Published: February 14 2012, 04:59 PM | 1 Comment(s)
by Marcel den Hartog

Definition: Frugalnomics is an environment where buyers demand quantification of benefits, significant ROI, fast payback and superior value from each purchase.

As most of you will agree, the term "frugalnomics" describes pretty much what the world looks like today. There is a very well written blog by Tom Pisello for those who want to find out more about some of the drivers of frugalnomics.

What we do not hear much about is the risk management that should be a part of frugalnomics, especially when we discuss IT.  For the past 4 years or so, IT Managers were constantly under pressure to invest in new technology whilst IT budgets remained flat. This of course meant that the cost of managing and maintaining IT and the people who were responsible for it had to be reduced. It is not the first time that we have had to cut down the cost of IT ─ we've had recessions before. But it IS the first time that Information Technology is so pervasive, so embedded that it really is an important ingredient of our daily lives. Add to this the fact that IT is progressively more complex than 15 years ago, and a cost cutting exercise without the right risk assessment can be very dangerous.

So, be honest, what type of risk-assessment did you do when you decided to reduce the amount of people managing processes like Application Performance Monitoring, Software Maintenance & Upgrades etc.? Did you have the right data to support those decisions? And more importantly, did you realize the impact of this on other IT processes? My anecdotal evidence tells me we did not, and allows me to give you an example from the field.

A financial institution decided to reduce the number of people managing IT. The groups hurt most were the mainframe, software upgrade & maintenance and performance management.
The reasons:

• Mainframe: 'cut before, no real impact on the availability and SLA's so let's cut again'.
• Software upgrade & maintenance; 'we do not really need to do upgrades every month, twice a year is sufficient, lets reduce staff by 30%'.
• Performance management; 'they analyze problems when they happen, but mostly delegate to the right teams to solve the problem. Let's just keep the right teams'.

The real impact:

• Mainframe: yet another reduction was the point where everything started falling apart. Where people were flexible and knowledgeable enough to just keep the mainframe going, this was cut to the bone and every issue now triggers another one, and that now takes longer to fix because the "operational knowledge", or "corporate memory" suddenly has rather a lot of holes.
• Software upgrade & maintenance; a very procedural team, always in control, never a surprise. But emergency fixes that were previously implemented in a managed way, are now installed in an ad-hoc way. And slowly but steadily, systems are becoming less stable, small but irritating conflicts take days to solve and it takes longer for new things to go into production.
• Performance management; after removing some of the cross-platform performance analysts, every performance issue results in finger pointing again. But the worst came only after 2-3 years later. Not proactively monitoring application components combined with ageing hardware that is not replaced in time, results in unpleasant performance surprises. Implementing new technology (mobile apps) results in unpredictable workloads and unexpected loads which in turn break the systems. Results: breached SLA's, an unhappy Business and unexpected hard- and software upgrades (cost).

Is this company an exception? Definitely not. Could this have been prevented? Probably. But cost cutting exercises always happen when a company is under pressure, there is no time to really analyze the impact and do proper risk analysis and more importantly, we do not have the right data to start with. What problems do our people solve today? What is the cost of doing things today? What cost did I avoid when I decided to do proper performance management and trending The problem is that I'm not really sure, in the past when I did complete performance management I know I was able to solve things more quickly, possibly more cheaply,  but I didn't measure it so I don't know. But now, I'm still not sure why I am paying through the nose - although I was warned that the more I tried to reduce the expense of doing things - over the long term, cutting corners meant that problems just became more expensive. Risk Management is now more important than ever because we simply can NOT deal with the financial consequences of a cost cutting exercise taking a bad turn. An error, caused by a lack of a proper risk assessment that makes you lose 20,000 customers in the middle of a crisis is a very, very expensive error...