Managed IT · Service levels
What response and resolution targets make sense for a 25 to 100 person company, the small print that makes an SLA look better than it is, and the questions that cut through it.
Response is not resolution
Watch the business-hours clock
Ask for real numbers, not targets
30 minutes of data collection, your report the next day, yours to keep with no obligation. Or pick a time on the calendar below. Or call 954-539-5678.
I should get the irony out of the way first. JLB tells its managed IT clients we reply within 5 minutes by email or phone and start a screen share within an hour, and I’m about to tell you that response time is mostly marketing. Both things are true. A fast reply matters, because waiting on hold or wondering whether anyone saw your email is miserable when you can’t work. But a reply only means someone has started. What you actually care about is how long until you’re working again, and that number rarely appears in a proposal.
So here is my advice, whoever you’re talking to, including us: listen to the response time, then ask to see last quarter’s resolution times, broken down by priority. A provider who measures them can show you. A provider who doesn’t is selling you the first five minutes.
A small distributor called us late one afternoon because its sales team couldn’t send quotations. The messages were leaving Outlook, but several major customers’ mail systems were rejecting them. We didn’t handle it as a routine desktop ticket, because quotes that can’t go out mean a sales team that can’t sell.
The first thing we did was capture the actual non-delivery report and read it. The rejection traced back to the sending domain’s authentication: a recent DNS change had left the SPF record inconsistent with the third-party system that sent the quotes. We confirmed which senders were legitimate, corrected SPF (without creating a second SPF record), checked DKIM signing and DMARC alignment, and tested delivery to outside accounts. While the DNS change propagated, the team sent time-sensitive quotes through an approved alternate sending path.
That workaround is the part I’d point to. The response that counted was working out which business process was at risk and getting people a safe way to keep quoting before the permanent correction was finished. In the language of the next section, that’s the resolution clock, and it’s the one the client felt.
A service level agreement usually describes up to three different times, and they’re easy to blur together. Response time is how long until the provider acknowledges your request, which, depending on the wording, can mean a person contacting you or just an automatic email with a ticket number. Time to engage is how long until a technician is actually working on the problem. Resolution time is how long until the problem is fixed or a workaround has you working again, and it’s the only one of the three you feel.
Even the large vendors are careful about this distinction. Microsoft’s published support terms for Power BI and Fabric, for example, commit to an initial response time by severity, under an hour for the most critical cases, and say plainly that Microsoft doesn’t provide a direct SLA for resolution. That’s a reasonable position for a software vendor dealing with every possible problem. It’s a weaker position for a company that runs your office’s IT and knows your setup.
None of this is necessarily dishonest. It’s just that the language of an SLA is chosen by the people who have to meet it. Watch for these.
“Four-hour response” usually means four business hours. If business hours are 8 to 5 on weekdays and you raise a ticket at 4:30 on a Friday, half an hour counts on Friday and the remaining three and a half count from 8 on Monday. The provider meets its four-hour promise at 11:30 Monday morning, 67 hours after you asked. Every word of the SLA was kept.
If “response” includes the automatic acknowledgment email, every ticket is answered in seconds and the statistic is worthless. Ask what counts as a response: a human, by name, who has read the request.
Fast targets usually apply only to “critical” or “priority 1” tickets, and the provider often gets to decide what’s critical. If your one accountant can’t open the accounting software during month-end close, is that critical? It is to you. Make sure the definitions in the agreement match what actually stops your business.
“We aim to” and “target” mean something different from “we will”. Look for what happens when the target is missed. Often it’s a small service credit, which is better than nothing, but don’t confuse a credit with compensation for a lost day.
“95 percent of tickets responded to within SLA” can be true while the remaining 5 percent were the ones that mattered. Ask about the worst cases, not only the average.
There isn’t an industry standard that everyone agrees on, so take this as my opinion of fair targets for an outsourced IT provider looking after a company your size during business hours. Your contract may differ, and if you’re in healthcare, law or anywhere with after-hours operations you may need tighter terms.
The resolution numbers are the ones a provider will push back on, and that’s fair; some problems really do depend on a third party such as your internet provider or a software vendor. What I’d expect in return is that the provider tells you when the clock is waiting on someone else, and chases them.
When you’re comparing IT providers, or reviewing the one you have, ask for these. Most good providers can pull them from their ticket system in an afternoon.
The fifth question tells you more about a company than the other five together. Everyone misses sometimes. The ones worth hiring can tell you when and what they changed afterwards.
If you’d like an independent view of how your current setup is actually performing, our free Technology Health Assessment rates IT operations along with cybersecurity, backups, spending and the rest. It takes 30 minutes of data collection, you get a plain-language report the next day with a short list of next steps, and it’s yours to keep with no obligation. Pick a time for the assessment.
Our managed IT service promises a reply within 5 minutes by email or phone and a screen share within an hour, because I think a person should know quickly that someone is on it. We price on users, locations, systems and support needs, and we can work alongside an existing IT person. But please ask us the same six questions you’d ask anyone else. If you’re reading your current agreement while you do, our guide to managed services agreements explains the clauses that usually matter, and break-fix versus managed IT explains why the payment model shapes how fast problems get fixed.
Book it now
Choose a slot that suits you. We’ll confirm by email, collect the data with you on the call, and send your report the next day. It’s free, and the report is yours to keep with no obligation.
Prefer to talk first? Call 954-539-5678, or send the form and we’ll contact you.
For a business of 25 to 100 people, I’d expect a person to respond within 15 to 30 minutes when the whole office is affected and within an hour when one person can’t work. Response time matters less than resolution time, so ask for both.
Response time is how quickly the provider acknowledges your request. Resolution time is how long until the problem is fixed or you’re working again. Many SLAs promise the first and say little about the second.
Often, yes, and it changes the picture a lot. A four-business-hour response on a ticket raised late on Friday can mean Monday morning. Check the definition of business hours in your agreement and whether urgent issues are covered outside them.
Usually a service credit against your next invoice, if the agreement includes one. Repeated misses may give you a right to terminate, depending on the wording. Read the remedies section, and ask the provider how often they missed last quarter.
