When executives talk about succession planning, they mainly limit the discussion to C-level staff (themselves) and who should succeed them if one (or more) of them get hit by the proverbial beer truck. Real worst case scenarios of the past include almost the entire executive staff of a major company in a car that was hit by a train. But that kind oif event is truly rare, and some companies still limit how many execs are "allowed" on the same flight itineraries. But sucession planning limited to such a small group of the enterprise population really belies a failure to understand what the BIA data should be telling them.
In a prior life, I was a Quality Project Manager in the defense industry. When the Cold War ended, my company was nearing the end of a huge contract, and announced a "golden parachute plan" for staff nearing retirement age. Among the most motivated takers was an engineer who designed a highly successful system used on Navy warships. He was also, however, the only person who fully understood the detailed architecture of the system and the antiquated programming language that was still running those systems, more than 20 years old. And the Navy was buying more of them. Engineering and Program Management executives realized they need this guy arround, but they couldn't just deny him the retirement package. So, they made him a lucrative offer to keep him around long enough to capture his genius on paper.
But that was an anomaly: few comapnies ever take a serious look at epople below their immediate reports. And that is a serious mistake. Companies, especially large and growing ones, have people who are running some process or operation that only they know how to run efficiently and/or properly. They probably don't think of themselves as irreplacable, but then, niether do thier bosses. That thinking needs to change: go back to the BIA data and see if it includes the names of people wo are deemed "critical." If not, perhaps it's a good time to scrutinize the BIAs and find those for the most critical processes, especially those that drive the revenue stream, customer satisfaction, and stakeholder and stockholder confidence. Then start looking at strategies to prtect against their loss, just like they were a mainframe running the financials.