“How fast will you respond?” is the question every business asks a prospective IT provider, and it receives the least useful answers in the industry: “very”, “usually within the hour”, “we pride ourselves on speed”. None of those is a number, none is contractual, and none survives the first bad Monday. Response time is genuinely measurable and genuinely benchmarkable, which means you can hold any provider (including a prospective one, including us) to figures rather than adjectives. This page supplies the figures.
First, the distinction everything else depends on
Response time is how long until a qualified human is actively working your ticket. Resolution time is how long until the problem is fixed. Providers advertise the first because they can control it; businesses experience the second. Both matter, but confusing them is how a provider with a “15-minute response” leaves your invoicing broken for three days while technically meeting its SLA: someone responded in 15 minutes, then the ticket queued.
Read every promise with that split in mind: a response commitment without any resolution ambition is a promise to say hello quickly.
Priority levels: why “it depends” is the honest answer
No sane support arrangement promises the same speed for a dead server and a wonky second monitor, so tickets are triaged into priority bands. The near-universal scheme:
- P1, critical: business stopped. Server or network down, phones dead, suspected security incident, nobody can work.
- P2, high: a team or key function impaired. A branch offline, email failing for a department, the line-of-business platform degraded.
- P3, standard: an individual impaired but working. One machine misbehaving, a printer refusing, software errors with workarounds.
- P4, low: requests and niggles. New starter setup, a licence question, “when you have a minute”.
The triage itself should be part of the promise: a provider whose P1 definition requires debate mid-outage has an SLA with a hole in the middle.
The benchmark numbers
What the UK managed-services market typically commits to, so you can place any quote against it:
- P1: response within 15 to 60 minutes, worked continuously until resolved, with escalation. Good providers answer P1s with a phone call, not an auto-acknowledgement email.
- P2: response within 1 to 4 hours, resolution ambition same or next business day.
- P3: response within 4 to 8 business hours, resolution typically inside 1 to 3 days.
- P4: response within 1 business day, scheduled by agreement.
Two calibrations on those figures. First, an auto-reply is not a response: the clock should stop at a human engaging, and it is worth confirming a provider measures it that way. Second, faster-than-benchmark headline numbers usually cost money or mean thin definitions; a “5-minute response, guaranteed” is either a premium tier priced accordingly or a promise about answering the phone rather than fixing anything. The contract guide covers how these numbers should appear in writing, including what remedy attaches when they are missed, because an SLA without a remedy is a mood.
The structural point underneath: in a managed arrangement, response times are contracted and monitored; on break-fix you are quoting the market’s queue on the day, which for a P1 means emergency rates and whoever can fit you in.
What actually determines whether you get those numbers
The SLA is the promise; delivery depends on mechanics worth probing before signing:
- Who answers? A staffed service desk beats a voicemail-and-callback loop; ask what happens at 9:01 on a Monday when everyone calls at once.
- Remote-first capability. The large majority of SME tickets resolve remotely, which is what makes fast response physically possible; remote tooling done properly is the engine under every good response number. On-site visits are the exception with their own longer clock, and the split should be explicit.
- Out-of-hours reality. Business-hours cover is standard and fine for most; what matters is that the boundary is written down, that P1 out-of-hours has a defined route (even if charged), and that businesses with real 24/7 needs buy actual 24/7 rather than assuming it.
- Reporting you can read. Monthly figures of actual response and resolution performance against the SLA, unprompted. Providers that publish their numbers to clients tend to be providers comfortable with their numbers, which is itself the signal. It belongs on the provider-choosing checklist near the top.
Frequently asked questions
What is a good response time for IT support?
For a critical, business-down fault: a human working it within 15 to 60 minutes, continuously until fixed. For standard single-user issues: same-day engagement with resolution inside a couple of days. The better question bundles both clocks: how fast do you engage, and how fast do you typically fix?
What’s the difference between response time and resolution time?
Response is when a qualified person starts; resolution is when the problem ends. SLAs guarantee the first tightly and the second loosely (faults vary too much for hard fix guarantees), which is why resolution *reporting* matters: the averages tell you what the guarantee cannot.
What are P1, P2, P3 and P4 issues?
Priority bands: P1 business stopped, P2 team or function impaired, P3 individual impaired, P4 routine requests. Each carries its own response commitment, and the definitions should be written into the agreement so a crisis is never a negotiation.
Are guaranteed fix times realistic?
Rarely, honestly: no one can guarantee a fix time for an undiagnosed fault, and providers offering blanket fix guarantees have usually defined “fix” carefully. What is realistic: guaranteed engagement, continuous work on P1s, escalation triggers, and published averages that show fixes actually happening.
What should happen if the provider misses its SLA?
The agreement should say: typically service credits, escalation rights, and at persistent-failure thresholds, termination rights. The remedy matters less for the compensation than for the incentive; a provider with money attached to its response times staffs accordingly.
How fast is your own response?
Contracted per priority band in the standard shape above, measured to human engagement rather than auto-reply, with performance reported monthly. The specific figures depend on the cover level agreed, which is exactly the conversation the health check ends with; we put the numbers in writing because this page would be embarrassing otherwise.
Get the numbers in writing
The free IT health check ends with a proposal in which every figure on this page appears as a commitment, not an adjective: bands, clocks, remedies, reporting. Bring your current provider’s SLA if you have one and we will benchmark it live. Get in touch; response to that is same business day, since you will reasonably be timing us from the start.
