Coffee Roasting and Dry Goods KPIs: Benchmark Ranges and How to Hit Them
Seven KPIs for coffee, tea and dry goods plants, from roast yield consistency to scrap: how to measure each one, what moves it, and which figures have a cited public source.
Seven KPIs run a roasting and dry goods plant: roast yield consistency, packaging line OEE, fill giveaway, energy intensity, labor productivity, changeover time, and first-pass quality with scrap. Use the ranges below as reference points for spotting a gap worth investigating, not as targets.
Every figure here shows its basis. A superscript number links to a public source in the Sources list at the end, and the sentence states that source's scope, such as its year and the industries it covers. A figure marked no cited source is kept exactly as originally published because no public source stating it was found, so treat it as illustrative.
At a glance
Coffee, tea roasting and dry goods KPI benchmarks at a glance
Ranges are as originally published unless a source is cited. Where the original used a word other than typical, such as world class, well run or target, that word is kept in the cell. The cited OEE data comes from consumer packaged goods packaging lines in general, not coffee lines. SMED means single-minute exchange of die.
Method: How the figures in this guide are sourced
A figure with a superscript number is stated by the cited source, listed under Sources with its publisher, date and access date. The cited OEE and minor-stop data covers packaging lines across industries, and the sentence says so. A figure marked no cited source is kept exactly as originally published because no public source stating it was found. It is illustrative, not measured industry data. Worked examples are plain arithmetic on the numbers shown.
Roast yield consistency
Roast yield consistency is the KPI that separates a controlled operation from a hopeful one. What matters is not just the average yield but a tight spread around it.
World-class roasters hold batch-to-batch yield within plus or minus 0.5 percent, while typical operations drift plus or minus 1.5 to 2 percent no cited source.
For example, on a 5.00 dollar effective green base, a 1 percent yield swing is 5 cents per pound of margin variance.
How to measure
Log green weight charged and roasted weight dropped for every batch.
Chart the spread batch to batch, not just the average.
What moves it
Automated drop-temperature control.
Consistent charge weights rather than operator feel.
Log batch yield and see the spread across your roasts. Roast Batch Yield
Packaging line OEE
Packaging line OEE is the throughput KPI: availability times performance times quality. World-class packaging lines run 80 to 85 percent1, 2, typical lines 55 to 65 percent1, 3.
Those figures are cross-industry, not coffee-specific. In PLC data from 725 consumer packaged goods packaging operations, measured from January to June 2009, the top quarter of lines averaged 82 percent OEE and the middle half 54.9 percent1. The classic world-class OEE threshold from total productive maintenance is above 85 percent2, and OEE software maker Vorne says most manufacturers score closer to 60 percent3.
Anything below 50 percent signals chronic film or changeover loss no cited source. In the same 2009 data, the bottom quarter of lines averaged 29 percent OEE1, and the average line had about 8.6 minor stops of 10 minutes or less per hour, a stop roughly every 7 minutes4.
For example, a line rated 900 bags per hour running at 60 percent OEE delivers 540. Cutting film splice and jam downtime from 15 percent to 8 percent adds 63 bags per hour at nameplate rate, somewhat less after performance and quality losses, at no capital cost.
How to measure
Count line output against nameplate rate.
Measure availability, performance and quality separately.
Attack the largest of the three first, which in this category is usually availability.
Fill giveaway is a money KPI hiding in a quality metric. Target average fill within 1 percent of nominal; world-class lines hold 0.5 to 1 percent giveaway, and typical lines give away 2 to 4 percent no cited source.
The legal floor sits below all of these. In the U.S., NIST Handbook 133 (2026 edition) requires the average net quantity of packages in a lot to at least equal the labeled quantity, rejects individual packages short by more than the Maximum Allowable Variation, and sets no limit on overfill5.
For example, 3 percent giveaway on a 12 oz bag is 0.36 oz of free product per unit, roughly 17 cents per finished pound at a 5.90 dollar cost.
