Ask most executives where automation should start, and they’ll point to the most visible process in the business — the one that shows up in the board deck. Ask where automation actually pays off fastest, and the honest answer is usually somewhere quieter: a shared services center, a procurement queue, an onboarding pipeline. Nobody outside the building talks about it. It just runs, constantly, at a volume that makes small improvements compound fast.
The uncomfortable part is that volume alone isn’t the deciding factor either. A process can run thousands of times a month and still be a poor automation candidate if every case is different. The processes where the ROI case is genuinely clear sit in a specific range — and it’s worth checking, honestly, whether yours is in it before investing in anything.
Why volume without repeatability doesn’t work
A high-volume process built on inconsistent inputs, undocumented exceptions, and ad hoc judgment calls will resist automation no matter how many transactions run through it. Volume just means you’ll hit every edge case faster.
Why repeatability without volume doesn’t work either
A perfectly standardized process that runs 40 times a month rarely justifies the investment. The math on hours recovered or errors avoided doesn’t compound enough to matter to the P&L.
A short self-check
Before sizing an automation business case, it’s worth answering these honestly:
- Roughly how many transactions does this process handle per month — invoices, applications, orders, tickets?
- Is there a named process owner who could describe the workflow end-to-end in one conversation?
- Do the systems involved already talk to each other, or does someone manually bridge them today?
- Are the exceptions well understood and finite, or does something new come up every week?
- Would the team managing this process describe it as “the same thing, over and over” — even if it doesn’t feel that way day to day?
For sales conversations: use this as the opening question, not the close
If a prospect can answer the volume question but not the systems or exceptions ones, that’s not a disqualifier — it’s the actual diagnosis. The first call should size the process, not pitch a solution.
Reading the range
Below roughly 1,000 transactions a month, the return rarely justifies a dedicated automation project on its own — it may be worth bundling with a related process instead. Above roughly 10,000 a month, the conversation usually stops being about a single automation and starts being about a broader operating model for that function. In between is where a well-scoped project tends to pay for itself fastest, because the volume is high enough to matter and the process is usually mature enough to be well understood.
None of this requires a vendor to answer. It’s worth running this check internally first — with or without outside help — before deciding what, if anything, to automate next.