# Sprint Forecasting Forecast what your team will actually ship this sprint, not what the spreadsheet says. iftrue blends historical velocity, AI lift, unplanned work rate, and carryover drag into a realistic commitment number. Sprint 24 forecast · Backend planning Mon 10am **Recommended commitment** P80 confidence **38pts** range 34 to 42 pts · stretch to 44 ## How we got to 38 - Last 6 sprint completion avg 36 pts - AI lift (sustained since Sprint 21) +7 pts - Unplanned work reserve (23% avg) −4 pts - Carryover drag (3 stuck tickets) −1 pt **Forecast commit 38 pts** 3 tickets have carried over 3+ sprints. Consider cutting or splitting before committing again. They're costing 1 pt of velocity per sprint. ### Sprint commitment forecast with confidence range ### Unplanned work ratio and carryover drag factored in ### Carryover radar for tickets stuck across multiple sprints ### Pre-sprint sanity check before commitments are locked ## Forecast inputs ### Velocity alone lies A six-sprint average ignores AI lift, unplanned work, and tickets that never close. iftrue blends four inputs so the forecast reflects how your team actually ships. ### Sprint completion trend Points completed vs committed, last 6 sprints. Outliers like vacation weeks and incidents are down-weighted. ### AI lift Measured velocity change since AI tool adoption. Applied to the baseline so the ceiling reflects today, not pre-Cursor days. ### Unplanned work ratio How much of each sprint has historically gone to unplanned bugs and support. Held back from commitment. ### Carryover drag Tickets bleeding across sprints drain velocity. The forecast accounts for the ones still lurking in your backlog. ## Sprint trends ### The charts your retro needs Story point completion ratio, unplanned work ratio, and work distribution by type over the last 6 sprints. Every chart is an input to the forecast, and every chart is a talking point for retro. - **Completion ratio**: Points completed divided by points committed. Consistent undershoot means the team overcommits. - **Unplanned work ratio**: Share of each sprint burned on work that wasn't in the plan. Rising trend is a planning or quality signal. - **Work distribution**: Who finished what, by issue type. Spot engineers carrying bugs while others ship features. ### Story point completion last 6 sprints S18 S19 S20 S21 S22 S23 Solid bars = completed. Dashed line = committed. Gap is overcommitment. **Unplanned work ratio** 23% avg Trending up. Worth a retro discussion about scope protection. ### Zombie tickets · carried across sprints 3 flagged - Bitbucket marketplace installation missing callbacks (4 sprints) To Do · 12 pts - Sprint average numbers in the team sprints tab is wrong (3 sprints) To Do · 5 pts - Unplanned work ratios don't add up to 100% (2 sprints) In Progress · 3 pts #### Suggested action Cut the 12-point Bitbucket ticket, or split it into a 3-pt spike plus separate implementation. 4 sprints is too long. ## Carryover radar ### Kill zombie tickets Every team has them: tickets that keep rolling over sprint after sprint, always "in progress", never done. iftrue surfaces them with a sprint counter per ticket so you can cut, split, or unblock before planning starts. One of these tickets drags down velocity more than it contributes. The counter is visible on every sprint board, every retro, and rolled into the forecast. No more discovering in month three that the same ticket has been "nearly done" since January. ## Pre-sprint sanity check ### Catch overcommitment before it ships Before your planning meeting ends, iftrue checks the draft commitment against the forecast and pings you in Slack if it's outside the safe range. iftrueAPP just now · DM to manager **Sprint 24 is over-committed** Draft commit is **48 pts**. Forecast ceiling is **44 pts (stretch)**. Last 4 sprints you missed target by an average of 6 pts when committing above forecast. **Suggest cutting 4-6 points before planning closes. PROJ-512 (5 pts) and PROJ-518 (3 pts) are the lowest-priority candidates.** ## Who it's for ### Built for engineering leaders - **Engineering Managers**: Walk into planning with a number you can defend. Stop overcommitting because the PM asked for more. - **Tech Leads**: See which tickets have been stuck across 3+ sprints and either unblock them or cut them. Stop the zombie ticket cycle. - **VPs of Engineering**: Aggregate forecasts across teams to plan realistic quarterly delivery. Calibrate roadmaps against actual throughput. ## See what your engineering org is missing. 30-minute walkthrough, tailored to your stack. No slides. Your team's data.