Boundary Explorer · Tool
Compare the current way of doing a recurring job with one or more proposed ways, under your stated numbers and useful lifetime. See the cost figures, the boundary where the preferred option changes, and an honest sensitivity reading.
The defaults below are illustrative — the same example shown on the home page. Open the inputs you want to change; the verdict updates live.
Comparison made
payback · 0.5 months
A lower modeled time cost is not overall business superiority. Quality, eligibility, and what the comparison did not model remain unresolved.
Change one assumption · verdict updates live
Comparison made
| Option | Total minutes over horizon | Per-period minutes | Future setup | Origin | Eligibility |
|---|---|---|---|---|---|
| manual(baseline) | 3600 | — | sunk | illustrative | — |
| assistedAssisted reusable workflow | 1632 | — | — | illustrative | eligible |
Variable: frequency_per_month (runs/month)
Searched range: 0.1 – 500 runs/month
Current value: 15 runs/month
Flip at: 2.9411764705882355 runs/month
Immediately below this transition
With frequency_per_month immediately below 2.941 runs/month, manual has the lowest modeled cost.
Immediately above this transition
With frequency_per_month immediately above 2.941 runs/month, assisted has the lowest modeled cost.
What to record: Record or estimate the recurring job frequency in runs per month.
What would change the comparison: Moving from the current interval to just below 2.941 runs/month would change the current preferred option (assisted) to manual.
What would remain unresolved: Remaining in the current interval would not change the modeled winner; crossing another listed boundary may change it. Quality and eligibility assumptions remain unresolved.
This break-even crossing does not establish that the target value is uncertain or worth measuring.
Defaults are illustrative. Open what you want to change.
Frequency 15 runs/month · Useful lifetime 8 months · Selected target Frequency (shared)
Description of the approach, not a quality ranking — e.g., "Manual today", "Assisted reusable workflow".
Description of the approach, not a quality ranking — e.g., "Manual today", "Assisted reusable workflow".
Add one named scenario with provenance and an optional frequency override. The engine re-runs the comparison under that scenario and reports whether it changes the preferred option. Use it to check how the verdict moves if your estimate is off in a particular way.
No scenario added.
Copy a link that encodes only numeric estimates, eligibility flags, fixed origin classifications, and version identifiers. Before you copy: option names and ids, scenario names, provenance, notes, and ineligibility reasons are omitted. A reopened link uses neutral labels while preserving the permitted numbers and their option/scenario relationships. Anyone holding the link can read that permitted state.