Do what It takes to succeed with IT projects. Rule one: don't make the same old mistakes

By William Seidman Why do companies keep making the same mistakes with IT projects? A colleague has been asked to help deploy a new financial management system, which is meant to replace several legacy systems. His client has set an aggressive deadline and hired a lot of resources from the vendor. They talk a lot about the business units requirements and are coding madly to try to get it done. They have three months to go — they don’t have a chance of being successful. They have no common purpose, a minimal plan and hardly any management infrastructure. They are trying to make this huge change while maintaining business as usual. There are only two ways these types of large, complex IT projects work, and companies consistently reject either approach. 1.  They can implement the vendors system with NO customization or adaption. I mean literally zero coding and accepting all of the system defaults as is. (This is actually the best way, but no one will do it because it means that the business has to be run according to what the software can and can’t do. Managers hate that!) We hear it so often – “You mean I have to run my business not by what the business needs, but only in a way that is acceptable to the IT system?” 2. They can really take the time to build a management infrastructure where there is clear focus on:

• ·         aligning on a purpose
• ·         getting organized and
• ·         NOT coding until everyone knows what to code

This is, inevitably, unpopular: it isn’t coding and doesn’t feel like progress, but it is the next best way to do IT projects. Even then the specifications are driven by a simple rule – do the MINIMUM possible to run the business. Only minimum requirements are possible which, again, is unpopular, because at first release, the system has less functional capability than the old system. Which brings up the objection, you mean that we put in all of this effort and it doesn’t work as well as the old system? Suffice to say, large complex IT projects are really hard to do. Few people really know how to do them, and most organizations resist what it takes to succeed.]]>