Best of Breed, Worst of Results
- Who really stands to benefit from best-of-breed software? Not you or your business. – By Richard White, Chief Executive Officer of logistics technology solutions provider CargoWise and technology development company WiseTech Global
Best-of-breed solutions sound good in theory, but just look at what happens in practice. If you’re not careful you’ll pay tens of thousands of dollars for a consultant to advise you on information technology systems. They’ll take your money, spend weeks, and sometimes months, assessing your requirements, only to come up with the advice they had probably planned to give you even before they walked through the door: buy best-of-breed.
It’s an easy recommendation to make. When someone says: “your organization needs the very best software in each category”, there’s not a single CEO who would argue. It’s also the most logical recommendation to make if you’re the consultant who will also need to tie all these separate systems together - at great expense and over a long period of time.
Of course you need the best - that goes without saying - but the assumption that an amalgam of the best software in each category will result in the best overall system is verifiably false.
The problem is not so much the software, but the time and effort it takes to hold it all together. When presented with the idea of best-in-breed, many fail to consider that systems might simply be connected together with the technological equivalent of glue guns, bailing wire and duct tape. Chances are, by the time you’ve realized this, you will already have invested so much time and energy in the project that it would be difficult to back out. As a result, the system will be delivered and implemented regardless of the dawning realization that it is not ‘best’, but expensive, poorly constructed and often dysfunctional at a whole-of-enterprise level.
Consider this analogy: Imagine trying to build a high-performance car using the top-of-the-line components from the world’s best automotive car and component manufacturers. If anyone actually managed to build it, the car would be unique, incredibly expensive, and totally unserviceable by anyone other than the original mechanic. Any support from the component manufacturer would be near impossible. If it functioned at all, the car would be subject to so many idiosyncrasies and problems it would be totally impractical for anything other than a collectable curio. It’s easy to see that the only people that can build cars are the automotive companies who can ensure that every piece works with every other piece.
So it is with software. We come from a world that, 20 years ago, had internal development teams who would build whole-of-enterprise systems from scratch. Today, whole-of-enterprise systems are so complex that they take hundreds or thousands of people working together with key customers and industry to build, maintain and continually upgrade, as many design choices can only be made with an understanding of the overall architecture.
It makes absolute sense that consultants propose best-of-breed systems. They require a lot more consulting work initially, and then a constant and expensive stream of maintenance and development. All this creates a steady stream of business for the consultant, and a constant stream of costs, risks and headaches for you.
Because of the way they are tied together, best-of-breed systems are only as strong as their weakest link; and a single failure in one system, or in one of the interfaces between systems, could bring the whole lot crashing down. When something goes wrong, as it invariably does with complex interlinked software systems from many different vendors, it’s almost impossible not to get caught up in finger pointing and buck passing between the suppliers who each support their own software only and the consultants and developers who support their interfaces.
There are management issues too. The differences between the software designs can exacerbate the silos that invariably spring up between different departments. Poor, incomplete or slow data transfer can result in miscommunication which can be costly, distressing and dysfunctional. Different departments may end up using different terminology for the same processes because they are essentially operating on different software systems, resulting in double entry, double handling, data errors, frustration, wasted time and resources.
If you’re offered or are considering a best-of-breed solution, stop and ask yourself: “what are the costs, risks and who really stands to benefit?” Chances are it won’t be you or your business, but the consultants who’ll come out on top.