Banking

IT Projects Up Against "Penny Wise, Pound Foolish"?


In response to a previous article (Geeks and Communications Skills), commenter Tony Hopkinson wrote:

"Now maybe I'm not business aware enough to realise why short term gains are preferable to long term success, may be one of these business types should explain it, then we can all row the boat in the same direction."

DirtClod also points out the connection between IT projects and research grants:

"Scientists who communicate get GRANTS.  IT people who want the boss to GRANT funding for projects might find the time to learn our language."

The two ideas are not unrelated. Tony’s complaint is a common frustration in IT. Even the best proposal in the world that shows great ROI numbers, productivity gains, reduced downtime, and all of the other things that a good IT proposal should show can be turned down. One would think that "the suits" of all people would jump on a chance to increase profits after overcoming a substantial upfront cost. After all, this is why companies build factories, invest in training, outsource workers, and so forth.

Unfortunately, increased profit is not the actual goal of many companies, particularly publicly traded companies. The actual goal of these companies is "to increase shareholder value." "Shareholder value" directly translates to "stock price." Look at the compensation plans for C-level executives (CEO, COO, CIO, etc.). Their bonuses are tied more directly to stock price than to profit, market share, revenue growth, or any other direct financial metric. In addition, a significant portion of C-level compensation is in the form of stock options. Additionally, the management of a publicly traded company has what is called "a fiduciary responsibility to the shareholders." That means that they are held accountable to the shareholders, not to the employees or customers. To put it in a more obvious way, "it is in management’s best interests to increase the stock price of a company at the expense of any other metric."

The end result is that a company acts in whatever way will be best for the stock price, which is not always best for the company. To make the situation worse, a C-level executive typically does not stay with a company for more than a few years. They have little incentive to worry about the long term health (or even the long term stock price) of a company. Shareholders now rarely hang onto a stock for the long term; they are looking to "buy low and sell high," which is hardly a recipe for having shareholders that care about the viability of a company’s business model or the sense of their business practices.

It is true that things like profit and loss, revenue growth, market share, and so on and so on play a role in the stock price of a company. This information gets released to the public once a financial quarter. IT projects, sadly, tend to be capital expensive. The inefficiencies that IT projects resolve are not line items on any spreadsheet though. Let me give an example. At one company I worked for, they computer that I was assigned was a Pentium 1 with 128 MB of RAM, running Windows 95. This computer was expected to be running the following applications throughout the shift: Microsoft Excel, Microsoft Outlook, Microsoft Word, Internet Explorer (two windows minimum, one of which contained a Java applet), a custom built Java application (using a non-standard Java VM, so I would have two Java VMs open at any given time), McAfee Anti-Virus, and a telnet client that was extremely heavy (it had all types of scripting, macros, etc. built in). Microsoft Word would be open about half of the time, as well as a few other applications. By the way, this was in the year 2003. To the best of my knowledge, that PC is still in use. This setup was so unstable, I had to start the applications in a particular order or face system lockup. Everyone had to come onto the shift 15 minutes early as well. So the company was paying 15 minutes of overtime per day, which comes to 65 hours of overtime a year. Even at our pittance of a wage, this inefficiency would have paid for replacement PCs within a year.

Why were those PCs never replaced? For the same reason that many IT projects never occur: the short term budgetary impact would have affected the stock price more than the efficiency gains. Inefficiencies simply do not show up in reporting figures. There is no line item in a quarterly report that says, "excessive staffing due to a process not being computerized," or "additional payroll because computers are slow," or "missed SLAs due to poorly trained users." In comparison, "three month project to automate a process," "replacement and upgrade of existing PCs," and "training classes for users," are all line items that Wall Street (or London, or Tokyo, or wherever) sees. It is sad, but it is true. This is why a great project can get shot down. The initial upfront cost is just too high and the ROI just does not come fast enough. I am not going to pretend to know what numbers "the suits" have in mind when evaluating an IT project, but off the cuff it seems to me that you need to deliver a 25% savings within a quarter of project completion, and 150% within one year of project completion on a one quarter long project to get approval.

This mentality hits IT everywhere you look. Projects get rolled out as "betas" that never seem to get finished because of it. Programs get written in "quick and dirty" languages like .Net that are just not appropriate for certain types of projects. In reality, if you want to write a highly scalable Web application, writing it in C++ in a CGI environment is your best bet. But now one will ever get approval for that project, because good C++ coders are expensive, and the project will take forever because you would need to re-write much of what JSP, PHP, and ASP already handle.

The only way to get the funding you need for your project is to do exactly what DirtClod says to do: learn management’s language. If you cannot show in words that management understands, in a format they understand ("Gentlemen, start your copies of PowerPoint!") that your project will more than pay for itself quickly enough to be palatable to the stockholders, it simply will not fly.

Do not think that this does not affect private companies, either. It does. Private companies with a small amount of ownership (a few partners) have a tendency for the owners to treat every dollar in the company as their own, which is understandable. The $10,000 you want to spend on servers is viewed by the boss as half a year’s tuition for his child’s college. Many, if not most privately held businesses think ahead to selling out or going public. A history of cost management helps them get the most return on their initial investment of time and capital. No matter how you cut it, IT projects that cannot be shown to have a big, quick ROI kicker are just not going to be approved, regardless of how well you write the proposal. I will discuss these kinds of "low hanging fruit" projects soon, stay tuned.

J.Ja

About

Justin James is the Lead Architect for Conigent.

0 comments

Editor's Picks