Let's get right to the point. TEN reasons not to replace legacy systems from a CIO's point of view:
1. It doesn't do what I want, but no replacement will either,,,
2. I really don't know what it does now, so how can I possibly replace it...
3. I don't relly know what the users want, and I am afraid to ask...
4. You can see the result of the last time I used a rigorous replacement selection process ...
5. If I replace one piece I will have to replace everything. With so many integration and mini (and not so mini) "workaround" apps, it will collapse like a house of cards.
6. I have implemented an SOA to allow easy application replacement; it has been about as successful as my EAI initiatve 10 years ago.
7. I know what I want, I know what I need, but I have neither the business analyst or solution architect talent to put it into place. And... I either can't afford or don't trust system integrators.
8. I am 52 years old and I think I can keep this thing going for another 8 years.
9. CIO's are supposed to be strategic. How can replacing a system be strategic? Maybe our our strategy is perfectly suited to our current systems.
10. Let my competition replace their legacy! I will be reeling in the customers when they do!
There is TWO reasons to replace them:
1. Eventually we will have to, and best do it when the situation is not dire (dying hardware, vendors going bankrupt, surprise competive moves by competition, new regulation).
2. Legacy IT is like a legacy business model - the only companies that can afford them have competitors that don't really compete (oligopolies, monopolies...). Eventually, someone will smell the opportunity and strike.
A FEW observations from my point of view,
1. One way or the other it is taking a bet.
2. With this in mind, portfolio management as applied to IT helps a CIO think through this morass.
3. Ask yourself, how important are IT investment decisions and the flexibility to change investments to corporate strategy?