Is your worst supplier really worse than the rest?
Three suppliers, eight goods-in lots each, defects per thousand units. One supplier is 48% worse than the best and the test settles it, and the follow-up test settles the pair everyone actually argues about.
Operations Starter Statistics free
After you install, this is the model to open.
Are These Three Suppliers Really Different?
- In your spreadsheet, click the Sortia icon in the strip of icons down the right-hand edge. No strip? Click the arrow at the bottom-right to open it. You can also use Extensions, then Sortia, then Open Sortia.
- Click Start from a template and put that name in the search box.
- Pick the card with that name and click Load this template. It arrives on a new tab with real numbers already in it.
The answer
- Calder against the best
- +48% 12.56 defects per thousand against Northgate's 8.46
- The group verdict
- F = 42.11 p = 0.000000045 against a critical value of 3.47
- Northgate vs Belmar
- p = 0.22 the gap everyone argues about is noise
- What Calder costs
- 262 extra defective units a year at current volumes
Three suppliers, eight goods-in lots each, defects per thousand units. The averages are 8.46 for Northgate, 9.08 for Belmar and 12.56 for Calder. Click Run: F comes back 42.11 against a critical value of 3.47, with a p-value of 0.000000045. Three identical suppliers would essentially never produce a spread this wide, so at least one of these three is genuinely different, and the extra defective units line beside the data prices it at about 262 extra defective units a year against your best supplier at current volumes.
But the test has told you that the group averages differ, not which supplier is the problem, and the argument in the room is almost always Northgate against Belmar rather than anyone against Calder. So run the second test: switch to t-Test: Equal Variances with Northgate as variable 1 and Belmar as variable 2. It returns t of -1.28 with a two-tail p-value of 0.2198, so the 0.61 gap between your two good suppliers is noise, and moving volume between them on this evidence would be moving it for no reason.
Calder is the finding; the Northgate against Belmar ranking is not. That two-step is how a one-way test is supposed to be used, and skipping the second step is how a supplier gets punished for a rounding error. What the test cannot tell you is why Calder is worse, and a defect rate has no opinion about tooling, transport or the shift that packed it.
To use your own goods-in log, paste one supplier per column with the name in the first row, keep the columns the same length, and widen the input range.
The model
It arrives on a tab called Template: Three Suppliers, One Question, carrying these columns:
- Northgate (defects per 1000)
- Belmar (defects per 1000)
- Calder (defects per 1000)
with the model computed beside the data:
| Extra defective units a year at Calder against Northgate | 262.4 |
Once it is in your sheet
- The model arrives with real numbers in it and runs as it stands, so you can press the button first and understand it second.
- Change the numbers to yours. The sheet marks which cells are inputs and which hold formulas, and most labels carry a note explaining the row.
- Press the run button at the bottom of the panel. It is labeled for the tool you are in, and the result lands on its own tab, with a written reading of it beside the figures.
Never used Google Sheets? Start here goes the whole way, in seven steps, and assumes nothing.
Next question
- Does this delivery note match a supplier on file?Which Rows in Two Vendor Lists Are the Same Supplier?
- What does an unreliable supplier cost you in stock?Safety Stock for a Plant That Cannot Stop
- Is the line running on a cycle nobody scheduled?Does Our Output Repeat on a Cycle?
- Is a flat daily plan good enough?The Season Repeats: Forecast It
- Is your contingency big enough if delivery slips?What If the Supplier Misses?
- How many tickets must you sell to break even?What Ticket Price Covers the Room?
Every model like this one, and the method behind them: Statistics in Google Sheets.
The method behind this one, worked end to end: ANOVA in Google Sheets.