Moving average fill from 12.4 to 12.1 oz on a 900 bag per hour line saves 0.3 oz x 900 = 270 oz an hour, about 16.9 lb. Over an 8-hour shift that recovers about 135 lb of product.
How to measure
Use statistical checkweigher data across a full shift, not spot checks.
Express giveaway as average fill above nominal, as a share of nominal.
What moves it
Target-weight control on the filler.
Regular scale calibration.
Energy intensity
Energy intensity benchmarks batch efficiency. Drum roasters target 1,800 to 2,500 BTU per lb when well run, with 2,500 to 3,500 typical and anything above 4,000 signaling excessive idle firing or a leaky drum no cited source.
Between-batch idle is the usual culprit, often 20 to 30 percent of total gas burned no cited source. Sequencing batches to eliminate a 10 minute idle gap can cut intensity 300 to 500 BTU per lb no cited source.
How to measure
Measure gas use per drop.
Divide it by roasted pounds to get BTU per lb.
Trend BTU per lb weekly.
What moves it
Back-to-back batching to cut idle firing.
Insulating the drum and ductwork.
Tuning burner modulation.
Labor productivity
Labor productivity is the people KPI, measured as finished pounds per labor hour. World-class small-batch roasters hit 250 to 400 lb per labor hour, typical operations 120 to 200, and under 100 usually means changeover or manual packing is eating the shift no cited source.
Inverted, 250 to 400 lb per labor hour maps to 0.0025 to 0.004 labor hours per lb.
The biggest lever is changeover and packing automation. Moving from hand-weighing to auger fill can double pounds per labor hour no cited source, and batching same-profile roasts back to back removes redundant warmup labor.
How to measure
Count finished pounds and all labor hours across full shifts.
Include cleaning time in the labor hours.
What moves it
Packing automation, such as auger filling in place of hand-weighing.
Changeover time gates flexibility and small-order economics. Target under 20 minutes for a same-product bag-size change and under 45 minutes for a full flavor or product changeover with cleaning no cited source.
Typical operations run 45 to 90 minutes, and world-class SMED-disciplined lines hit 10 to 15 minutes on size changes no cited source. For reference, single-minute exchange of die (SMED) sets the target of changing over in a single digit of minutes, under 10, by moving internal setup work to external work done while the line runs6.
Pre-staging the next film roll, bags and flavor totes offline can strip 30 to 50 percent off the stopped time no cited source.
How to measure
Time size changes and full flavor changes separately.
Separate internal tasks that need the line stopped from external prep that does not.
What moves it
Converting internal work to external work.
Pre-staging the next film roll, bags and flavor totes offline.
First-pass quality and scrap round out the scorecard. Target first-pass yield above 98 percent on packaged units and total scrap under 1.5 percent; typical operations sit at 95 to 97 percent first pass with 2 to 4 percent scrap no cited source.
Each point of scrap on a 5.90 dollar cost base is about 6 cents per pound.
Pareto the defect reasons. Off-profile roasts, seal failures and stale returns usually dominate.
How to measure
Divide good packaged units by units started.
Code every scrapped unit by reason and Pareto the reasons weekly.
What moves it
Drop-temperature discipline.
Seal-bar temperature and dwell control.
Tighter finished-goods rotation to cut staling returns before they become scrap.
Treat these seven KPIs as one dashboard, not isolated numbers, because they trade against each other. Pushing packaging OEE by running film hard can raise seal-failure scrap; chasing energy intensity with tight back-to-back batching can starve changeover flexibility.
World class in this category is not a single hero number. It is holding roast yield within 0.5 percent, giveaway under 1 percent and scrap under 1.5 percent no cited source, with packaging OEE above 801, at the same time, quarter after quarter.
How to build your own baseline
Set a weekly review that trends all seven KPIs together.
Write down each KPI's definition and keep it fixed.
Collect several months of data before you set a target.
Compare against your own trend first and published ranges second.
Record what changed in the process whenever a KPI moves.