Chapter 10 – Choosing tools: the questions that matter
The Quantitative Schedule Risk Analysis Handbook · Edition 1, August 2026 · by the team behind [Epoch SRA](https://epochsra.com), an Excel add-in for schedule risk analysis · corrections welcome at contact@epochsra.com
This chapter is deliberately vendor-neutral, ours included. The SRA tool market is small and genuinely differentiated – enterprise platforms, scheduler add-ins, Excel-based tools, and general simulation engines all coexist because they answer different organizational shapes. The wrong question is "which tool is best?" The right question is "who will operate it, on what schedule data, under what constraints?" – and the checklist below turns that into vendor conversations.
Where does my schedule data go? Local computation, vendor cloud, or on-premise server are different answers with different consequences – for export-controlled programs, often decisive ones. Ask precisely; "secure" is not an architecture.
What happens to constructs the tool cannot handle? Every import or integration has edges – link types, calendars, constraints, resource logic. The differentiating answer is not "we support everything" (no one does) but "here is what we do with what we don't support." Silent approximation and honest reporting are different products at the same feature list.
Which of your numbers are calibrated, and against what? If the vendor's answer to "how do you know your distributions are honest?" is a description of the input dialog, Chapter 7 applies. Ask whether any percentile has been backtested against actual outcomes, and whether the output distinguishes evidence-backed numbers from model estimates.
What does the monthly cadence cost? Not the license – the labor. Re-import a changed plan: do mappings survive? Re-run: are results reproducible and diffable against last month? A tool that makes the first analysis easy and the twelfth expensive optimizes for the demo, not the program.
Can my schedule quality be assessed before my risk is? Chapter 4's argument in procurement form: ask what schedule-health checks run, and what happens when they fail.
What does a seat cost per year, all-in? Published pricing, quote-based pricing, per-seat versus suite, maintenance terms. None of these answers is wrong; unavailable answers are.
Can I evaluate on my own data without a sales process? The fastest tool evaluation is your real schedule through a real trial. Vendors who make that possible in an afternoon are making a statement about their product; vendors who route it through three calls are making a different one.
Weight these by your organization's shape. A risk office running a thousand-task portfolio weights integration depth and enterprise features. A program manager who is the risk process weights time-to-answer, cadence cost, and data locality. Both are right, and most tool disappointment traces to buying for the other organization's shape.
--- In practice: send this checklist, verbatim, to every vendor on your shortlist and to your own team about the current process. The current process is also a tool, and it should survive the same questions.