Ask a cloud or data-infrastructure company who owns pricing and the honest answer is usually finance, by default rather than by design. Finance sets the discount ladder, models the enterprise minimum commitment, and reconciles usage against revenue recognition, because that work touches the number the board asks about first. What rarely has a named owner is the decision underneath it: what the pricing model is supposed to signal about the product, which customer behavior it is built to reward, and what happens to margin once usage scales past the assumptions the model was priced on.
Those are strategy questions wearing a finance department's clothes. Left unowned, they get answered by default too, usually in the direction of whatever keeps this quarter's revenue recognition clean rather than whatever keeps the business durable three years out.
Why the mechanics and the model keep getting confused
The mechanics of usage-based billing (metering, invoicing, true-ups, the discount schedule sales negotiates deal by deal) are genuinely finance and revenue-operations work, and they are usually done competently. The confusion starts when the model itself, not its execution, gets treated as settled the day it launches.
A consumption model is a bet about behavior: that customers will expand usage as they get value, that the unit economics hold as usage scales rather than eroding, and that the pricing signals which workloads the product wants more of. That bet needs revisiting on a cadence, the same way a go-to-market plan does, because the answer changes as the customer base changes. A model tuned for a handful of early enterprise accounts prices very differently than one built for a long tail of self-serve usage, and few companies formally re-underwrite the model as that mix shifts. It just keeps running on the assumptions it launched with.
What buyers already assume about your pricing
Buyers of infrastructure and platform spend are not evaluating a consumption model in a vacuum. Flexera's State of the Cloud Report has put wasted cloud spend at 27% of the average budget every year since 2019, holding in that 27% to 32% range across multiple annual surveys. Enterprise buyers read that kind of finding, and it changes how closely they scrutinize a new usage-based line item before they sign it.
That means a consumption pricing model gets evaluated on legibility, not just price: can the buyer's finance team forecast next quarter's bill with confidence, or does the model behave unpredictably enough that it reads as one more line item that might turn into waste. A model that is cheap on average but volatile month to month carries a trust cost that a flatter, slightly more expensive model does not. Pricing strategy that ignores this is optimizing for a number the buyer's own finance team has already been trained to distrust.
Three questions that separate the model from its execution
Before the next pricing review, three questions belong to strategy, not to whoever runs the billing system:
What behavior is this model actually rewarding right now? Not what it was designed to reward at launch. Usage patterns drift, and a model that once rewarded expansion can quietly start rewarding whichever customers are best at gaming the metering unit.
Where does margin actually break as usage scales? Every consumption model has a usage level where the unit economics stop working the way they did in the pricing deck. Knowing that number before a large account hits it is a strategy exercise, not a finance report generated after the fact.
What does this model signal about who the product is for? A pricing structure tuned for high-usage enterprise accounts sends a different message than one tuned for a broad self-serve base, whether or not anyone intended to send it. That signal shapes which customers show up, which is a positioning decision wearing a spreadsheet.
Where this fits
Pricing and packaging decisions for cloud and data-infrastructure businesses sit inside the same discipline as any other durable positioning call: strategy and management consulting is where the model gets re-underwritten against how the business and its customer base have actually changed, not just how the billing system currently runs it. It pairs with marketing consulting for the go-to-market and packaging side of the same decision, and with data strategy consulting where the question is which usage data the model should actually be built on.
If your pricing model hasn't been re-underwritten since it launched, that is worth a direct look before the next renewal cycle forces the question. Start a conversation: we would rather help you find where the model already breaks than let a large account find it for you.