Key takeaways
Most enterprises underestimate contingent workforce costs because significant spend occurs outside the VMS through rogue engagements, SOWs, and unmanaged suppliers.
Cost leakage comes from several sources at once, vendor sprawl, off-program spend, weak rate card governance, compliance risk, and operational inefficiency, and they compound each other.
True spend visibility requires tracking every contingent category: staffing agencies, independent contractors, AOR, EOR, and SOW engagements.
Effective cost control starts with visibility and governance, then moves to supplier consolidation, SOW oversight, continuous monitoring, and stronger compliance processes.
If you manage a contingent workforce program or own the external workforce spend category, you already know the number your leadership sees is not the real number.
The VMS report looks clean. The QBR deck shows approved spend against rate cards. But somewhere between the requisition and the invoice, money is moving outside the program, through hiring managers who went around the process, engagements with contingent workers that nobody classified properly, and suppliers who drifted off rate while nobody was watching.
This article is for the people responsible for fixing that. Contingent workforce program managers who need full visibility to run a tight program. Procurement leaders who need hard data to justify consolidation and negotiate from strength. If that's you, here's where the money actually goes, and how to get it back under control.
Why contingent workforce spend is harder to control than you think
The illusion of control: rate cards vs. reality
Rate cards feel like governance. They set agreed markups, standardize billing, and give you something concrete to point to in a supplier review. But a rate card only controls spend it can see.
As a program manager, you've probably run the numbers on your approved supplier invoices and felt reasonably confident about where things stand. The problem is that your approved suppliers aren't the only ones being paid. Hiring managers working outside the VMS, contractors engaged directly by business units, Scope of Work (SOW) agreements signed through legal without touching procurement, none of that flows through your rate card framework.
Procurement managers face the same problem from a different angle. Category spend analysis looks manageable until you start pulling AP data and comparing it against VMS records. The gap between the two is where your negotiating leverage is quietly disappearing.
Why contingent workforce spend is disconnected by design
The contingent workforce wasn't built as a unified system. It grew over decades as business units solved individual talent problems with whatever was available, a staffing agency for temps, a freelance platform for project work, a few SOW-based agreements with consulting firms that got renewed automatically.
The result is a patchwork that nobody designed and nobody fully owns. The VMS captures some of it. HRIS has another slice. AP has invoices that don't map cleanly to either. Reconciling these sources into one accurate number is a manual exercise most programs don't have the bandwidth to complete properly.
For a program manager, this means the headcount data you report to leadership is probably understating the real picture. For procurement, it means the spend-under-management figure you present in category reviews is missing a material portion of actual external labor costs.
The shift from staffing spend to total contingent workforce spend
"Contingent workforce spend" used to mean staffing agency invoices. That definition no longer reflects how enterprises actually engage external talent.
A complete picture of contingent spend today includes:
Temporary workers sourced through preferred staffing agencies
Independent contractors engaged directly or through AOR providers
Statement of Work and project-based engagements
EOR fees for globally engaged talent
Platform-sourced talent that bypasses traditional supplier channels
Internal program administration, headcount, VMS licensing, and overhead
Program managers who only track agency invoices are typically counting roughly half the total cost of their contingent workforce. Procurement managers who haven't mapped AOR and EOR spend into their category analysis are presenting incomplete numbers to leadership.
The visibility gap is bigger than most programs expect
According to research from Ardent Partners, unmanaged or off-program contingent spend regularly accounts for 20–40% of total external workforce costs at large enterprises. For a program spending $50M annually, that's $10–20M operating outside approved channels, rate structures, and compliance guardrails, and outside your ability to govern it.
What counts as total contingent workforce spend?
Before you can control spend, you need an honest accounting of everything you're responsible for. Here's what that full picture looks like, and where each category tends to fall through the cracks.
