Replace story-point estimation with measured cycle time to get concrete delivery dates and confidence levels, using simple spreadsheet formulas and scatter plots.
Technical leaders waste time translating story points into dates while managers demand concrete delivery timelines. The article shows that cycle time, a direct measure of how long work actually takes, eliminates the magical thinking behind velocity and gives reliable predictions. It starts with raw spreadsheet data, demonstrates how to calculate cycle time even when a story finishes in a day, and explains a simple IF formula (or MAX) to avoid zero-day results.
The author walks through building a scatter plot with dates on the X-axis, letting teams see whether their cycle time has settled into a narrow band. By eyeballing the data they can draw 50%, 80% and 90% confidence lines, turning vague estimates into actionable forecasts that managers can trust. This approach surfaces hidden work-like the week spent finding expertise for code reviews-that velocity completely masks.
Practical steps include gathering at least three months of data, adding start and end dates, handling weekend days, and using a one-day floor for calculations. The article also warns against penalising teams for longer cycles when they work over weekends, emphasizing morale impact. By adopting cycle time, teams gain a transparent metric that aligns expectations, improves decision-making, and reduces the frustration caused by unreliable story-point forecasts.
Check out the full stdlib collection for more frameworks, templates, and guides to accelerate your technical leadership journey.