SLA breaches: why you hear after the customer

Monday’s support report shows eleven SLA breaches last week. Four came from customers with a one-hour response time in their contract, and one of those customers has already emailed their account manager to ask what they’re paying for. Everyone in the review agrees it was a busy week. Nobody can do anything about it, because the report arrived after every one of those tickets could have been saved.

Response times are the part of customer support your customers can hold you to, because they’re written into the contract.

What is SLA breach alerting?

SLA breach alerting gives each ticket a target from the customer’s contract, and runs a clock that pauses only while the customer owes you a reply. It warns the person assigned at set points before the target, while there’s still time to act. It also records every breach and near miss, so targets and staffing can be argued from data.

Why you hear about breaches after the customer

1. Everyone gets the same target. A premium customer and a free trial on the same clock means one of them is wrong.
2. The alert fires at the breach. By then it’s a report, not a warning.
3. The clock pauses for the wrong reasons. Pausing while a ticket waits on another team hides the delays you most need to see.
4. Warnings go to a channel. A warning everybody sees is one nobody owns.
5. Near misses aren’t counted. They show where the next breach will come from, and nobody looks.

What it costs

A breach is a broken promise in writing. It comes up in the renewal conversation, sometimes as a service credit, and always as a reason to doubt the next promise.

It also costs the chance to fix the pattern. When breaches only appear in a weekly report, the review argues about last week instead of staffing for next week.

What to fix first

1. Read the target from the contract. Pull each customer’s response and resolution times from the CRM onto the ticket.
2. Warn at half and at eighty percent of the target. Send the warning to the assignee by name, with a link to the ticket.
3. Pause the clock only while the customer owes you a reply. Internal waits keep the clock running.
4. Escalate the second warning. If the ticket is still open at eighty percent, tell the team lead too.
5. Record breaches and near misses. Report both by customer tier, team and time of day.

The guide to building SLA breach alerts has the prompts for each step.

Where it fits

SLA alerts need routing to set the right priority and intake to match each ticket to the right contract. When a ticket is waiting on engineering, its clock is still running, which is why the escalation process needs status flowing back to the ticket.

For the whole customer support process, from first message to fix, see the customer support process.