I’m a member of the Saugatuck Technology Cloud Research Board which, in their own words;
is a broad-based community of interest focusing on Cloud Computing in all of its aspects. Membership in the Saugatuck Cloud Research Board is open both to vendors and end-users, who share an interest in this evolving phenomenon
The good folks at Saugatuck sent me a report the other day which is well worth a read for those in larger enterprises looking to make the move from hosted to cloud computing solutions. The report can be purchased here but I thought I’d give a precis.
Saugatuck found that the recurring motivator for user intention to move to cloud based solutions was the expectation of significant cost reductions. I’ve long said that selling SaaS and Cloud Computing based on their cost implication was a sure road to disaster – instead vendors should be selling the value add of using these services.
Saugatuck’s research found that cost benefits were expected due to cloud computing’s inherent economies of scale and ability to scale up/down at will and hence not pay for excess, but unused, capacity. Saugatuck however pointed out three reasons that cloud computing could be expected to cost a premium to a similar in-house solutions;
- Price of convenience – Saugatuck used as an analogy the price difference between convenience store and grocery chains
- Vendor overheads – the fact that cloud vendors have a full set of overheads and hence there may be duplication in the overall expenditure
- Vendor profit margin – obviously vendors need to return a profit which internal IT departments do not
Saugatuck have come up with a nice modelling table looking at different scenarios – I’ll not reproduce it here but suffice it to say it provides some clarity for those looking at a move to the clouds – of course that clarity looks at only one issue, the pricing one which we have contended is over-emphasised.
The bottom line, as always, is due diligence. Saugatuck have estimated that the same amount of compute capacity rented on-demand for a month may cost up to 4 or 5 times as much as a similar capacity traditional hosted. I’d love to see the maths for that calculation – but either way it shows that cost is a thorny subject.