The automatic pilot thinking during a merger says: lets take the best of company A and the best of company B to create the new company C. The small detail is that the best of A plus the best of B may not be good enough for C. To put it unkindly, the sum of the two “best” may still be crap! The focus should be: what the new C should look like? By all means, let’s take any goodness from anywhere where available, but here you’ll have (usually) an opportunity to build something above and beyond the past. If we don’t realise that, we may settle for an ‘obvious’ (and easier) common denominator, and this does not make it automatically good. Ok, it’s harder to think this way, I admit. Also, I would bet my money that, in 9 out of 10 cases, a resulting ‘C hybrid best + best’ is not a good idea. Hybrids are bad news. They still contain A people and B people and their endless references: ‘We in A did, We in B run…’ etc. Reboot the system. Make it brand new, shiny C. The best of the past may not be the future you need.
Main drawback is that ‘the best of B’ may not be as good as ‘the worst of A’, or at least overlap. One approach would be to select them blind, but you’d still run up against NIHS – the Not Invented Here Syndrome. Two examples of that from computer industry: ICL (now part of Fujitsu) had two operating systems, each developed in different locations. Too late management decided to ditch one. Similarly with Microsoft, when the two rival operating systems were Windows 98 et seq. and Windows NT, which was technically superior and more robust, but viewed as a techie solution. Surprisingly, MS chose the more sensible but less popular path, to develop that and drop the other.
This problem also manifests itself in areas such as TV/radio quizzes, where the runner-up in one week could easily have succeeded against the opposition of other weeks.