I read the piece in New Yorker titled “Why Doctors Hate Their Computers?” and enjoyed it very much. It is by Atul Gawande who is a surgeon and an author. It describes the issues we all face every day – technology is changing fast and we want our respective communities to adopt them, but it is a monumental challenge. I am of course simplifying it, but thats the crux of it. There is one thing in the article that stuck with me – “Mutation and Selection”.

Basically the author compares how the medical profession operated under a very different paradigm early on, where, every physician basically operated independently that suited their particular modes of operation. This is mutation part. Electronic medical record (EMR) systems tried to bring standardization, better sharing of information amongst the physicians and most importantly, gave access to information to the patients readily. This is the selection part. Obviously this is not a trivial adjustment for those who operated independently and the fact the EMR systems, which are in their infancy, are not optimal. At least not yet.

Higher Ed institutions face exactly the same issues. The whole issue of centralization of systems is the “selection” part and the proliferation of multiple systems (Best of Breed) is the mutation part. What is the right balance between the two is so complex and dependent on the institution. But, the article describes how a neurosurgeon and his team is trying to “mutate” the “selection” system (EMR) so that their needs can be accommodated. This is what we would call customization in the old ERP systems, which turned out to be a terrible idea for a variety of reasons. However, in the more modern systems, such as Workday or Salesforce, accommodations to mutations are much simpler to manage through “configurations” and “business processes”. This would be a “controlled mutation” of sorts.

But, whats the problem with supporting best of breed?

One of the major issues of best of breed is the word “Best”! It is in the eyes of the beholder. In general, the choices for best of breed originate from functional offices or influential people in the organization who have heard about or like a particular product. Exactly how it fits in with the rest of the institutional and IT strategy, what are the costs related to integration, sustainability, and data security and controls all become an afterthought. In most cases, a handful of enthusiasts drive the adoption and look at IT as the villain when integrations either don’t work well or take too long to implement.

And, almost always the enthusiasm on the functional side declines because everyone gets absorbed in their day job and then look to IT organizations to own the systems. Such portfolio creeps are impossible to manage.

Vendors take advantage of this big time by directly engaging with the functional offices and promise the world and say they have APIs that make data integration simple or how well it works with the ERP systems etc. And of course, everything is GDPR compliant these days, but you drill deep and you will find holes. And you find them even with some of the most reputable large companies and therefore you can imagine the extent of this issue with smaller companies and startups.

Having a clear strategy and plans is essential to manage this. This requires a careful articulation of preferences, a method to assess the needs, a process for evaluating options (including existing systems and local capabilities for development), and finally governance, a group that will decide whether to move forward or not. Will this slow things down, of course, for all the right reasons! Will this be perfect, of course not. Because someone more influential in the governance committee may steer a decision in a direction that is not optimal. But it is unlikely to happen as often when such a process doesn’t exist.

The best of breed results in increased local satisfaction and efficiencies, almost always at a fairly large overall institutional cost. Centralized systems in comparison may have overall institutional efficiencies but at the cost of decreased local satisfaction and efficiencies. Finding the right balance is the key and the strategy and plan can help.

At Wellesley we are trying very hard to find that balance. We are small and highly centralized for technology support and the portfolio cannot keep expanding. We have to constantly remind the departments that they are lean themselves and hire staff for their domain knowledge and not for the staff to act as systems administrators. We have a strategy and a plan and so far we are keeping with it. We are also realistic and do not erect unnecessary barriers when we feel that a best of breed is the best choice for the institution. Even with all of this, some days feel like we are fast losing the battle!