Reviewed for technical terminology, Turkish factory field reality and published EUR scope boundaries (not statutory engineering sign-off).
1. Step one: separate essential pass/fail gates from weighted preferences
No amount of polite email responsiveness or attractive payment terms can rescue a supplier who physically cannot roll your required 3.0 mm structural section, refuses orders below 100 tonnes when you need 12 tonnes, or lacks mandatory product conformity documentation required by your project’s importing market.
Before touching any numeric point scale, define three to five Essential Conditions (Pass / Fail / Pending Clarification). A candidate that clearly fails an essential condition is recorded as Not Suitable for Current Brief with the exact reason documented—rather than carried into a weighted average where secondary perks could make an unusable offer look competitive.
Weighted scoring belongs strictly to candidates who either satisfy your essential gates or have one bounded pending confirmation that can be checked in the next stage.
2. Standard 100-point criterion table for building-product shortlists
Below is a balanced 100-point weighting model tailored for construction materials, fabricated steel and modular building components. Teams can adjust weights before issuing an RFQ—for example, raising Documentation & Traceability for regulated façade systems—provided weights are locked before supplier responses are read so nobody tweaks weights retroactively to favour a preferred vendor:
| Criterion ID & category | Max weight | What is evaluated at shortlist stage |
|---|---|---|
| C1 · Technical specification & geometry fit | 25 pts | Alignment with drawings, dimensions, base grades, coatings and clear labelling of any proposed deviations |
| C2 · Manufacturing role & process capability | 20 pts | In-house production vs trading role, relevant machinery, coil/panel size limits and subcontracting transparency |
| C3 · Commercial clarity & normalised unit basis | 15 pts | Transparent unit pricing, tooling separation, MOQ fit, lead-time realism and Incoterms® clarity |
| C4 · Documentation & traceability alignment | 15 pts | Relevance and consistency of datasheets, test references, entity names and revision control |
| C5 · Export packing, pallet & loading readiness | 10 pts | Defined pallet dimensions/weights, moisture/edge protection and truck/container loading method |
| C6 · Technical communication & responsiveness | 10 pts | Direct answers to numbered RFQ questions without evasive boilerplate |
| C7 · Sample & factory-visit cooperation | 5 pts | Willingness to supply traceable samples and confirm plant access for a scoped visit |
| Total available weight | 100 pts | Assessed weight coverage tracked explicitly whenever any criterion is marked Unknown |
3. The 0–5 + Unknown rating scale (and why Unknown stays out of the denominator)
Each criterion with available evidence is scored on a 0 to 5 integer scale, converting to weighted points via the formula Earned Points = Criterion Weight × (Score ÷ 5). Whenever a criterion has not yet been evidenced—for example, export pallet specs have not yet been sent, or physical process capability awaits a factory visit—mark the rating Unknown rather than guessing 0 or 3.
- 5 (Strong fit, documented): Directly meets the requirement with clear written/documentary or observed support.
- 4 (Good fit, minor detail open): Meets core requirement; minor non-critical detail to confirm at order stage.
- 3 (Acceptable with stated trade-off): Workable option with an explicit commercial or technical trade-off.
- 2 (Weak fit / material deviation): Significant gap or unapproved substitution requiring engineering/commercial review.
- 1 (Poor fit): Barely addresses the requirement; high operational risk.
- 0 (Verified non-compliance): Evidence shows the supplier does not meet this criterion.
- Unknown (Not yet assessed): Evidence missing or awaiting verification—excluded from assessed weight denominator and flagged via coverage %.
4. Worked scorecard arithmetic: Supplier A vs Supplier B (fictional training example)
Fictional training example only: The table below uses invented scores for fictional Supplier A and Supplier B to demonstrate why reporting Earned Points / Assessed Weight alongside Coverage % prevents bad shortlisting decisions. Neither column represents a real company or completed project.
Notice how Supplier A has not yet provided documentation (C4, weight 15), so C4 is marked Unknown. Only 85 of the 100 weight points have been assessed (85% coverage). Across those 85 assessed points, Supplier A earns 64 points (20 + 16 + 12 + 6 + 6 + 4 = 64 / 85 assessed weight, or 75.3% of assessed weight). Meanwhile, Supplier B has supplied evidence across all seven criteria (100% coverage) and earns 69 / 100 assessed weight (69.0% of assessed weight):
| Criterion (Weight) | Supplier A rating & calculation | Supplier A pts | Supplier B rating & calculation | Supplier B pts |
|---|---|---|---|---|
| C1 · Spec & geometry fit (25) | 4/5 → 25 × (4 ÷ 5) | 20 | 3/5 → 25 × (3 ÷ 5) | 15 |
| C2 · Process capability (20) | 4/5 → 20 × (4 ÷ 5) | 16 | 4/5 → 20 × (4 ÷ 5) | 16 |
| C3 · Commercial clarity (15) | 4/5 → 15 × (4 ÷ 5) | 12 | 4/5 → 15 × (4 ÷ 5) | 12 |
| C4 · Documentation (15) | Unknown (excluded from denominator) | Unknown | 4/5 → 15 × (4 ÷ 5) | 12 |
| C5 · Export packing (10) | 3/5 → 10 × (3 ÷ 5) | 6 | 3/5 → 10 × (3 ÷ 5) | 6 |
| C6 · Responsiveness (10) | 3/5 → 10 × (3 ÷ 5) | 6 | 3/5 → 10 × (3 ÷ 5) | 6 |
| C7 · Sample / visit readiness (5) | 4/5 → 5 × (4 ÷ 5) | 4 | 2/5 → 5 × (2 ÷ 5) | 2 |
| Total & coverage summary | 64 / 85 assessed weight (85% coverage) | 64 / 85 | 69 / 100 assessed weight (100% coverage) | 69 / 100 |
5. How to interpret the result without turning arithmetic into autopilot
Look at what the two-part notation reveals immediately. If an analyst carelessly treated Unknown as 0, Supplier A would appear as 64/100—losing to Supplier B (69/100) even though Supplier A actually matches the drawing geometry (C1: 20 vs 15) and visit readiness (C7: 4 vs 2) better! Conversely, if an analyst converted 64/85 into a bare percentage (75.3%) without stating 85% coverage, a reader might award the contract to Supplier A while overlooking that 15 points of critical product documentation (C4) remain completely unverified.
Displaying 64 / 85 assessed weight (85% coverage) vs 69 / 100 assessed weight (100% coverage) points directly to the rational purchasing action: hold final award, require Supplier A to submit the missing C4 product datasheets/test references (or run a €190 Desk Check), and visit the winning plant (€490 core-area Factory Visit, or combined inside PK-SOURCE at €1,290) before releasing a deposit.