Quality at source

First-pass yield tracker

First-pass yield tracker is licensed CC BY 4.0. Attribution: Katafacts (katafacts.com).

1 · What it is

What it is

First-pass yield is the share of units that make it through a process stage without rework, rejection, or scrap — and rolled yield is the same idea across every stage combined, which is always lower than any single stage's rate because losses compound. A first-pass yield tracker asks for one thing per stage: how many units entered, how many advanced. Everything else — the per-stage conversion rate, the rolled yield, and which stage is weakest — is computed from that.

2 · When to use it

When to use it — and when not to

Use it when

You can count units entering and leaving each stage of a multi-step process — a production line, an approval workflow, an intake pipeline.
You want to know not just the overall pass rate, but specifically which stage is losing the most.
You want a rolled yield figure that reflects compounding loss across every stage, not an average of each stage's rate.

Not when

You only have a single before-and-after count with no stage-level breakdown — that gives you an overall yield but not the 'which stage' finding this tool exists for.
The process only has one step — there's no rolling to do, and a simple pass-rate figure will do the job without this tool.
You want time-based analysis (lead time, flow efficiency, where WIP is piling up) rather than count-based conversion — that's value stream mapping.

3 · How to fill it in

How to fill it in

Scope: What's in and out of this process — which product line, which start and end point? A first-pass yield tracker with no stated boundary blends processes that don't belong together into a number that describes neither.
Stages: List each stage in order with how many units entered it and how many advanced without rework or rejection. Pull real counts from a defect log or system of record where possible — a guessed count defeats the point of measuring at all.
Rolled yield · Weakest stage: Computed for you: the per-stage conversion rate for each stage, the rolled yield across all of them, and which single stage is weakest — nothing to enter here.
Narrative: Drafted from the weakest-stage finding and your observation notes — what you've seen or heard about why that stage specifically is losing units.
Countermeasures: What will you do about the weakest stage, ranked by impact and effort, owned and dated? A countermeasure that doesn't trace to the weakest-stage finding is a coaching failure mode, not a real action.

4 · What good looks like

What good looks like

The example below is a four-stage assembly line's first-pass yield: real entered/advanced counts per stage, a rolled yield figure with the arithmetic shown, the weakest stage named with its loss count, and a countermeasure that traces directly to that finding rather than a generic quality push.

Same example, as a downloadable xlsx workbook.

Download .xlsx

First-pass yield tracker

Bracket assembly line — first-pass yield

Dana Whitfield, Line Supervisor · 2026-02-10

Team: Marcus Lee, Quality, Priya Nair, Production · Sponsor: Alex Romero, Plant Manager

Scope

Scope runs from raw component prep through final inspection for the model-B bracket assembly line only. Excludes packaging and shipping, which are a separate value stream with their own defect log.

Stages

Stage	Entered	Advanced	Conversion
Component prep	500	480	96.00%
Sub-assembly	480	450	93.75%
Final assembly	450	430	95.56%
Inspection	430	410	95.35%

Rolled yield · Weakest stage

Rolled yield

82.00%

Weakest stage

Sub-assembly

480/500 × 450/480 × 430/450 × 410/430 × 100 = 82.00%

Weakest stage

Sub-assembly

93.75% conversion · 30 lost

Narrative

Sub-assembly is the weakest stage at 93.75% conversion, losing 30 of 480 units — the single largest loss of any stage in absolute terms. Operator notes point to fixture drift as the likely cause, not a one-off quality issue, which suggests a calibration schedule would address the root cause rather than just catching more defects downstream. Rolled yield across all four stages is 82.00%, meaning roughly one in six units started is lost somewhere before inspection.

Countermeasures

Add a mid-shift fixture recalibration check at sub-assembly, logged on the station's standard work sheet.
Linked finding: Sub-assembly — weakest stage at 93.75% conversion, 30 units lost
Impact: high · Effort: low · Owner: Priya Nair · Due: 2026-02-24
Set a rolled-yield target of 90% and re-run this tracker after the recalibration schedule has been in place for a full month to confirm it actually moved the number, not just sub-assembly's local rate.
Linked finding: Rolled yield — 82.00% overall, below the 90% standard for this line
Impact: medium · Effort: low · Owner: Marcus Lee · Due: 2026-03-24

5 · Common mistakes

Common mistakes

Averaging each stage's conversion rate instead of computing the rolled (compounding) yield.
Losses compound multiplicatively, not additively — four stages each at 95% conversion roll to about 81%, not 95%. Averaging overstates the real end-to-end yield every time there's more than one stage.
Treating every stage below 100% as equally worth fixing.
Same vital-few discipline as every other artifact in this catalogue (master-plan §1.3) — one weakest stage named with conviction beats a list of every stage with some loss.
Counting units at the start and end of the process only, skipping intermediate stages.
That gives an overall pass rate but destroys the one thing this tool is for: naming which specific stage is where the loss actually happens.

6 · What it connects to

What it connects to

upstream

Defect log
The per-stage entered/advanced counts this tracker needs are exactly what a defect log already records — this tool turns that log into a stage-by-stage yield finding.

downstream

A3 problem solving
Once the weakest stage is named, an A3 is where you'd run root cause analysis in depth if the fix isn't already obvious from the observation notes.

7 · Where AI helps

Where AI helps

Judgement — stays yours

Deciding whether a stage's loss is worth acting on now or monitoring for a pattern first
Committing to which countermeasure to act on first

Analysis — AI helps

Drafting the narrative from observation notes and the weakest-stage finding
Pressure-testing whether a proposed countermeasure actually addresses the named weakest stage

Drudgery — automated

Computing every stage's conversion rate and the rolled yield, with the arithmetic shown
Identifying the weakest stage from stage data
Exporting to xlsx in the house format

9 · Rate this kata