Spend Category | Who Typically Tracks It | Frequently Missed? |
|---|---|---|
Staffing agency labor (temp employees, contractors) | VMS / Program Manager | Partially, off-preferred-list agencies often missed |
Independent contractors (1099 / IC / freelance) | AP or direct business units | Often, bypass VMS entirely |
SOW / project-based engagements | Procurement / Legal | Frequently, coded as services spend, not workforce |
AOR fees (Agent of Record) | Varies, often no clear owner | Often, newer model, not mapped in most programs |
EOR fees (Employer of Record) | HR / Global mobility | Often, sits outside traditional program scope |
Internal program management costs | Overhead / G&A | Almost always excluded from program spend totals |
Visible spend is what shows up in your VMS dashboard and gets discussed in supplier QBRs. Actual spend is the sum of every row above, including the ones that have no clear owner and never appear in your reports. For most programs, the gap between those two numbers isn't small.
Where contingent workforce cost leakage actually happens
Cost leakage isn't usually a single problem with a single fix. It's several problems running at the same time, and they compound each other.
When business units have access to too many suppliers, or add new ones without going through procurement, the volume that drives your negotiating leverage gets split across too many relationships.
Two business units hiring for nearly identical roles can end up paying meaningfully different rates because they're each using agencies they've worked with independently, under different contract terms. Your preferred supplier list exists on paper. In practice, the supplier ecosystem is larger than anyone has formally mapped.
For procurement managers, this is a direct hit on category savings targets. For program managers, it's a governance problem: you can't audit what you don't know exists.
Maverick spend, hiring managers engaging contractors or agencies without going through the VMS or the approved supplier list, is one of the most consistent sources of cost leakage across enterprise programs.
It happens for understandable reasons. The approval process is slow. The hiring manager has worked with a particular niche agency for years and trusts them. They need someone in 48 hours and the requisition workflow takes a week. So they go around it.
The consequences land on you, not them:
Off-contract pricing that your rate cards can't touch
Co-employment and misclassification risk from workers with no classification review
Hidden headcount that skews your reporting to leadership
No audit trail if a compliance question arises later
The program manager's challenge is making the approved process fast and easy enough that going around it isn't worth the effort. The procurement manager's challenge is finding the rogue spend that's already happened and quantifying the cost impact.
Governance of SOW-based engagements — where third-party vendors manage workers to deliver contractually defined outputs rather than the company managing its own workforce — is where procurement programs tend to have the biggest blind spot. Statement of Work contracts are structured as project deliverables, so they get routed through legal, coded as services spend rather than workforce spend, and never touch the VMS.
The complication is that a significant share of SOW engagements are functionally staff augmentation with a different label. The supplier provides named individuals, those individuals work alongside internal teams, often on-site or within internal systems, and bill on a time-and-materials basis. Because the engagement is classified as a project, it can bypass your rate cards, your classification review, and your headcount reporting.
For procurement managers, this represents spend that should sit inside your governance framework but doesn't. For program managers and in-house legal, it can carry co-employment considerations sitting just outside your line of sight.
Rate cards don't enforce themselves. The most common failure modes:
Rate cards benchmarked against market data that's well out of date, while market rates have shifted
No automated workflow to flag or reject requisitions that exceed card rates
Rate structures that cap base pay but not total bill rate or markup
Different rate cards across regions or business units with no central reconciliation
Without system-enforced controls, rate cards become the starting point for negotiation rather than the ceiling. Urgency from hiring managers creates the opening, and rate drift is the result.
Your VMS knows what was requisitioned and approved. Your ERP knows what was invoiced and paid. Your HRIS may or may not carry any record of contingent talent at all. These three systems typically don't reconcile cleanly, which means an accurate end-to-end picture of your contingent workforce requires manual work most programs don't have time for.
The consequence for program managers: headcount reports to leadership are probably understating the real picture. For procurement managers: spend-under-management metrics are based on incomplete data, which affects both how you're evaluated and your leverage in supplier negotiations.
