23 February 2026
Forecasts that survive a dull reading
If the only way the capital buffer holds is the upside case, the application does not have a capital buffer.
E-wallet volume forecasts are sensitive to three numbers that founders like to keep optimistic: conversion from download to funded wallet, average stored value, and the share of value that sits overnight rather than passing through. Move any one of them toward the median of a quiet month and the float story changes. Move two of them and the capital story often changes with it.
A dull reading is the point. We take the commercial case as the upside, then rebuild a base case from slower onboarding, lower stored value, and a merchant payout cycle that does not always complete on the advertised day. Capital has to cover set-up costs and early operating losses in that base case, not only in the deck.
Related-party receivables deserve a special mention. Counting a due-from-shareholder balance as if it were cash is a habit that does not survive the first clarification. If the money is not in the applicant’s account, it is not paid-in capital.
The workbook we hand back is rarely prettier than the one we were given. It is labelled. Each assumption has a range. The capital bridge can be explained in a meeting without a second screen.