Lessons from a Bad MSP Trip
August 8, 2011 Dan Burger
A few weeks ago, I wrote an article about hosted services in the IBM i market. Although there are many offerings, the only area where there’s a noticeable customer interest is in data backup and recovery. You can find plenty of vendors offering everything from exclusively backup to disaster recovery and high availability. Competition is good, but comparisons are not always easy. Services and prices vary and sometimes surprises are not pleasant.
The perception of managed service providers is pretty good overall. There are many more companies that are pleased than those that believe they were mistreated. One of the mistreated companies came to my attention recently. It makes a good example of what can go wrong.
I’ll mention no names and completely avoid the “he said, she said” can of worms. Doing so does not change the lesson to be learned. So how do you get the expected value from the agreed upon price?
In the circumstance I was told about, the company was expecting to get a managed disaster recovery service. What it got was a lot of trouble: multiple backup losses that resulted in reseeding data on both iSeries and Windows servers. A lot of time and effort was expended, and in the end the contract with the managed service provider was canceled and the search for a new MSP began.
I don’t know any more of the details of this failed project. But I’ll tell you what I think MSPs should deliver. They need to deliver project management skills. There needs to be continual communications on project status and especially on potential problems. A vendor who calls itself a managed service provider better be big on service.
Buyers of a service better be sure the vendor they choose provides good service. The number of MSPs has expanded rapidly in the past few years. There will be a shakeout coming and a primary reason for some companies leaving this business will be an inability to provide service. Applying the bigger is better rule doesn’t substitute for getting references from companies of similar size and business requirements. It’s just as faulty to assume big vendors don’t do well in the small to midsize market.
Buyers should also beware of basing decisions on the lowest price. Low prices can help quickly develop a customer base, but it can also indicate a vendor that is cutting corners on infrastructure and services. One of the price point issues to pay attention to is testing. A disaster recovery system has to be tested. Four times per year is a best practices standard. Companies that don’t test get bit the hardest when unplanned downtime comes calling. Some managed DR providers have a specified number of system tests built into their pricing. Others charge separately for each test and that is over and above the monthly contract price. Closely compare how vendors describe testing. There can be considerable differences. Anything less than a test that proves your business can run off a target machine is not indicating DR readiness.
You have to solve a problem before you can talk about the value you attained.