What an escalation matrix is
An escalation matrix is a one-page table that tells everyone in support who owns a ticket, and when it must move to someone else. Each row is a severity level with a definition, a first response target and a resolution target; each column is an escalation level with the time it gets involved and how to reach it. When a P1 lands at 3 a.m., nobody has to guess whether to wake the on-call engineer: the matrix already says so.
Functional vs hierarchical escalation
Functional escalation hands the ticket to more skill or access: a senior engineer, a specialist team, the vendor. Hierarchical escalation brings in more authority: a team lead, a support manager, the head of support, who can free up people, approve exceptions or speak to the customer's leadership. In the matrix above, Levels 2 and 3 are usually functional and Management is hierarchical. Keep both: a ticket that only goes up the management chain gets attention but not a fix, and one that only goes sideways can sit with an engineer while the customer hears nothing.
How to set severity levels and times
- Define severity by impact and urgency: how many customers are affected, whether they can work, and whether a workaround exists. Write one concrete example per level.
- Start from the targets you have promised in contracts or on your website, then set escalation times inside them, so a ticket reaches the next level before the target is missed.
- For the top level, escalate early: involving Level 2 immediately costs little, while finding out at the deadline costs the customer.
- Leave higher levels empty for low severities. A how-to question does not need management, and alerting them for it trains people to ignore alerts.
- Name roles and shared channels (queue, Slack channel, on-call rota), not individuals, so the matrix still works during holidays and after team changes.
Example: a SaaS P1 incident, minute by minute
With the SaaS preset, P1 has a 15-minute first response target, a 4-hour resolution target, Level 2 immediately, Level 3 after 15 minutes and Management after 1 hour. A customer reports that nobody can log in at 09:00. At 09:00 the frontline agent tags it P1 and posts it in the escalations channel, so the senior support engineer owns it from the start. If there is still no fix at 09:15, the on-call engineer is paged; by then the customer must have had a first reply, or the response target is breached. At 10:00, if the service is still down, the head of support takes over customer communication and decides on status page updates. The 4-hour resolution target expires at 13:00. Set the "Try it" box to P1 and 20 minutes to see the same answer.
Common mistakes
- Escalation times later than the targets they protect, so the next level only hears about a ticket once it is already late.
- Severity chosen by how loud the customer is instead of by impact; keep a separate flag for key accounts if you need one.
- Personal mobile numbers in the matrix: they go stale and expose people's data. Use rota tools and shared channels.
- No owner for the hand-back: after an escalation, say who updates the customer, or everyone assumes someone else did.
- Never testing the path. Walk through one fake P1 each quarter to check that every contact still works.
Limits of an escalation matrix
The presets are starting points, not standards. Your targets must match what your SLA contracts and support plans actually promise, including whether the clock runs 24/7 or in business hours only (this builder counts calendar time). Review the matrix at least quarterly, and after any incident where the path failed. Check how you are doing against the targets with the SLA compliance calculator and watch the share of tickets that need escalating with the escalation rate calculator: a rising rate usually points to a training or tooling gap at Level 1.
Frequently asked questions
- What is an escalation matrix in customer support?
- An escalation matrix is a table that says, for each severity level, how fast the team must respond and resolve, who takes over when a ticket is not moving, and after how long. Rows are severity levels (P1 to P4), columns are the escalation levels (frontline, specialist, engineering or manager, management) with the time each one gets involved and how to reach them.
- What is the difference between functional and hierarchical escalation?
- Functional escalation moves a ticket to someone with more skill or access, such as a senior engineer or the vendor, because the frontline cannot fix it. Hierarchical escalation moves it up the management chain, such as a team lead or head of support, because a decision, resources or customer communication at a higher level are needed. Most matrices use both: functional first, hierarchical when the clock keeps running.
- How many severity levels should an escalation matrix have?
- Four levels (P1 Critical, P2 High, P3 Medium, P4 Low) is the most common choice because agents can tell them apart quickly. Three can work for small teams; more than five usually means people argue about the level instead of working the ticket. Write one plain example trigger per level so classification is consistent.
- What should an escalation matrix include?
- For each severity: a definition with an example, the first response target, the resolution target, and the escalation path with the time each level gets involved. For each level: the role that owns it and the channel to reach them (queue, Slack channel, on-call phone). Use roles and shared channels, not personal names or numbers, so the matrix survives holidays and team changes.
- How do I make an escalation matrix in Excel?
- Put one severity per row and these columns: severity, definition, response target, resolution target, then one column per escalation level with its trigger time. Add a second sheet as an escalation log (ticket, opened at, severity, owner, status) with a formula for hours open and a lookup to the matrix to flag breached targets. The builder above creates exactly that file.
Related
About the author
Abdessamad Ghanem
Customer Experience & Customer Success Consultant · Founder of Clarivoxx
Abdessamad Ghanem works across Customer Support, Sales, Customer Success, Account Management and Partner Management. He writes Clarivoxx Insights for B2B SaaS professionals who own retention, adoption and customer outcomes.