Newsletter April 2005

Pop Quiz

• How Not to Bet Your Business on New Software

For those of you who have never dealt with a software vendor, here's a quick pop quiz:
You spend $100,000 on software. The software causes your company to lose $1,000,000 in profit. You decide to sue. You should expect to get

1. $1,000,000
2. $100,000
3. $0

The answer -- most likely an amount between (b) and (c). Software vendors, unlike most companies, are not liable for the harm caused by their products.

At least not yet. A few weeks ago, the Wall Street Journal reported that large companies are starting to seek damages from vendors when bad software costs them a lot of money. This "news" may surprise those of you who don't negotiate software contracts. If you bought a new truck and the axle fell off, injuring people and property, you'd probably want some compensation from the guy who sold you the truck. But in the software world, expectations are lower. If your expensive software costs you millions of dollars in lost sales, the most you can generally expect is a refund. Maybe. If your fall season is shot, that's life.

Now, if I were an attorney, I might advise you on how to better negotiate your contracts. (For more on this, check out a friend's article on this topic in CIO Magazine). But since I'm not, I'll advise you on ways you can avoid losing a large chunk of business to bad software. Of course, you want to make the best choices when you start out, but for today, I'm going to discuss practical ways to avoid disaster when you're about to start using new technology that you've already chosen.

How Not to Bet Your Business on New Software

• Don't install too much all at once. Often, businesspeople replace multiple systems simultaneously, which is asking for trouble. For example, I know a retailer who put in a new point-of-sale system while changing accounting software at the same time, under the theory that "it's better to get through the pain all at once and quickly." Let's just say that the pain lasted over a year.
• Don't rely on outsiders bailing you out. In a rush to go live, businesspeople tend to count on their consulting teams or vendors to "get them through." Now, I have often been the beneficiary of such situations, but if you're smart, the best thing for you to do is see if you can use the system in test mode without any consultants standing by. Perhaps, several weeks before you go live, you could declare a CFD--a consultant-free day. On this day, you and your staff try to use the new system to do what you think it should do. If you're able to cut checks or print invoices successfully, you know you're ready to go. If you get caught on the first screen, it's time for more training.
• Don't be afraid to change your date. It's human nature to get wedded to a particular go-live date, and while I definitely agree that there's nothing like a hard deadline to keep people focused, the benefits of such a deadline are always outweighed by the risks.

Red Three Consulting: Transforming Information Technology into Answer Technology