When the IT budget is tight, good enough has to be enough

When IT resources are scarce and work needs to get done, stop sneering at easy-to-use end-user tools and start putting them to use

Dear Bob ...

Like just about everyone else, our IT budget has been shot to pieces. That's OK -- I've had to lay people off before and I've had to kill promising projects before. These things happen.

[ Bob Lewis has more IT management suggestion: "When does running IT as a business make sense?" | Get sage advice on IT careers and management from Bob Lewis in InfoWorld's Advice Line newsletter. ]

What I'm not used to is a team of executives (and I use the term "team" pretty loosely -- these people have no sense of teamwork at all), each of whom act as if my budget was just doubled. I'm serious. Every one of them is on my back to take care of his or her crash-and-burn project faster than we could have done it fully staffed.

I've tried explaining the facts of life, over and over again, and what I get in response is, "A high-school kid could put this together for me in Excel in a couple of days. Why are you telling me it needs seven people for six months?"

Any thoughts on how to bring a dose of reality into the discussion?

- Overwhelmed

Dear Overwhelmed ...

First, make a list. Sit down with each exec and ask them if they could get only one project done, what it would be. The rule for the discussion: They describe their projects in terms of business change, not in terms of software requirements ("We need to improve productivity in the warehouse by picking items more efficiently," not "We need an inventory picking system enhancement.")

Next, call a meeting with your business analysts. Walk them through the full list, then parcel out the requests based on each analyst's expertise and ability to get along with the various execs. In this discussion, let them know you're looking for quick solutions that are good enough, not elegant solutions that will withstand the test of time. Their job is to figure out how to get each exec most of the improvement they're looking for and quickly, not all of the improvements they'd like done the "right way."

This means that if a twice-a-day batch extract into Microsoft Access is good enough, there's no need to create a real-time SOA-driven interface. It means that a once-a-night dump-and-load into Excel might be a better answer than enhancing the data warehouse and its business intelligence interface.

It might mean nothing more than teaching their staff how to assign tasks to each other using plain-vanilla Microsoft Outlook, instead of deploying a full-blown, enterprise-scale integrated project management solution.

Very important: When talking with the business execs, the business analysts are to make sure they don't get distracted by any requests for system features. Their mantra is, "My job is to help you get the business improvements you need. The technology might not look like what you're expecting, but I will make sure you get the improvement."

Give your BAs a week to sketch out the basic shape of a solution.

Now comes the payoff. Get the execs, including the CEO and CFO, into a room with you and your business analysts. Lay out the full program:

  • The projects.
  • The kludgy solutions your BAs have come up with that will get everything done on the cheap.
  • The long-term risks associated with relying on a swarm of band-aid-and-chewing-gum automation.
  • Your estimate of the cost of cleaning up the mess after the crisis has passed.

Tell the CEO and CFO they have two choices: They can have IT take care of just one or two projects the right way or all of the projects the wrong way. If they choose the latter, recommend they start reserving budget dollars now for the effort you'll have to expend in a year or two cleaning up the kludge.

When they make their decision (which will almost certainly be to go ahead with the kludges), accept it and make it work.

We've all been saying for years that these are business decisions. It's time to make that real.

- Bob