Home / Guides / Estimate vs actual
Estimate vs actual
Most teams estimate every sprint and never look back. A quick comparison of estimates with what the work really took is the cheapest way to get better at it.
What to record
After a story is done, the team gives it an actual in the same unit as the estimate: “in hindsight, that was a 13”. Keep it relative. You are not tracking hours, you are checking whether the sizing held.
What to look for
- How many landed close. Count the stories whose actual is within one deck step of the estimate. That’s a fair definition of “accurate” for relative sizing.
- Which way you miss. If most misses run long, the team is optimistic. Bring that up at the next estimate, not as blame but as calibration.
- Which cards drift. Often one value is the problem: your 8s are fine but your 3s keep turning into 5s. Revisit the reference story for that card.
Keep it safe
- Look at the team’s numbers, never at individuals.
- Never turn estimates into deadlines after the fact.
- A handful of stories tells you little. Wait for a few sprints before drawing conclusions.
In PointHuddle
The Summary tab lists every estimated story with a field for its actual. The Stats tab then shows how many stories landed within a step, which way you tend to miss, and how each card held up, along with session pacing (time per story and stories per hour).
Try it with your team
Create a room, share the link, and vote. No accounts, no sign-up, free.
Start a planning poker room