These systems are already in place
The mainframes have performed efficiently enough to secure the right numbers and metrics for a business. User training and acceptance requires resources in terms of time and money and unless absolutely essential, they may not justify the spends. Peripherals, however, may be upgraded and could often provide a much more justifiable change.
Would a transformation bring about a reasonable ROI in terms of business?
This is a question sometimes less considered during transformation exercises; however, very often, upgrades to tried and tested legacy systems may well be all that is necessary. There exist both pros and cons as with every decision required for change, and both of these must be weighed in case a completely new software becomes something to be blamed for productivity problems.
Are all of the newly offered functionalities necessary for your business?
There may certainly be occasion when the functionalities are less than adequate and it is not advisable to “make do” because it has worked till a certain point. Legacy systems are not necessarily the only thing to blame in these cases, though, and upgrades to existing systems may be the only thing required for a boost, rather than an overhaul. The bouquet of functionalities that are offered by the new system may be tempting, but it is necessary to consider if all of them are actually important and if they translate to implementation.
Reskilling is an exercise in itself
It is also easier to find skills required to handle legacy systems that drive profitable output and this makes sense to several companies that understand that retraining and rehiring for systems upgrade purposes only to be rendered redundant in a few more years without a strong market motive may not make the best market sense. This is not to say that change is unnecessary, but that it must be tempered with business consideration being more important than the insecurity of missing out on newer technology offerings.
Legacy systems are not really going anywhere very soon, especially considering the data maintenance on legacy systems and the legal complications of upgradation. In other areas as well, Legacy systems already have supported the mainframe structures for long enough to be adjudged useful. The costs of replacing these systems is not a marginal increase; it is a manifold increase and unless measured against potential ROI, it may not be worth the risk. According to Dharmesh Mistry who had previously been closely involved in technology strategy at Natwest and Lloyds TSB, "We felt that if it was not broken, why fix it? If you look at the lines of code involved in migrating to something new, why bother?"1
Justifying the need for more than an upgrade to the Legacy systems requires careful consideration of both pros and cons. The main question must of course always be, “Is it necessary?”. If it is deemed so, this must have strong rationale that ensures that the ROI justifies the change. Legacy systems needn’t scare any company, especially if they are being used by a large workforce that provides an output that may not be improved by a technology upgrade. As they say, if it ain’t broke, why fix it?