Per-stage performance table

One row per stage: runs completed of started, average planned, average actual, average variance, how many ran long versus short, days lost, and photos taken.

What it is

The heart of the panel. Ran-long and ran-short are shown as counts beside the average because 'the average hides one disastrous job inside forty fine ones'. Days lost carries an (i) when some delays were logged without a cause or a cost, so the reader knows how much of the total is missing. The top delay cause is printed under the stage name.

Also called: planned vs actual per stage · which stage runs long · stage variance table

See it
Per-stage performance table
Apr
May
Jun
Jul
Delia Yoder — 30×40 Garage
Framing
Ivy Brubaker — Riding Arena
Concrete
Ivy Brubaker — Barndominium
Trim
Ivy Brubaker — Barndominium
Framing
Sutter Kline — Riding Arena
Concrete
Framing crewConcrete crewTrim crew
The table with columns Stage / Runs / Planned / Actual / Variance / Long-short / Days lost / Photos, variance coloured red or green with a leading sign. StageAnalyticsTab.tsx. Sample data — no customer information appears here.
How it works
  1. 1Stage logs are grouped by task type; only rows with a recorded actual count toward the averages.
  2. 2avgPlanned reads baseline_days first, falling back to planned_days.
  3. 3ranLong / ranShort / onTime are counts of positive, negative and zero variance.
  4. 4Delays with neither a cause nor a days impact are counted separately and excluded from the days-lost total.
  5. 5The Runs column reads completed/started so the denominator is always visible.
Why we built it

Two comments in the route carry the reasoning: on the counts, ' "Ran long" is the count that drives a conversation; the average hides one disastrous job inside forty fine ones'; and on the missing-detail counter, 'Surfaced rather than dropped: it is the measure of how much of this data can be trusted.'

The problem
  • Stage duration estimates were never checked against outcomes.
  • Averages concealed the frequency of overruns.
  • Incomplete delay records silently understated days lost.
Sound familiar?
What you get
Stage estimates can be corrected from real history.
Frequency and magnitude of overruns are both visible.
The panel states how trustworthy its own delay data is.

See it on your own jobs

Twenty minutes, your numbers, no slide deck. We’ll build one of your real buildings in front of you and send you the estimate link at the end — yours to keep either way.

or keep browsing features →