A low price per minute does not necessarily produce a low cost per useful outcome. Long calls, repeated attempts, incorrect bookings, and human rework can change the economics. The right comparison includes what happened after the conversation.
Define the outcome first
For a scheduling workflow, count correctly saved appointments. For a reception workflow, count complete requests delivered to the responsible person. Do not compare those outcomes as if they were the same service.
Use your current process as the baseline. Include employee time spent collecting missing information and correcting errors. If the business mainly wants better coverage after hours, measure that improvement directly rather than force it into a staffing-reduction story.
Include the whole cost chain
| Cost | What to inspect |
|---|---|
| Conversation | Duration and model or speech usage |
| Telephony | Number, carrier, and call charges |
| Integration | Business-system access and upkeep |
| Review | Staff time checking or finishing work |
| Failure | Repeat calls, incorrect actions, recovery |
Ask suppliers which of these are included. A demonstration may omit the recurring cost of system maintenance or an external service used during the call.
Use a transparent calculation
Suppose an illustrative pilot costs 400 units of currency and produces 100 reviewed, correct outcomes. Its observed direct cost is four units per correct outcome before any excluded overhead. If only 60 outcomes are correct, the denominator changes. These are arithmetic examples, not Golub pricing or customer results.
Track the distribution too. A few very long unresolved calls may matter more than the average. Separate ordinary inquiries from complex escalations so the comparison remains meaningful.
Decide what improvement is worth paying for
Faster response, fewer missed requests, and cleaner handoffs may create value without reducing headcount. Agree on the business objective and revisit the calculation with real usage.
Golub Voice is best evaluated through a defined call flow. Contact us with expected volume, call types, and the outcome you want to measure. That supports a grounded scope discussion without inventing a universal savings percentage.