Revenue and Margin Are the Same Problem

In most companies, pricing and software spend are handled by different people who rarely meet. Pricing sits with product and finance. Software and cloud spend sits with IT, procurement and engineering. They report through different structures, run on different calendars and are measured on different numbers.

They are both working on gross margin. One raises the top line, the other lowers what sits underneath it, and the arithmetic does not care which team moved first.

The separation used to be harmless

When serving a customer cost almost nothing, cost control was an overhead question rather than a margin question. Software spend was something to keep tidy. Pricing was where the interesting decisions were.

That is no longer true where AI and infrastructure costs scale with use. In the State of B2B Monetization in 2026, Kyle Poyar found that the median target gross margin for AI products is about 50%, and only 12% aim for 80% or higher.

At 80% margin, cost is a detail. At 50%, cost is half the business, and a team that cannot see it is managing half a number.

Both sides move the same line

The same research found that 54% of companies name internal costs and margins as the most important factor when pricing AI, ahead of competitor pricing at 36%.

That is a pricing statistic describing a cost input. It is a direct measure of the two disciplines collapsing into one.

The spend side of the equation is not small. Poyar also found that 70% of AI spending comes from software budgets, with AI-native companies drawing from services budgets at 35% and headcount budgets at 15%. AI spend is largely being absorbed into existing software budgets rather than appearing as a new line, which is precisely the condition under which it grows without scrutiny.

In text: about 50% median target gross margin for AI products. 54% name internal costs and margins as the top factor in AI pricing. 70% of AI spending comes from software budgets.

The market is pricing the connection

On 2 December 2025, Stripe announced an agreement to acquire Metronome at a reported $1 billion. Stripe did not disclose terms. Patrick Collison said: "Metered pricing is the native business model for the AI era."

Metronome is usage metering infrastructure. The thing being valued at that price is the ability to measure what customers consume and turn it into a bill. That capability sits exactly at the junction of the two disciplines: it is simultaneously a pricing system and a cost attribution system, because the same measurement that tells you what to charge tells you what it cost to deliver.

One data foundation, two questions

The practical reason these should be handled together is that they need the same inputs.

Pricing needs to know what customers pay, how they bought, what they use and what serving them costs. Spend optimization needs to know what the company pays for, who actually uses it, what the contracts say and when they renew.

Both are a join of financial data, usage data and contract data. The entities differ, with customers on one side and vendors on the other, but the shape of the problem is identical, and so is the reason it is hard: the data lives in separate systems with no shared key.

A company that builds that foundation for one of these questions has most of what it needs for the other. A company that builds it twice, in two teams, with two tools, has paid twice for the same thing and will get two answers that do not reconcile.

In text: 1. Connect billing, usage, contract and cost data 2. Ask what customers should pay 3. Ask what the company should spend 4. Read one gross margin number

What this changes in practice

Three things.

The margin number becomes something one person can explain. When pricing and spend are reported separately, a margin movement is attributed after the fact by whichever team is asked first.

Decisions stop being made in isolation. A pricing change that attracts high-consumption customers is a cost decision. An infrastructure change that reduces quality for the premium tier is a pricing decision. Neither is visible from inside one team.

The review cadence can align. Pricing reviews and renewal cycles both run on the contract calendar, and both benefit from the same freshly joined data.

The practical step

You do not need to reorganise to start. You need one view where revenue per customer and cost to serve that customer appear together, and one meeting where the people responsible for each look at it at the same time.

That view is what StackIQ produces, and it is the reason our pricing work and our spend work rest on the same foundation.

If revenue and margin are being managed separately in your company, our margin optimization work and our pricing and packaging work start from the same connected data.

Related reading: Why AI Gross Margins Sit Near 50% and Companies Create More Value Than Ever but Capture Less of It.