Test P&L: would this survey have prevented a significant drawdown?
The fact that a survey of 78 people was filled out does not guarantee that the conclusions about document checks are actionable data for decision-making.
Such a small and potentially biased sample does not justify blind trust in a proportion of 75% or 25% that could have operational consequences.
If an investment is based on this data, the risk of slippage is real, because the elegance of data collection does not translate into real profit if the base is fragile.
For example, a document reminder service targeting the 25% could be a financial sink if these 78 respondents do not represent the market.