Insights and opinions on Information Technology from George Tomko, a veteran CIO turned renegade consultant.


Killing Ants with Anvils

Problem management. Incident management. Issue management. They are all names for the process of taking something that is “out of whack” and somehow making it get “back into whack”.

Unfortunately, this is an area where business professionals, especially IT professionals, perform miserably. We’ve all heard the slurs i.e. “I have got to call the ‘helpless’ desk” or “…I hate dealing with those geeks in IT”. It goes on and on, day after day, year after year, as regular and as sure as the Sun comes up in the morning.

Of course, the IT folks have to deal with their “BDUs” (basic dumb users) and users have to have their cheat sheets listing out every keystroke and one iota of difference in what shows up on the screen means, you’ve got it, a call to the ‘helpless’ desk. The “solution” will be a ticket number which implies that some unknown, unseen, geek is going to do some unknown, unseen thing to make things work again. Or, the command will come to reboot whether there is any indicator or not that prescribes such brutal action and the wait time associated with the shutdown and start-up sequence.

As futile, perhaps, as the user who keeps clicking their mouse button or hitting the enter key over and over.

What I have ultimately come up with as the CAUSE of all of this mess are ants. Yes ants. Let one ant live here and one ant there and pretty soon, lots of them. And you thought that a computer “bug” was just a made-up term.

Anyway, the secret is to never let an ant live a nanosecond longer that the moment it shows up. Unfortunately, ants are very resilient, pesky creatures that can lift 10 times their own weight and can fit into nearly microscopic places. They run fast, too. So, you have got to take decisive action with a guaranteed outcome – the ant must die and there must be no doubt about it.

That’s where the idea of the anvil comes in. I have (unscientifically) measured a 100% kill rate when I drop an anvil onto an ant. The ant cannot run fast enough to get out of the shadow of the anvil and the weight of the anvil not only kills the ant but buries it too. In short-that ant is not going to bother anybody again.

This takes me back to problem, incident and issue management. The whole thing implies a process, giving tickets, counting, categorizing and storing what are essentially ants! But, it doesn’t do a good job of killing ants. Whether it is due to complex processes, roles and interactions of highly specialized and narrowly focused geek squads or waves of ants that overrun the ability to kill them one-by-one, there are just not enough anvils being thrown around.

The bottom line in the business of delivering and supporting IT services is to kill ants with anvils. Knock ‘em dead right then and there. Pursue each one to the ends of the earth.

Not doing so means a loss of the killer instinct. I remember, early in my career, when I was a systems programmer, supporting the operating system of a large mainframe environment for a regional bank. When the systems were running slow or some reports of problems started coming in, we mobilized like a surgical team. The stakes were high: there were hundreds of users and many people performing financial transactions, so time was money.

We would open up the live system and try to find where a bad value in memory was causing the problem. We would carefully consider our options (and rebooting was the last on the list) and the bravest of us would, using the master console, “zap” the memory location with a new value.

We basically threw an anvil at it. Then, crisis averted, we all went back to our offices and got back to productive work.

We could have rebooted and the system would be back up in an hour, not available to anybody. And, ultimately we would not have any insight into what the actual problem was. Instead, we would have a problem ticket open and wait for the dreaded event to happen again – more future ants would be coming because we didn’t kill the one we had in our sights.

Of course, we apply this to any problem we face in the world. Remember the TV commercial for oil filters where the tag line was “You can pay me now or you can pay me later”. How many things that start small do we let grow into big, ugly monsters? In our personal life it might be ignoring the slow leak in a tire; running your car below the “E” indicator on the gas gage; bypassing OS patches and updates on your PC; disregarding warning messages.

In business life, it might be an issue with a late shipment from a supplier for equipment that is needed to make the deadline that was not tracked earlier when something could have been done about it. It might be a boss who does not coach an employee about undesirable behavior the first time it is observed and 5 years later is involved in a messy termination of that same employee for the same behavior.

It is all about taking care of things when they are manageable, visible and resolvable.

The next time you see an ant, drop an anvil on it. You won’t ever have to worry about that ant again.