Pre-Qualification Model, Expiry and Evidence Integrity for Electrical Contractors
PortCity Utilities must pre-qualify electrical contractors before they can receive purchase orders. There are 180 contractors, most re-qualified annually. The proposed model has three areas: financial stability assessed from audited statements, three-year safety incident rates, and a current insurance certificate. Thresholds are: 70 or more qualified; 50 to 69 conditionally qualified with a safety improvement plan; below 50 disqualified. During the pilot, four problems appeared. Twenty-two contractors show expired qualifications yet still hold open purchase orders released last quarter. One contractor submitted an insurance certificate that the insurer later confirmed had been altered, after a $340,000 order was issued. Three assessors - a safety officer, a category manager and a finance analyst - score the same answers differently, producing a 30-point spread. Two disqualified contractors have asked for the specific reason their answers failed.
1. Is a single weighted score across the three areas adequate? Recommend how the model should be structured, including how expiry should be represented.
Read the worked answer
A single weighted average is inadequate here because it lets a strong financial score offset a missing insurance certificate. Insurance currency is a legal prerequisite, not a matter of degree. Structure the model with two layers: mandatory pass/fail criteria (valid certificate of insurance with adequate limits, no disqualifying safety event, financial statements supplied) and scored criteria inside each area carrying weights. A supplier that fails a mandatory criterion is disqualified regardless of its average.
Define outcomes explicitly: qualified, conditionally qualified with stated restrictions (for example, eligible only for orders below a value threshold until the improvement plan is accepted), and disqualified with reason codes. The reason codes are what later satisfies the two contractors asking why they failed, without disclosing anyone else's data.
Represent expiry as data, not as a status note. Each qualification carries a validity period driven by the shortest-lived evidence - typically the insurance certificate - and re-qualification must be initiated well before that date, not after it lapses. Build the questionnaire per area so evidence attachments are mandatory for the scored items.
Sanity-check the thresholds against the pilot distribution; if 50 to 69 captures most of the population, the model is not discriminating and the weights need revisiting before rollout.
2. Twenty-two contractors are expired with open orders, and one certificate was altered. What controls do you put in place, inside and outside the qualification process?
Read the worked answer
Accept that qualification status is not a spend control by itself. Expiry does not retroactively block an existing order, and the qualification record will not stop a buyer creating a new one unless procurement checks it. Two controls are needed: a check at requisition or order creation that warns or errors when the supplier is expired or disqualified, and an expiry report reviewed on a fixed cadence by the category team. Decide the grace policy in advance - for example, no new orders once expired, existing orders allowed to complete - and write it down so exceptions are policy, not favors.
Start re-qualification earlier. Trigger initiatives 60 to 90 days before expiry so the population never reaches a cliff edge.
For the altered certificate: suspend or block the supplier immediately, freeze further releases, escalate to compliance and legal, and record the finding against the supplier record. Do not rely on the expiry date to retire them quietly. Then fix the evidence control: insurance must be verified with the broker or insurer rather than accepted as an upload, and sampled re-verification should run quarterly. Upload alone is evidence of receipt, not of truth.
Report the exposure - 22 suppliers and the open commitment value - to the budget owner.
3. Three assessors produce a 30-point spread on identical answers. What changes make assessment consistent and defensible to suppliers?
Read the worked answer
Split responsibility and constrain judgment. Assign each assessor to the area they own - safety officer to incident rates, finance analyst to statements, category manager to commercial and delivery items - rather than having three people score everything.
Anchor the scoring. Define a rubric per question describing what a 2, a 3 and a 5 look like, with reference to the evidence required. Unanchored 1-to-5 scales are the main cause of spread this wide. Run a calibration session where all three assessors score the same two completed questionnaires and reconcile differences before live assessment begins.
Make evidence mandatory so a score can be traced to a document. Require justification text on any score at the extremes, and lock responses after the deadline so the basis does not shift mid-assessment.
Add a review layer for consequential outcomes: any conditional or disqualified result is second-reviewed by someone who did not assess it, which also gives you a defensible appeal route. Handle appeals with a reviewer outside the original panel and a documented outcome.
On transparency, give suppliers their own scored areas, the reason codes and the required remediation - not assessor identities, internal notes or other suppliers' data. That satisfies the two complaints while keeping the process fair and confidential.