How to Kill an IT Project in 10 Very Easy Steps
By David Clark
October 6, 2008
Want to join the ranks of the IT-project challenged?
Want to become an oft-cited statistic in software selection horror stories?
Well, here’s how to get into the “Our-IT-Project-Went-South-and-I-Don’t-Know-Why” Hall of Fame, in 10 very easy steps…
1. Fail to get user buy-in.
Users, bless their souls, just won't use an application they don't want to use. Especially if they feel it's been foisted on them by a team who doesn't understand their needs. The result? Workarounds. And a system that only works on paper.
Getting user buy-in is crucial at the requirements-gathering stage, as well as during the rest of the selection process. The best way to get buy-in from a group is to offer them a voice—and proof that you're listening.
2. Fail to get executive buy-in.
Going maverick, are you? If senior management doesn't explicitly buy into the need to implement a new enterprise application, you will not have the backing you need to make the difficult decisions all major projects inevitably require. And that can translate into an incapacitating lack of room to maneuver. Head, meet chopping block. Chopping block, meet—whoops!
3. Automate the wrong business processes (AKA "Look Ma, I'm doing the wrong thing, much, MUCH faster!")
The benefit of implementing an enterprise software application (such as enterprise resource planning) is not simply that you are automating your business processes, but that it forces you to examine how—and why—you do what you do. Make sure you map out your existing business processes to ensure they correspond to real-life requirements. If they don't, for goodness-sake don't automate them.
4. Over-express your requirements
You don't hear much about this one—in fact, the accepted wisdom is to enumerate and prioritize every one of your requirements. Which only makes sense. But sometimes you can get bogged down in the way you do things (see the point above). You can also get bogged down in costly (and often unnecessary) customization. Don't forget, many software vendors spend years—and tens of thousands of labor-hours—in determining best-practice processes for your industry. Take advantage of that.
5. Succumb to undue influence
Is a certain someone on the board whispering in the ear of your CEO? Is that someone whispering something about, oh, I don't know, a certain software vendor? Is that vendor suddenly on your very very short list? There's no nice way to say this, so I'll just come out and say it: Cover. Your. Assessments. Use a software selection methodology that provides full accountability through each step of the selection process.
6. Let software vendors set the pace in software demonstrations
Two words for you: demo script. Draft one. Send it to the vendors on your shortlist. And make sure they understand they will be penalized for not sticking to it.
7. Look for the magic bullet
ERP will boost your bottom line! ERP will cure what ails you! ERP will vacuum under the couch cushions, too, and dunk a tastefully-rendered miniature umbrella into your drink.
If a vendor is trying to sell you on a magic-bullet* cure-all "solution"**, you may wish to probe a little deeper.
* Hint: there isn't one.
** An ERP application is not a "solution," in spite of what all those glitzy white papers may say. An ERP system is a tool. Have you ever tried to hammer a nail with a "solution"? "Hey, honey! Pass me the solution, wouldja"? Solutions are your department—given the right tools.
8. Set unrealistic deadlines
This is a sure-fire project killer—nothing says "bad decision" quite like stress and haste. Make sure you've allotted enough time to conduct a thorough software evaluation. I recommend you carefully consider each step of the process, from requirements-gathering to go-live, and ensure every stakeholder is aware of—and respects—the key project milestones.
9. Compare apples to oranges
So one product spits out reports only on Tuesdays during months ending in "R" or "Y."
And the other will get you stats whenever Britney Spears' publicist issues an unspun press release.
Which one are you going to choose?
Make sure your requirements are well-defined—asking vendors poorly defined questions can lead them to give answers that are too vague to be useful, which makes it difficult to compare them to answers from other vendors.
10. Overlook the fatal flaws
All businesses have a few essential requirements that do not appear in some (or most) software applications. These items are non-negotiable. And if the packages on your shortlist don't fulfill these requirements, they're fatally flawed relative to your specific needs. For instance, if e-business is part of your strategy, proper security may be a fatal flaw.
Don't think this could ever happen to you? If your requirements are not explicitly defined, it can. A vendor may claim to meet all your requirements—in all good faith—but if you're saying tomato, and the vendor's saying tomahto, you may end up with, uh, egg on your face.
Want to find out how a software selection tool works? Take TEC's eBestMatch for a spin. It's fast. And it's free. No, wait, I lie. It'll cost you the time it takes to type in your e-mail address.
Originally published December 6, 2007