Do MSPs Need Service Level Agreements?
During a panel discussion at the Kaseya User Conference, an interesting debate popped up: Do managed service providers really need to offer service level agreements (SLAs)? And if so, how should they be written? I heard a range of opinions, but here are the three I heard most often.
Opinion 1 – Keep It Vague: Offer “half” an SLA, promising that you’ll hit key metrics (for instance, perhaps you can offer three-nines reliability or one-hour emergency response support). But don’t put in writing the penalties you’ll pay for failing to meet those SLA metrics.
One member of the MSPmentor 100 told the audience that he uses that approach, and no customers have ever balked at the vague SLA agreement. His big concern was paying back-fees to customers if his one-hour emergency response actually required 65 minutes. As long as he was in the ballpark with his SLAs, customers were happy, the MSP said.
Opinion 2 – Include the Semi Colon: In other words, this service level agreement outlines what you’ll offer; and what penalties you’ll pay if you don’t fulfill the SLA.
Opinion 3 – Try to avoid SLAs altogether: Is this really possible in the managed services industry?
I’m sure there are other options and approaches. I’m all ears. Let the debate continue.
Joe,
I think an SLA is vital if your objective is to optimize network configuration, reduce service tickets and maximize asset availability.
If you are using Managed Services to augment your break-fix operation than an strict SLA may not be as critical for you or the customer.
I just heard a analyst say that in a recent study they determined that 85% of time in fixing a problem is in the identification and start of remediation. Only 15% of the time was actually fixing root cause.
A properly written Service Level Agreements will probably help the MSP manage and scale their business better, allow you to have greater upticks in service scope and upsell. I have seen first hand how a well defined SLA can eliminate the guess work out of your teams decision of how and when to take action on a particular incident.
I think MSP CEOs need to look in the mirror and ask: “What do I want this business to be when we grow up?”
SLAs certainly are important, but the approach will vary greatly based on the size, geographic reach and expertise of an MSP.
Small MSPs seem worried about getting burned by agreements they can’t fulfill. (Then again, that’s the sign of deeper problems at the MSP…).
I think your contract should be consistent with your marketing. If you emphasize and quantify during the sales process how your monthly fees are justified by an offsetting reduction in downtime, then you owe it to the client to actually measure downtime, and the client deserves some compensation if you fail to deliver on your promise. On the other hand, if you position your service from the beginning as one that is delivered on a best-effort basis, then an SLA will probably not be necessary. It really comes down to what the client thinks they are buying. Are they buying a service/utility that can be measured in terms of on/off, working/not working, and early/on-time/late? Or are they hiring a group of experts to work for them part-time? The answer to this question has implications not only for SLAs, but also for pricing and other aspects of the relationship.
I am in the process of launching the MSP practice and would like to see what others around the country are offering as their SLA. I am in the Houston market and would appreciate some insight from anyone who might wish to help me. My phone number is 281.870.9050
and my best email contact is [email protected]
Thanks for your help in advance
You can reach me at joe [at] ninelivesmediainc.com if you want any thoughts from MSPmentor.
Joe Panettieri
Editorial Director
MSPmentor