Skin In The Game November 9, 2009
Posted by Chuck Musciano in Leadership, Technology.Tags: Cloud Computing, Customer Service, Operations
trackback
With clouds on everyone’s mind these days, more and more CIOs are beginning to consider cloud-based services. There are still a lot of concerns with this, depending on the system or service you seek to move to the cloud. In particular, what happens when the cloud goes down?
When negotiating with cloud service providers, the conversation inevitably turns to service level agreements. Typically, a vendor will promise some level of availability, with some prorated refund if the service is unavailable for an extended period of time. Thus, if a service is unavailable for more than 24 hours, you might get one-thirtieth of your monthly service fee refunded. Less than twenty-four hours? You might get nothing at all.
Does anyone, except for the service provider, think this is a good deal?
The cost of an outage is not the actual cost of the underlying service. The cost of an outage is the value of the business impact you suffer. If your e-commerce platform goes down for an hour, costing you $100,000 in sales, you should get $100,000 from your service provider. Needless to say, when you mention this to potential providers, they tend to get a bit defensive. “You can’t expect us to fully reimburse your lost business, can you?” Well, yes. Yes, you can.
If your service is good enough for a client to bet their business on, they’d expect you to have some skin in the game. If you aren’t willing to put money on the table that says you are as good as you claim to be, why should they be doing business with you? Does anyone want to be the CIO that, while explaining a multi-million dollar outage to his board, concludes with “but we got a check for $1,200!”
What is baffling is that this would be an easy guarantee for a qualified vendor to make. Hedging risk against failure is an actuarial problem. Why wouldn’t a vendor purchase an insurance policy against just such an occurrence, in an amount that would cover the exposed risk up to a certain point? Roll the insurance costs into the service fee and proudly market your “Million Dollar Guarantee” far and wide. I suspect you’d get some business. I also suspect that you’d get really good at providing exceptional service.
A lot of CIOs are naturally reluctant to deal with service providers who refuse to share the risk equally. Vendors who find a way to put their money where their mouth is will gain the respect, and business, of discriminating CIOs.
[tweetmeme source=”EffectiveCIO” alias=”http://bit.ly/cio129″ only_single=false]
The Effective CIO 
Cloud computing should be part of a mix of services a CIO will use. Moving 100% to the cloud is probably unlikely for most of the reasons you mention. At least today. Leveraging hosted, SaaS, dedicated on-premise / VM, and cloud environments in the answer. The CIO should place critical systems (where outages cost millions) in the environment that works best for his/her situation. Don’t put your trust in the glossy brochure or the penalties appendix of your contract. A good operations team has a pulse on complexity and stability of their environment and can identify the appropriate solution based on your data. In some cases, the cloud might be far more stable than your SaaS or on-premise system! Go with the traditional balance of cost, risk, and experience. There’s a fit for the cloud in every company. The CIO needs to find out what is right for them.
At the end of the day few service providers are going to make the sacrifices necessary, or make the hard decisions that sometimes cost significant $, to resolve a major outage in the quickest manner possible. The SP is constantly weighing the cost benefit of the contract with the client, the client is always weighing the cost benefit with business continuity. Priorities are misaligned at the start.
For my part, I keep as much internal control over critical services as possible. I staff with people who will make decisions based on my firm’s needs, not on the language in a contract.
This isn’t always possible, but where I can, I do.
Chuck – agreed on the SP insurance. I’d still say though that a relationship managed by an SP contract, versus one managed by an employment contract, is more complicated and performs at a less efficient level, regardless of the guarantees.
Great post CM!
All IMHO,
David.
Thanks for a great post Chuck. The prorated refund mentality is what will surely put a cloud provider at the end of the consideration list. Honestly, if a provider can’t spend the time with a potential customer to develop an understanding of what a hour-long outage during a Thursday afternoon means to a retail business, then they should just quit now.
Also, agree with the insurance idea. I’ll have to suggest it to one of my specialty insurance clients 🙂
For another angle on things hampering cloud adoption, here’s a thought on the top culprit: The Biggest Barrier to Cloud Adoption
-Chris
[…] there! If you are new here, you might want to subscribe to the RSS feed for updates on this topic.Skin In The Game by Chuck Musciano on The Effective […]
[…] Skin in the Game and Never Secure Enough – Chuck Musciano (cloud SLAs and security) […]