The Asymmetry Problem
Four places your enterprise contracts earn the vendor more than you think they do. Each walks through the math, and ends with a check you can run against your own contracts tonight.
Every enterprise contract has two sets of numbers inside it. The gap between them is what the vendor earns, quietly, over the life of the deal.
Yours, built from budget and last year's rate card. Theirs, built from investor promises, margin mix, and sales-team comp. Both price the same contract. They arrive at different totals.
Buyers negotiate inside their own set of numbers. That works only if the vendor's economics look like yours. They don't. In every category below, the buyer negotiates hard against the one number they can see. The vendor concedes because it isn't where the real money is anymore. The money moved somewhere else: a clause, a line item, a commitment mechanism, or an invisible markup the vendor's CFO is measured on.
Your SaaS vendor's CFO publicly promised investors to grow your spend 20% a year.
You sign a three-year deal with a big SaaS vendor. You push the price down 25%. You sign. Procurement books the savings. Two numbers matter: the price you just negotiated, and a metric called "net revenue retention" (NRR) the CFO talks about every quarter on their public earnings call.
NRR is simple. Of the customers the vendor had twelve months ago, how much is that same group spending today? Leading enterprise SaaS companies report 115% to 125%, meaning the existing base pays 15% to 25% more this year than last. That's not a number on a slide. It determines what the company is worth.
How? The contract is full of mechanisms, engineered in by the vendor's pricing team before your negotiator walked in. Two examples.
You buy 500 seats at your discounted price. By year two, you're at 650 seats. The contract has a "true-up" clause: extra seats bill at list, not at your discount. Those 150 extra seats come in above your discounted rate, so you've added more than 30% to your annual spend without renegotiating.
Year two, the vendor announces their AI assistant (previously included in your tier) is now a separately priced add-on at $20 per user per month. Your tier price didn't go up. The scope of your tier got smaller.
Each of these is a recovery mechanism. Procurement optimized for a one-time event (signing). The vendor optimized for a three-year compounding outcome (harvesting). The discount lives on page two of the order form. The 20% compounding recovery lives in the schedules and the master agreement nobody re-read.
Find every clause that lets your spend grow without a new negotiation
Take your top 20 SaaS agreements by spend. For each, list every clause in these five buckets:
- Clauses that charge more as your team grows ("true-up," "additional seats").
- Pricing that steps up when usage crosses a threshold (API calls, GB stored, transactions).
- Features that could quietly move from "included" to "priced add-on" at renewal.
- Renewal caps (like "price won't rise more than 5%") with exceptions or carve-outs.
- Modules where the price is protected but the included scope isn't.
Your industrial distributor makes its real money on the items you didn't negotiate.
Your maintenance team is on the factory floor. A machine breaks. They walk to the Grainger (or Fastenal, or MSC) vending machine bolted to the wall, find the part, and order it. Nobody in procurement saw the order. Multiply that by thousands of orders, across dozens of sites, across a year. That's the tail.
Two numbers matter: prices you negotiated on specific parts (your "contracted items"), and catalog prices the distributor charges for everything else.
Catalog margins are wider than contracted margins by design. The public filings of the big distributors are explicit about the pattern.
They're not hiding this, and they're building infrastructure that makes it harder to leave. Fastenal has roughly 134,000 FMI devices (vending machines and smart bins) placed inside customer facilities. Once the hardware is on your wall, the path of least resistance is to grab whatever is in the box at whatever price is in the box.
Concrete example. You negotiated $8.50 per box on a specific brand of nitrile gloves. Good win. But your team also buys 37 other glove SKUs from the same distributor at catalog. Similar items on catalog commonly run 15% to 25% higher than contracted pricing. Your negotiated savings on the one SKU is real. The unnegotiated spend on the other 37 is larger, and the distributor kept the margin there.
Sort every line into two buckets, then audit the biggest catalog items
Pull twelve months of invoice-level data from your top industrial distributor. Sort every line into:
- SKUs on your negotiated price list.
- SKUs bought at catalog.
Cap the agency's markup, and the margin moves somewhere you aren't looking.
You need a software developer for six months. You call a staffing agency. You pay a bill rate. The agency pays the contractor a wage. The agency keeps the difference, after payroll taxes, benefits, and overhead. Three numbers matter: bill rate (what you pay), pay rate (what the contractor earns), and the agency's cost of employing them.
Most procurement teams try to lower the bill rate by capping the agency's "markup," the percentage added on top of the wage. A 40% markup on a $50/hr contractor makes the bill $70. Procurement caps markup at 30%. Bill drops to $65. Savings booked.
The quality story is downstream of both. The agency's best contractors get steered toward accounts without a markup cap, where margins sustain on pay rate alone. Your role, now less profitable per hour, gets filled from a less experienced bench. Procurement sees bill rates come down further than the cap required and books the win. Engineering gets weaker talent and a slower ramp, on a different budget.
Pull 50 contractor records and line up four numbers
- Bill rate · what you paid.
- Pay rate · what the contractor earned. The agency will resist sharing, and the number you get may be inflated.
- Pass-through and non-markup charges · burden, benefits, admin, PTO recovery.
- Time to fill and tenure in the seat.
Your ad agency stopped making money on the fee twenty years ago.
You hire a big ad agency. They do two things: plan and create ads (labor, billed as a "fee" or "commission"), and buy the ad space with your money (the "media spend," which flows to Google, Meta, TV networks, publishers). Two numbers matter. The fee and the media spend.
For twenty years, buyers have squeezed the fee. The big holding companies (WPP, Omnicom, Publicis, IPG) responded by building a second revenue engine that's mostly invisible to you.
Instead of buying ads on your behalf (as your "agent"), the agency buys inventory itself, with its own money, then resells it to you at a markup. That's "principal media." The invoice says "$1M media spend." The agency bought it for $700k. The $300k difference is their margin, hidden inside the media line.
They also get rebates from the platforms: spend $100M a year and the platform quietly writes the agency a year-end check. That money goes to the agency, not you.
In late 2025, a former WPP executive filed a whistleblower lawsuit alleging GroupM (now WPP Media) generates close to $1 billion a year in "non-product related income" from these mechanics. Industry analysts estimate Omnicom's principal trading business alone at around $3 billion in annual revenue.
You negotiated the fee down. The agency let you. Most of their margin moved into a part of the relationship you don't see, don't audit, and don't have contractual rights over.
Pull your MSA and check four things
- Agent vs. principal. Does the contract restrict the agency to acting as your agent, or does it allow principal buying with markup?
- Rebate disclosure. Must the agency disclose rebates from media owners and pass them back?
- Audit rights. Can you audit the agency's books on a campaign? Who pays? Can they refuse on confidentiality grounds?
- Affiliate transactions. If the agency buys from a related company in the same holding group, must they disclose it?
Four categories.
One structure.
The vendor's real economics run on a variable the buyer's attention isn't aimed at. The visible number gets negotiated. The invisible number compounds. None of the four checks above is harder than the work procurement already does. Each takes under an hour per category, and the pattern, once seen, is hard to unsee.
We built Infinity Loop because these patterns repeat across every spend category we've looked at. Aggregate visibility across hundreds of enterprise contracts surfaces what's invisible from inside any single company.
If you want a second set of eyes on your own portfolio, the fastest way in is a 30-minute call. We'll walk through which pattern most likely applies to your spend profile, what a pilot would look like, and what we'd expect to find. If the fit isn't there, we'll tell you.
