Just when we thought there was nothing but good things to say about legacy systems modernization, along comes a report by Celent’s Craig Weber based on a recent survey of industry participants. It’s a needed dose of reality, we think. Here’s some of the headline information from the preliminary results (thanks, Craig, for sharing them).
First, there is a disconnect between vendors’ and practitioners’ perceptions of the effort required for, and results from, modernization projects. For example, 68% of vendors have a ‘rock solid’ belief that their clients will achieve hard dollar savings from such projects, whereas only 23% of insurer respondents have the same level of confidence. There are similar disconnects with regards soft dollar benefits and success rates.
Reading some of the literature in the past few years, one would tend to believe that the old surround and supplement strategy (e.g., building new front ends to legacy systems) had gone away. The Celent results suggest that’s not quite the case. While the majority of respondents feel that a full replacement of legacy technology would improve functionality more than the surround strategy, the latter was competitive in certain functions, such as supporting new service models or new distribution channels.
So, does this mean that insurers have become less concerned about legacy systems? Hardly. When it comes to support, over 50% of respondents feel that the lack of developers skilled in legacy technologies (COBOL, Assembler, etc.) either is a critical problem now or will become so within 3 to 5 years. More significantly, one in six (17%) believe it is critical now.
But when it comes down to it, the respondents don’t blame technology, or resources, or weak business cases for the challenges in legacy modernization projects. When asked what was the factor which contributed to modernization projects under-performing, the most commonly cited (by 64%) element was: ‘difficulty in changing the culture’.
“We have met the enemy, and he is us.” -Pogo