Most HR teams put real thought into the Internal committee training. Far less thought goes into how long members serve, what happens at handover, who ends up sitting in that room at all.
In my work with organisations across sectors, composition turns out to matter more than curriculum.
Two failure modes, and most organisations have already chosen one by accident
Consider a committee that turns over completely every year.
Nobody recognises a name that has come up before. Nobody notices that a pattern in one function mirrors something that surfaced in another two years ago, because the person who would have made that connection rotated off before the second instance came up. Every case is treated as the first case, because for that committee, it is.
Now flip it. A committee that never turns over runs into the opposite problem. Give people enough time in the role and familiarity with how the business really works starts shaping what reads as concerning and what reads as just how things are done here. Nobody deliberately decides to look the other way. It happens slowly, and it happens to good people.
Both of these are decisions the organisation has already made, by default. Someone set the term length once, when the committee was first constituted, and nobody has touched it since.
A committee with no memory cannot see a pattern. A committee with no turnover stops recognising one.
This has stopped being a quality problem
For years, this was a question of how well individual cases were handled. It is now closer to an evidentiary one.
Under the Companies Act, every company already reports cases filed, disposed of and pending in its Board’s Report. For the top 1,000 listed companies by market capitalisation, SEBI’s BRSR framework requires the same numbers under Principle 5, with the previous year’s figures set alongside for comparison.
That comparative column is the whole point. A board signing off on those numbers is vouching, whether it realises it or not, for the competence of whatever produced them. And the moment two years sit side by side, the question stops being “was this complaint handled correctly” and becomes “did the organisation see what was building.”
A committee with no institutional memory cannot answer that. Not because anyone did anything wrong, but because the information was never held anywhere it could be found again.
Three fixes, none of which need a change in the law
Stagger the terms. Two-year terms, with members rotating off at different points, so the committee never turns over all at once. Memory survives without the group calcifying around a single worldview.
Write a handover note. Short, anonymised, from the outgoing members to the incoming ones. What kinds of cases came from where. What escalated fast. What the last group wishes it had caught sooner. Right now that knowledge walks out of the building with the people who leave.
Set a composition rule. No more than two members from the same function. An external member who is genuinely rotated, not the same name renewed for six years running. A committee drawn too narrowly develops exactly the blind spots of the business it is meant to be checking.
None of this is expensive. All of it is structural, which is precisely why it gets skipped. Nobody’s annual goals contain the phrase “committee term length.”
The test
Build the committee this way, and two years from now it can answer the question a board or a tribunal might actually ask.
Build it on defaults, and you will not find out that it cannot until someone asks.
Train the committee well. Then build it so that what it learns outlives the people who learned it.