," even when no question had actually been sent. Lumen reviewed the original tickets, timestamps and customer messages. It reclassified the unsupported pauses as exceptions, corrected the month's report and discussed the agreed remedy with the customer.
The support lead changed the workflow so a pause required a recorded request for information and an audit trail. Critical tickets were also escalated before they reached the response deadline. The corrected compliance percentage was lower, but the customer could see the issue and its repair. Later reviews compared exceptions by cause. Accurate measurement became part of the service.
Watch out
Common mistakes.
- Changing timestamps or labels after a miss to keep a dashboard green without evidence.
- Calling every exception a contractual breach before checking definitions and exclusions.
- Counting an operational fix as complete while the affected customer and contractual treatment remain unresolved.
Questions
People also ask.
Is every missed target an SLA breach?
Not necessarily. Check the agreed calculation, exclusions, notice rules and remedies alongside the facts.
Who should own an exception?
Assign operational recovery, customer communication and contract review to named owners, even if different people handle each.
How can repeated exceptions be reduced?
Group them by cause, test corrective actions and track whether recurrence falls without weakening the measurement rule.
From the founder's library
Take it further with the book.
Build your financial confidence beyond this definition. Shihan's full-length guide, Accounting Fundamentals, takes the same plain-English approach and turns it into a complete, practical playbook for non-finance managers, business owners and students - with chapter-end quiz answers and presentation slides included.
25% off with code MMHQ25, applied at checkout. Priced in USD - checkout may show the equivalent in your local currency.
View the book and save 25%