For contingent workforce managers, misclassification is the risk that keeps them up at night. Treating an independent contractor in ways that resemble employment — providing equipment, setting work hours, supervising the work directly — can create exposure under frameworks such as IR35 in the UK, the FLSA in the US alongside overlapping state wage and hour laws, and equivalent rules across most major markets where global enterprises engage talent.
The potential financial consequences of getting this wrong can include back taxes and social-security contributions, benefits liability, fines, unpaid overtime, statutory penalties, and in some cases personal liability for those who approved the engagement. No single factor is determinative; classification generally depends on the totality of the circumstances, and the specific tests vary by jurisdiction.
For procurement managers, this translates to exposure that's difficult to quantify in advance and very visible after the fact. A single misclassification matter in a high-risk jurisdiction can produce a liability that dwarfs the cost savings the engagement was intended to deliver.
The fully loaded cost of running a contingent workforce program includes more than supplier invoices. Factor in:
Internal headcount dedicated to managing the program and supporting VMS adoption
Time spent on manual data reconciliation across systems
Legal review of non-standard or off-template contracts
Idle contractor time caused by delayed IT provisioning or background-check clearances
That last one is worth examining. When a contractor sits idle for a week waiting for equipment or clearance, you're paying for capacity you can't use. Multiply that across hundreds of annual engagements and it becomes a material line item that never appears in a cost report.
The real impact: how leakage affects contingent workforce spend
The seven sources above don't operate independently. They compound. Off-program spend makes rate governance harder because the volume that gives you leverage is fragmenting. Data fragmentation makes compliance harder to demonstrate during audits. Weak SOW governance increases co-employment considerations. Operational inefficiencies erode the program's internal credibility, which makes hiring-manager buy-in harder to maintain — which feeds more rogue spend.
The figures below are widely cited industry estimates, not Lifted data, and the actual impact varies by program:
Leakage Source | Estimated Impact (industry estimates) |
|---|---|
Vendor sprawl / markup inflation | Reported by industry sources at ~5–15% above optimized rates |
Rogue and off-program spend | Industry sources note 20–40% of total spend can be unmanaged |
SOW mismanagement | Industry estimates of ~10–20% of services spend ungoverned |
Weak rate card enforcement | Industry sources note meaningful year-over-year rate drift |
Misclassification-related penalties | Variable — potentially significant per jurisdiction |
Operational inefficiencies | Significant indirect cost, rarely quantified |
The sum of these isn't a rounding error. For most mid-to-large enterprise programs, it can represent millions of dollars that the program is theoretically responsible for but has no current mechanism to recover.
How to get started: a simple 5-step spend control framework
You don't need to overhaul the program overnight. Spend control follows a sequence, and the sequence matters.
Map total contingent workforce spend. Pull data from every system with a record of contingent labor, VMS, HRIS, AP, ERP, procurement platforms, and any global payroll systems handling EOR or AOR engagements. The goal is one number: what did the enterprise actually spend on contingent talent across all categories in the last 12 months? The first pass requires manual reconciliation, that's expected. What you're building is a baseline that captures all spend categories, not just what flows through the VMS.
Identify off-program and rogue spend. Compare your complete spend map against your approved supplier list and VMS records. Any spend that didn't flow through approved channels is rogue spend. Quantify it by business unit, supplier, and category, then understand why it happened. Most rogue spend is predictable: the same business units circumvent the program for the same reasons, usually slow cycle times, limited supplier options for niche skills, or a frustrating VMS experience. That diagnosis tells you where to fix the program, not just where to chase down spend after the fact.
Consolidate vendors and standardize rates. Use your spend data to build the case for supplier consolidation. Look for multiple agencies covering similar skill sets or geographies at different rate structures, low-volume relationships that generate no leverage, and suppliers in regions where your preferred partners have equivalent capability. Enforcing rate card compliance across 40 suppliers is a full-time job. Enforcing it across 8 preferred partners is a process.
