Sponsor banks approve programmes one at a time, carefully. What gets examined less often is the position the portfolio has quietly produced: overlapping activities, similar customer bases, shared vendors, correlated failure modes.
Concentration is not an accident. It is the cumulative result of individually reasonable approvals.
Approve the marginal programme, not the average one
The question at committee is not whether this programme is acceptable in isolation. It is what this programme does to the bank's position given everything already approved.
That is a Regulatory Delta question, and we think it is answerable. The same activity added to two different portfolios produces two different consequences, because the institution on each side is different.
Exit conditions are part of approval
A programme approved without stated re-test triggers or exit conditions leaves the bank with a position it can only manage all at once. Setting those at approval is cheaper than negotiating them under pressure, usually with a fintech whose revenue now depends on the answer.
The part that remains difficult is judging correlation across programmes. That still needs experienced people. What should not need reconstructing every quarter is the underlying reasoning about what each programme changes.