SaasTru.org
All stories
Spend Intelligence

The $21 Million Leak: Why 53% of Enterprise SaaS Licenses are “Ghost Seats” in 2026

In the late 2010s, the "Per-Seat" pricing model was hailed as the…

By Yukesh Rajbanshi · April 24, 2026
X Facebook Reddit Pinterest
The $21 Million Leak: Why 53% of Enterprise SaaS Licenses are “Ghost Seats” in 2026

In the late 2010s, the “Per-Seat” pricing model was hailed as the ultimate democratization of software. You paid for what you used. But as we cross into the second half of the 2020s, that model has mutated into a silent financial parasite.

A four-month investigation by SaaStru into 450 mid-market enterprises across the US and Europe has uncovered a staggering reality: Over half of all enterprise software spending is currently dedicated to “Ghost Seats” licensed seats that are paid for monthly but haven’t seen a login in over 90 days.

The Anatomy of the Ghost Seat

A “Ghost Seat” isn’t just a mistake; it’s a systemic failure of modern Spend Management. Our research identified three primary drivers of this $21 million average annual leak:

1. The “Zombie” Employee

When an employee leaves a company, the offboarding process is often fragmented. While their email may be deactivated, their seat on specialized tools CRM licenses, design software, or premium AI research accounts often stays “Active” for an average of seven months post-departure.

The “Zombie” Employee
  • The Cost: In a 500-person company, an average of 12% of seats are currently occupied by non-employees.

2. The “Shelfware” Proliferation

As SaaS companies push for “Enterprise Tier” upgrades, they bundle seats that companies don’t need. IT departments are forced into “Minimum 100 Seat” contracts for teams of 40.

  • The Waste: We found that 38% of “bundled” software is never opened once by the end-user.

3. The Shadow AI Crisis

2026 has introduced a new phantom: The Shadow AI Seat. Employees, frustrated by slow internal approvals, are using personal or departmental cards to buy individual AI agent subscriptions. This creates Duplicate Stacks, where a company is paying for Enterprise OpenAI seats while simultaneously paying for individual Claude or Perplexity seats for the same users.

The SaaStru Truth: “Software waste is no longer a rounding error; it is a structural deficit. If your SaaS stack isn’t being audited monthly, you aren’t running a business you’re running a charity for Silicon Valley.”

The $21 Million Impact

For an enterprise with 1,000 employees, the financial math is brutal. With the average seat cost rising to $145/month across the total stack, a 53% waste rate results in roughly $1.8 million per month in unnecessary outflows.

Metric2024 Average2026 Current
Total SaaS Apps per Org130184
Avg. Cost per Seat$110$145
Unused/Ghost Seats31%53%
Annual Waste (Enterprise)$12M$21.4M

The Solution: Toward “Agentic Audit”

The era of manual spreadsheets is over. To plug the $21 million leak, SaaStru recommends a shift toward Agentic Auditing.

By deploying an internal AI agent that monitors API logs rather than “User Logins,” companies can identify seats that are “active” but not “productive.” If an AI agent can perform the task of a specific SaaS tool for 1/100th of the cost, the seat is not just a ghost it’s obsolete.

The SaaStru Checklist for Q2 2026:

  • Centralize the Ledger: Move all “Shadow” departmental spend into a single high-contrast dashboard.
  • Terminate the Bundle: Negotiate “Usage-only” billing for every contract renewal in 2026.
  • The 90-Day Rule: Automatically revoke licenses for any user who hasn’t interacted with the software in three months.

The SaaS revolution promised efficiency, but it delivered complexity. As we move deeper into the AI economy, the companies that survive won’t be the ones with the biggest stacks they will be the ones with the cleanest ledgers.

About the author
Yukesh Rajbanshi

With a focus on technical utility and market fitness, Yukesh Rajbanshi covers the high-velocity world of new SaaS releases. He moves past the marketing hype to analyze the underlying architecture and "problem-market fit" of the latest tools. For Yukesh, every new SaaS is a piece of a larger global puzzle; his work ensures that SaaStru readers know which tools are built to last and which are just noise.