Usage Based Models: Why They Suck as a Customer

Speed meets control. Finally.

As a CFO, I've always hated how much of my day is choosing between control and speed. You either get real-time visibility or you get your team moving fast… never both.

My Brex card is different. Brex gives me the controls I need without slowing down important expenses, like branded swag, a new podcast mic, or Walter’s ice cream stipend. With Brex, receipts are automated, expenses hit in real time, and categorization doesn't require three people and a spreadsheet. I can see exactly where money's going while we focus on growing the Mostly Metrics world empire.

Being smart and being fast aren’t mutually exclusive.

See why 30,000+ global businesses, including 1 in 3 US startups and 250 public companies, use Brex to spend smarter and move faster:

On paper, usage-based pricing sounds like the pinnacle of modern SaaS. Flexible. Elastic. Customer-friendly.

In practice? It can be a trap.

One founder recently shared how their team was sold a hefty annual commitment for cloud infrastructure. Halfway through the year, it became clear they were only using about 50% of it. When they approached the vendor with a simple ask: “Can I pay for what I actually use?”… they were met with stiff resistance.

He goes on to describe the interaction:

This wasn’t an isolated rant. It’s the quiet part of usage-based pricing, said out loud.

And it's true - as customers we asked for a system where we only pay for what we used. Like a taxi meter.

But in SaaS land? You’re still charged for a ride to Hoboken, even if you get out in Midtown.

Instead we got all this red tape around when you have to use it, and it's all slanted towards protecting the downside of the vendor.

I’ll say that one more time:

Yes, in theory it’s the pricing we requested. But in reality, it’s not a symmetrical exchange of risk.

It all comes to a head when you are forced into renewing in order to use what they said you wouldn't lose.

If you’ve ever been on the other end of a commitment gone bad, it follows an all too familiar playbook:

The math is simple (and only upside for the vendor). You either burn the unused credits or re-up for more than you need just to keep access to what you already paid for.

That’s not flexibility. It’s infrastructure hostage-taking.

Why do vendors love this model so much? Because it props up Net Dollar Retention (NDR) like a toddler sitting on top of two Yellow Pages to reach the dinner table.

That “130% NDR” stat isn’t always driven by product love or organic growth; it’s often enforced contractually. You renew high, or lose your sunk costs.

Yet it gets mistaken in a financial sense for “durable revenue.”

The concept of durability in technology is characterized by a product or service becoming almost indispensable due to its utility, integration, and the significant role it plays within its user base.

Like a Bloomberg terminal. It’s not going away.

Many of these usage based infra and AI contracts might be able to squeeze a few temporary net dollar retention points out of a longer contract duration; throw in some ramped usage in years two and three and you’re looking pretty! But that’s not the same essence of durability.

A handful of major public SaaS companies were exposed in 2022 and 2023, with NDR dropping sometimes more than 30 points.

So make no mistake - if your product is not “durable” in a qualitative sense, we eventually see who’s swimming naked when the tide goes out.

For some usage based companies, NRR is a fleeting metric built less on usage and more on breakage.

Let the mistakes of other operators be your starting point next time you find yourself in a usage based negotiation. Here’s your checklist:

Worth restating: don’t buy the “you can always use it later” line unless it’s spelled out clearly in the agreement.

Usage-based pricing isn’t broken, but the way it’s implemented often is.

It’s the reason why we can’t have nice things.

I recently wrote about how SAFEs are destroying cap tables. They were a tool built for one purpose, and distorted to fulfill another.

The same can be said about usage based pricing.

When flexibility is used as bait, and breakage becomes the business model, it’s no longer a partnership… it’s hide the ball.

And when vendors optimize for short-term metrics over long-term trust, don’t be surprised if customers walk the first chance they get. Like the angry reader above, eventually you’ll see customers churn out the top, no matter how “mission critical” you are.

Ed Park was the first FP&A hire at Facebook. He was working for Mark Zuckerberg when servers were literally melting off the walls.

In this episode, I jam with Ed Park, Head of FP&A at Facebook during its explosive hypergrowth years, Head of Finance and Internal Operations at Asana, the current CFO of Daffy, and a “glass-half-full” finance leader.

We discussed:

The conversation covers how a company's business model shapes its destiny, finding the balance between what’s urgent and what’s important, and Ed’s glass-half-full approach to being a finance leader who enables teams by saying yes.

Ever notice how the CFO title keeps swallowing up the responsibilities of the COO?

I’ve had guests come on the pod and ask me to tweak their title after recording — CFO to COO, COO to CFO. Sometimes it feels like the roles are doing the cha-cha on LinkedIn.

CFOs are in far more demand. And yet, many CFOs secretly (or not so secretly) covet the COO badge.

Why?

Maybe it’s about optics. COO sounds more “in the mix,” less back-office. Maybe it’s about influence — COOs get ops, GTM, and people stuff. But in modern tech orgs, especially where capital is tight and growth is scrutinized, the CFO is driving strategy too. And strategy is the new operations.

So yeah — in 2025, it’s not that the CFO wants to be COO. It’s that they already are.

Wishing you a usage based contract based on what you actually use,

CJ