Every benefit claimed for accounts payable automation is either arithmetic you can do from your own figures or an assertion somebody else made about a different organisation. The first kind survives a finance director's questions; the second does not. Separating them before writing a business case is what turns a rejected proposal into an approved one.
The arithmetic benefit
Monthly volume, times median minutes per invoice, times the fully loaded hourly cost of the people doing the work, applied only to the share of invoices that could actually clear without a person. That last restriction is what makes it credible, because claiming a saving on invoices that will still need somebody is exactly what gets a case questioned.
The capability benefits
Shorter elapsed time, so terms are met reliably and accruals are less speculative. An approval trail that survives a departure. Visible exception ageing. A defence against duplicate payment and bank detail fraud. All real, none easily monetised without inventing a figure, and all stronger stated as capabilities than as dollars.
The benefit that is not yours to claim
A higher touchless rate than your purchase-order coverage allows. That ceiling is set upstream, and an automation case that assumes past it is assuming a change in another department's behaviour. If that change is part of the plan, say so and let it be resourced; if it is not, cap the claim at the measured number.
Questions people ask about accounts payable automation benefits
Can we use industry averages at all?
As context in a sentence, not as the basis of a calculation. The first question about an average is always whether it applies here.
What if the arithmetic is marginal?
Then the free upstream improvements are the project, and they may make the software case later. Both outcomes are useful information.
How do we measure minutes per invoice?
Time ten invoices end to end, including interruptions, and take the median rather than the best.