Bring SOW under governance. Start with a clear, written definition of what constitutes an SOW engagement versus staff augmentation. Apply it to every active SOW and build it into the intake process for new ones. Any engagement involving named individuals, time-and-materials billing, or direct management by an internal stakeholder generally warrants treatment as contingent workforce spend, regardless of how the contract is structured on paper. A light-touch competitive bidding requirement in the approval workflow creates a checkpoint that reduces hidden margins and keeps scope from drifting.
Implement continuous monitoring and reporting. Spend control isn't a project you complete; it's a cadence you maintain. Build a monthly or quarterly review that tracks:
Total contingent spend vs. prior period, across all categories
Off-program spend as a percentage of total
Rate card compliance by supplier and business unit
SOW spend as a percentage of total contingent spend
Supplier SLA performance scores
Classification audit flags and open risk items
Time-to-fill trends across preferred suppliers
The objective is early warning. Catching rate drift in month two is a supplier conversation. Catching it in month fourteen is a budget problem.
Most contingent workforce savings start with visibility and control
The best-run contingent workforce programs aren't the ones with the most suppliers or the most complex technology. They're the ones where the program manager and the procurement lead work from the same complete data set, and where every dollar of external labor spend is visible, governed, and traceable.
That kind of visibility is what makes every other part of the job easier. Rate negotiations. Supplier consolidation. Compliance audits. Reporting to leadership. All of it depends on knowing what you're actually spending.
Supplier selection is part of that equation too. Lifted, an Upwork Company, is a tech-enabled contingent workforce supplier whose proprietary technology automates the sourcing, classification, and engagement workflows that traditional agencies handle manually. That operational efficiency passes directly to clients, engagements that are 10–30% more cost-effective than standard legacy suppliers. And because Lifted plugs directly into your existing VMS or MSP program with zero disruption, adding us to your preferred supplier list doesn't require a program overhaul. It just requires onboarding one supplier.
Spend control starts with one accurate number. Build that baseline first. Everything else follows.
Frequently asked questions
What is contingent workforce spend and what does it include?
Contingent workforce spend covers the total cost of engaging non-permanent workers: agency-sourced temps and contractors, independent contractors, SOW and project-based engagements, EOR and AOR fees, and the internal cost of running the program. Most enterprises undercount because they only track what flows through the VMS.
What is maverick or rogue spend and why does it happen?
Rogue spend occurs when hiring managers engage talent outside the approved program — directly through niche agencies, freelance platforms, or personal networks. It typically happens because the approved process is too slow or limited. The result is off-contract pricing, hidden headcount, and compliance exposure that falls back on the program manager to resolve.
What is the difference between visible spend and actual spend in a contingent workforce program?
Visible spend is what's captured in VMS reports and reviewed in supplier QBRs. Actual spend includes everything else: off-program hiring, direct IC engagements, SOW, EOR and AOR fees, and internal program costs. The gap between the two is where most unmanaged risk and budget leakage sits.
How can enterprises reduce contingent workforce cost leakage?
Start by mapping total spend across every system, then identify off-program spend, consolidate vendors to standardize rates, bring SOW engagements under governance, and maintain continuous monitoring. Visibility and governance come first; consolidation and savings follow.
Author
Lee Willoughby
Senior Marketing Director, Lifted
Lee Willoughby is the Senior Marketing Director at Lifted, an Upwork company helping enterprises source, engage, and manage contingent talent across every contract type. With a background as a co-founder and workforce technology entrepreneur, Lee focuses on the future of contingent workforce management, helping organizations navigate the complexities of global talent, compliance, and workforce transformation.
This content is for general informational purposes only, and is not intended to be and should not be viewed as legal or tax advice. Readers should contact their attorney or tax professional to obtain advice with respect to any particular legal or tax matter. Information discussed can change frequently, and Lifted cannot guarantee that all information is current at all times.
