Vendor Reliability in Freight Sourcing: A Scorecard Pricing Teams Can Use
Two vendors can quote the same movement and create very different amounts of work for your pricing team. One replies promptly but leaves destination charges unclear. Another takes longer, yet provides the requested cost breakdown, validity and routing in its first response.
Price matters, but so does whether the team can use the quote before the customer's deadline. A freight vendor evaluation scorecard helps pricing teams assess that part of the relationship using evidence from previous sourcing work.
Start with a defined question: how consistently does this vendor provide relevant, clear quotations that the team can evaluate without avoidable rework?
Define reliability for the sourcing task
In this scorecard, vendor means the carrier, airline, overseas agent, co-loader or other provider approached for a buy rate. The assessment covers the RFQ and quotation process. Booking execution and shipment delivery need their own evidence.
A quick quote does not establish that a shipment will arrive on time. Always assess container carrier schedule reliability and vessel delays using vessel-arrival data. A truck response history measures a different part of the service.
Keep both assessments available when selecting a provider, but label them clearly. A pricing team may value a vendor's quotation responsiveness while operations has concerns about its handling of bookings or exceptions. One record should not erase the other.
Build the scorecard from observable responses
Gather objective supplier-performance information and consider the supplier's experience of working with the buyer. The framework below applies that approach to freight RFQs, using the original request, replies and amendments as evidence.
Agree the required response before collecting results. A request for destination handling needs a different charge breakdown from a door-to-door quotation. Specify the shipment, service scope, required commercial details and response deadline so vendors are evaluated against a clear request.
Define an eligible RFQ as a complete request sent to an appropriate vendor. Keep one record per vendor per RFQ; amendments are versions of that record. Count each quotation once, even when it needs several corrections.
| Measure | Definition | Evidence to retain |
|---|---|---|
| Response reliability | RFQs receiving a substantive response by the agreed deadline / eligible RFQs sent. | Send time, deadline and first substantive reply. |
| First-quote completeness | First quotations containing all required information / quotations received. | Required fields and charges checked against the original RFQ. |
| Commercial clarity | Quotations with explicit, unambiguous applicable commercial terms / quotations received. | Currencies, charge units, validity, routing and inclusions or exclusions. |
| Correction burden | Quotations requiring correction because of vendor errors / quotations received. Lower is better. | Original quote, corrected version and reason for the change. |
| Lane and service fit | Quotations matching the requested lane, mode, scope and service constraints / quotations received. | Requested movement compared with the offered option. |
Report each result with its numerator, denominator and review period. For example, "12 complete first quotes out of 16 received" tells the team more than an unexplained four-star rating. Keep the measures separate initially; a single weighted total can conceal a weakness that matters for the current enquiry.
Record missing evidence as unknown. Do not award a good score because nobody logged a problem, or score a new vendor poorly simply because it has little history.
Measure a useful response rather than an acknowledgement
An automated "received, will revert" message should not count as a substantive response. A quotation, a clear decline or a specific clarification count, provided your definition is applied consistently. Report their outcomes separately: a timely decline helps the team move on, but it is not an available rate.
Alongside response reliability, record when a quotation becomes usable for comparison. A reply at 10:00 may still require two exchanges about local charges before pricing can evaluate it at 15:00. Retaining both timestamps shows the difference between replying promptly and completing the requested sourcing work.
Use a complete RFQ and an explicit deadline with a time zone. Record missing customer information, vendor working hours and changes to the request. If the customer changes the destination or cargo-ready date, identify the resulting revision separately from a correction caused by an error in the vendor's quote.
Apply the same distinction to rate amendments. A documented change in availability or an expired validity period needs commercial review; it should not automatically be classified as an avoidable vendor mistake.
Compare vendors within the work they actually handle
Review results by relevant lane or lane group, mode and service scope. An agent with strong airfreight responses should not inherit the same assessment for ocean exports. A provider quoting freight only should not be marked incomplete for omitting destination handling that the RFQ never requested.
Include the branch relationship where it affects the process. Response arrangements, contacts and local working hours may differ across offices. A global vendor average can hide those differences.
Track quotation availability alongside the scorecard: how many eligible requests produced a quote, a decline or no response? Frequent declines may indicate that the vendor is being approached for unsuitable work. Check RFQ targeting before treating every decline as poor performance.
Keep the sample size visible. A vendor with two excellent responses has promising evidence, but its record is less established than one assessed across dozens of comparable enquiries. Use a consistent review window and require minimum exceptions for discussion.
Read the results alongside price and current requirements
Consider an illustrative review of 20 comparable RFQs sent to each of two vendors. Vendor A provides 18 substantive responses on time and 15 quotations, of which 12 are complete on the first submission. Vendor B provides 17 substantive responses on time and 14 quotations, of which 11 are complete.
A has the stronger on-time response record. B provides more quotations and a higher proportion of complete first submissions. That difference gives pricing a reason to investigate the extra clarification work associated with A. It does not establish which provider will offer the right option for the next shipment.
For the current enquiry, compare the complete buy cost, validity, routing and service requirements separately. If a charge is unknown, keep the comparison provisional until it is resolved. Historical responsiveness cannot make an incomplete or inapplicable quote suitable.
Use the review to improve preferred vendor decisions
Give the pricing or procurement lead responsibility for reviewing the evidence with branch users. Share specific examples with the vendor and ask whether unclear RFQs, changed requirements or contact arrangements contributed to the problem. Agree a corrective action and review date where improvement is needed.
Make preferred status specific enough to help the next enquiry. Record the lane or service scope, reason, owner and review date. A vendor may remain preferred for one movement type while the team sources alternatives for another.
Keep exclusions and unresolved issues visible when deciding whom to approach. The scorecard should help a pricing executive choose relevant sources and anticipate follow-up work. The executive still needs to assess the quote offered for the actual shipment.
Bring vendor history into the Wend AI sourcing workflow
Wend AI keeps vendor RFQs, responses, reminders and rate comparisons connected to the enquiry. That sourcing history gives the team context when reviewing which providers to approach and how their quotations compare.
Use the scorecard above with a set of recent RFQs, then review the vendor workflow with the Wend AI team. Start with the evidence you need to retain and the decisions it should support. Agree how that information will be captured and reviewed, while keeping vendor selection and commercial judgement with your team.
About the author