The Q2 MSP scorecard for a 2,000-contractor IT program came back green across the board. Time-to-fill: 12 days. Fill rate: 84%. Rate card compliance: 96%. Every SLA met, every column in range. Three weeks later, the Procurement Director was in a budget meeting explaining why cost-per-hire for IT roles had climbed 14% in two quarters while headcount stayed flat.

The scorecard never showed that coming. It wasn't designed to.

What MSP Metrics Actually Measure

Most MSP program scorecards track fill rate, time-to-fill, rate card compliance, and submission volume. These are the right metrics for what they do: they tell you whether vendors are showing up, responding to reqs, and staying within agreed billing parameters.

What they don't capture is how many submittals it took to produce a hire, how many interview rounds were needed before a submittal passed hiring manager review, or what happened to effective cost-per-hire as the program pushed more volume through a concentrated supplier base. Those gaps matter more than the SLA columns suggest. Fill rate tells you whether the req was closed. It says nothing about how efficiently it was closed.

How Vendor Tier Concentration Works, and Why It Feels Like a Solution

MSP programs typically organize vendors into tiers. Tier 1 suppliers receive the highest volume of requisitions, based on their ability to respond quickly, handle multiple reqs simultaneously, and meet program compliance requirements at scale. This makes operational sense: fewer vendors to manage, more pricing power at volume, more consistent administrative process.

The problem shows up when req distribution becomes mechanical rather than merit-based. When Tier 1 vendors receive first right of refusal on every IT requisition regardless of how their recent submittals have been performing, the program's spend trajectory and the scorecard's performance signals start to diverge.

Volume concentration makes sense as a starting point. It becomes a cost problem when it's the only criterion you're using to route reqs.

Where the Cost Drift Hides

The mechanism is straightforward. When a vendor submits five candidates for a role and one passes the hiring manager review, the submittal-to-hire ratio for that req is 5:1. When a different vendor submits two candidates for a similar role and both advance to the first interview, the ratio is 2:1. The second vendor's work costs the program less: in screening time, in hiring manager hours, in coordination overhead spent on candidates who don't advance.

Most MSP program scorecards don't surface this comparison. They capture fill rate and time-to-fill. They don't routinely capture submittal-to-interview conversion, interview-to-offer rate by vendor, or the ratio of hiring manager hours consumed per hire. Those numbers, accumulated across a high-volume IT contractor program, are where cost-per-hire diverges from what the rate card predicts.

A vendor submitting at 8:1 against a program average of 3:1 is generating real cost: administrative overhead, panel time, extended req aging when a weak slate returns and the req has to go out again. That cost doesn't appear in the fill rate column. It accumulates in the budget reconciliation.

The Visibility Gap That Lets It Compound

Part of what allows this pattern to persist is a structural visibility problem in contingent workforce management. According to research published by HRO Today, more than half (56%) of organizations have limited or no visibility into their contingent labor and external spend. When organizations can't see where spend is actually going and why, they can't identify which part of the program is generating the drift.

A procurement team watching fill rate and time-to-fill stay green while cost-per-hire climbs has no obvious diagnostic path from those numbers. Finding the driver requires a different set of metrics: submittal efficiency by vendor, interview-to-offer rate by vendor tier, req-aging data broken down by first-submittal quality. Most VMS implementations don't surface those by default.

This is distinct from maverick spend. The program is managed. The vendors are on contract. The cost is still leaking, just through a mechanism the scorecard was never built to catch.

What Structured Screening Does to the Math

The simplest intervention at the submittal stage is requiring vendors to document what screening actually happened before a candidate entered the VMS. Not "reviewed and approved for submission," but structured criteria: were the required technical skills verified through a structured conversation? Were availability, rate expectations, and location constraints confirmed before the submittal was entered?

Vendors who apply this consistently produce submittals that convert at a higher rate. Not because they're better at guessing which candidates will pass, but because more of the obvious disqualifiers are caught before the submittal reaches the hiring manager. That shift in submittal quality reduces the total number of submittals the program processes per hire. It reduces hiring manager review load. It compresses req aging for IT roles, which carry the highest submittal-to-hire ratios in most contingent programs because technical criteria are specific enough that unstructured screening fails at a consistently higher rate.

The cost reduction doesn't show up in fill rate or time-to-fill. It appears in cost-per-hire, hiring manager capacity, and the req-aging curve for your hardest-to-fill IT contractor categories.

The Hackett Group's benchmarking research on procurement performance found that organizations with the strongest spend controls and compliance processes experience 59% less savings erosion from buying and contract noncompliance compared to their peer organizations. The same pattern holds inside contingent programs: the variance in cost efficiency is driven more by what happens at the submittal and screening stage than by what's in the rate card.

What to Ask Before the Next Scorecard Review

If your current scorecard doesn't capture submittal-to-interview conversion rate by vendor, that's the first gap to close. This metric can be calculated from data already in your VMS: submittals entered versus candidates advanced to the first interview stage. It reveals vendor performance variation that fill rate and time-to-fill cannot.

The second question is for your Tier 1 vendors: what does their screening process look like before a candidate enters the VMS? A structured answer, one where criteria are documented and confirmation steps are defined, is a different thing from "our recruiters review the JD and assess fit." The first can be audited. The second is a description of effort, not a screening process.

The third question is about whether your MSP scorecard currently rewards submittal volume or submittal quality. Programs that reward responsiveness, submittals within 24 hours of req opening, without pairing that metric with conversion rate create the conditions for the cost drift described here. Faster is not better when most of what's submitted doesn't advance.

The vendors at the top of your tier structure got there by performing well at scale. That history matters. What it doesn't tell you is whether their performance at scale is still producing the cost efficiency your program was built to deliver. Submittal quality data is the diagnostic your scorecard is missing. Tracking it is where the fix begins.

Want to see how structured screening changes your submittal-to-hire ratios? Book a pilot and we'll run your next role through the Eximius workflow.