Academy · ARTICLE
Brand Scorecard explained: apples-to-apples tool comparison
The math behind the Brand Scorecard, in plain English. Cost-per-part is the right denominator. Here's how Niyamis TMS computes it — and what changes when you use it instead of sticker price.
Philippe, Niyamis · 8 min read
The one question every shop asks wrong
"Which brand is cheapest?"
Every precision shop in New Zealand has had that meeting. The vendor quotes come in, someone lines up the sticker prices in a spreadsheet, and the shop picks the cheapest line. Six months later, scrap rates on a certain alloy suddenly look wrong — because the cheapest end-mill on the invoice was the most expensive end-mill in the spindle.
The Brand Scorecard answers a better question: which brand costs you the least per part produced on this specific job?
The formula
cost_per_part = unit_price / average_tool_life_in_parts
That's the whole thing. Everything else on the Brand Scorecard is a way of feeding real numbers into that fraction.
- Unit price comes from your PO history. Niyamis TMS pulls it in on ingest — every line item you've received in the last eighteen months.
- Average tool life comes from your transaction data. TAKE means the tool went into service; SCRAP means it came out. The difference is a service window. Cross-reference the tool with the job it was on and you get parts-per-tool.
Neither number needs a sensor. Both numbers are already in your shop's operational records. TMS just reads them together.
Example: a specific 10 mm end-mill
This example uses placeholder numbers until Accord clears the real ones.
| Brand | Unit price | Avg. life (parts) | Cost per part |
|---|---|---|---|
| Brand A | $48 | 72 | $0.67 |
| Brand B | $63 | 110 | $0.57 |
| Brand C | $39 | 44 | $0.89 |
The cheapest brand on the invoice (C, at $39) is actually the most expensive brand per part ($0.89). The middle-priced brand (B) wins. The scorecard makes that visible at a glance.
Why the winner changes with the alloy
The scorecard is computed per SKU per job context, not per SKU globally. When Accord compared Kennametal and Sandvik on a specific 10 mm end-mill, Sandvik's cost-per-part was 22% lower on the 80-part aluminium job — but the opposite was true when the same end-mill ran on 316 stainless. (Again, placeholder numbers pending sign-off.)
This is the result most shops suspect but can't prove: tool performance is alloy-specific, and comparing brands on a global average hides the difference. The scorecard keeps the context attached, so you see the winner on aluminium next to the winner on stainless next to the winner on titanium.
What the screenshot shows
- Filter bar. Pick the SKU, the alloy, the machine class. The scorecard reshapes under you.
- The leaderboard. Brands stacked by cost-per-part, with unit price, average life, and scrap rate as columns.
- Confidence. Each row has a confidence indicator — green when the sample size is large enough to trust, amber when it isn't. A brand with 8 parts of history is tentative; a brand with 800 is not.
- The delta. For every non-winning brand, the scorecard shows what you'd save by switching, annualised at your current volume.
What to do with it
The first time a shop sees a Brand Scorecard, two things usually happen. One, something changes on the next PO — often a commitment volume moved from the cheaper-on-paper brand to the cheaper-per-part brand. Two, a conversation starts about the SKUs where the sample size is too small to trust, and whether those SKUs should be rationalised.
Both are the right outcome. The scorecard is meant to drive procurement decisions, not replace them — it gives your buyer the one number per SKU that makes the conversation with the vendor shorter and sharper.
Next in the series
- Tool life prediction without sensors — how Niyamis TMS forecasts remaining life for every tool in service.
- A field guide to scrap — what your SCRAP notes are trying to tell you.