Quantify your operational advantage.

Check the two places vendor models normally cheat. The first is the error term: ours prices an error at 8–15% of order value, because a bad quote costs you a remake — material, technician hours, freight, conceded margin — and not the whole order. Models that book the full order value on every error produce numbers an order of magnitude larger and cannot be defended in a room with a controller in it. The second is effort reduction: our most optimistic scenario claims 35% of quoting desk time, not 85%. Both constants are published above, both are adjustable, and the resulting payback lands in months rather than days.

Yes. The one-time implementation and ERP migration charge is a slider like everything else, it is subtracted at month zero of the NPV schedule, and it sits in the numerator of the payback calculation. It is excluded from only two figures — the annual net benefit and the return multiplier — because both are recurring-year ratios by definition, and that exclusion is stated in the audit notes rather than buried. What the model does not price is your own project manager, your data cleanup, and the productivity dip during changeover. Add those before you treat payback as a planning number.

Conservative, unless you are willing to change how quoting works. The scenarios are not product tiers — the software is identical in all three. They describe how much of the old process survives. Conservative assumes the platform runs beside existing spreadsheet habits and removes 10% of desk time. Moderate assumes the price book migrates and the team actually stops rebuilding quotes by hand, at 20%. Aggressive assumes the quoting process is redesigned around parametric configuration, at 35%. Budget on conservative; treat the difference as upside you have to earn.

Nothing leaves your browser. The calculator holds state in memory, the memo is assembled client-side and written to a local object URL, and there is no analytics call carrying your inputs. If you close the tab, the scenario is gone. When you engage formally, dispatch and quoting data sit in Canadian regions and the platform is PIPEDA-aligned.

Rate cards, price books, customer records, and open work orders come across as structured exports — CSV or a direct API pull where your ERP supports one. Historical quotes are imported read-only so that reporting continuity survives the cutover. We do not require you to retire the ERP; most customers keep general ledger and inventory where they are and replace only the quoting and dispatch layer.

Nine percent sits mid-band in the 8–10% range most Canadian mid-market finance functions apply to internal software investments. It is compounded to a monthly equivalent rather than divided by twelve, which is the difference between a schedule that reconciles and one that quietly drifts. If your committee uses a higher rate, the NPV falls and the payback and multiplier do not move at all, because both are undiscounted.

No, and you should treat any vendor who does with suspicion. What we can commit to contractually is the behaviour of the software: the margin floor is enforced at the line, the quote, and the approval; the rate card is published and reproducible; and the arithmetic is deterministic, so the same inputs always produce the same output. Whether recovered hours turn into margin depends on decisions inside your business that no software vendor can underwrite.