The Rule of 40 Still Applies to AI Startups but the Math Changes

Yes, the Rule of 40 still applies to AI startups, but the bar is harder to clear. The metric itself does not change. What changes is the profit side of the equation, because inference and compute costs sit inside gross margin and pull it down. That forces AI-native founders to lean harder on growth to reach the same combined score of 40.

Does the Rule of 40 Apply to AI Startups?

Yes. The Rule of 40 applies to AI startups just as it does to any software business, and it stays a useful shorthand for balancing growth against profitability. The rule says that a company's revenue growth rate plus its profit margin should add up to at least 40. A company growing 60 percent a year while running at a 20 percent loss still scores 40, and so does a company growing 20 percent with a 20 percent profit margin. The reason it matters more for AI companies is that their gross margins tend to be structurally lower than classic software-as-a-service, so the profit half of the equation starts from a weaker position and growth has to make up the difference.

What the Rule of 40 Actually Measures

The Rule of 40 is a balance test, not a growth target or a profit target on its own. It became a standard health check for software businesses around the middle of the last decade after investor Brad Feld wrote about it, and firms that study software companies, including McKinsey and Bessemer Venture Partners, have used it since. The formula is simple. Take your year-over-year revenue growth rate as a percentage, add your profit margin as a percentage, and check whether the total reaches 40.

The profit side can be measured a few ways. Common choices are EBITDA margin, free cash flow margin, or operating margin. The exact definition matters less than using the same one consistently every quarter. The point of the rule is to stop founders and investors from celebrating growth that is bought with unsustainable losses, or from praising profitability that comes at the cost of no growth at all. A healthy software company is allowed to trade one for the other, as long as the two numbers together clear the bar. In practice, sustaining a score above 40 year after year is the exception rather than the norm, which is exactly why it works as a filter.

Why AI Margins Change the Math

Here is the AI-specific twist. Classic software companies serve a product that costs almost nothing to run for each additional customer, which is why their gross margins commonly land between 60 and 80 percent or higher. AI-native companies are different. Every model call has a real variable cost, and heavy products can spend meaningfully on inference, GPUs, third-party model APIs, and human review. When those costs are booked correctly as inference inside cost of goods sold, gross margin falls.

Andreessen Horowitz documented this pattern in its widely cited 2020 essay, The New Business of AI, reporting that the AI companies it studied ran gross margins in the range of 50 to 60 percent, well below the 60 to 80 percent or higher typical of comparable software-as-a-service businesses. That gap is the whole story. A lower gross-margin ceiling means the profit half of the Rule of 40 has less room to work with, so AI startup gross margins near 50 to 60 percent quietly raise the growth rate you need just to stay healthy.

The Growth Premium AI Startups Pay

Work through a simple illustration. Imagine two companies that both want to post a 20 percent profit margin. The classic software company with an 80 percent gross margin has plenty of headroom to fund sales and marketing and still keep 20 percent of revenue, so to score 40 it needs to grow only about 20 percent. The AI company with a 50 percent gross margin has far less left after serving each customer, so reaching that same 20 percent profit margin is much harder. If its realistic margin is closer to breakeven or even a small loss, it may need to grow 40, 50, or more percent a year just to keep the combined score at 40.

This is why growth-stage discipline looks different for AI founders. The same Rule of 40 score demands a higher growth rate, which usually means more spend on go-to-market, which pressures the burn multiple at the same time. The trade-off is real, and pretending AI economics mirror software economics is how founders end up surprised by their own numbers on a board slide.

When the Rule of 40 Does Not Apply Yet

One honest caveat. The Rule of 40 is built for companies that already have real, recurring revenue at some scale, usually from a Series A onward. A pre-revenue or very early seed-stage AI startup cannot meaningfully score itself against it, because a tiny revenue base makes growth rates look enormous and margins swing wildly on a single contract. At that stage the rule is a direction to grow into, not a test to pass.

Founders should still track both inputs early, though, because the habits that protect gross margin are far cheaper to build in from the first line of code than to retrofit once the cost base is locked. Honest COGS accounting and disciplined model spend are exactly those habits. Treat the Rule of 40 as the standard your unit economics should converge toward as revenue becomes real, and let it shape the product and pricing decisions you make long before you formally clear it.

How to Apply the Rule of 40 to an AI Startup

• Book honestly. Put inference, GPU, model API, and human-in-the-loop costs in cost of goods sold, so your gross margin reflects reality rather than a software-era assumption.
• Pick one profit definition. Use EBITDA, operating, or free cash flow margin, and hold it constant across every board update so the number stays comparable.
• Read growth and margin together. A score of 45 built on 60 percent growth and a 15 percent loss is a very different company than a 45 built on 20 percent growth and a 25 percent profit.
• Watch the trend, not just the level. A score that rises as you scale matters far more than a single quarter that happens to clear 40.
• Defend the margin on purpose. Model routing, caching, and prompt discipline can lift gross margin over time, and every point of margin you recover relaxes the growth you are forced to chase.

Where a Venture Studio Fits

The reason the Rule of 40 is worth internalizing early is that AI-native margin structure is a design decision, not just an accounting outcome. Which models you route to, how aggressively you cache, and whether you fine-tune a smaller model instead of paying for a frontier one on every request all move gross margin, and gross margin is what sets how hard you have to grow.

That is the trade-off Avante Ventures engineers from day zero. As a venture studio that co-founds AI-native companies for Brazil and LATAM, Avante builds cost discipline into the product before the first customer, so the margin side of the Rule of 40 starts stronger and the growth premium a founder has to pay is smaller. A studio that has already lived through inference bills and model-routing decisions across a portfolio brings that pattern to each new company rather than letting every founder rediscover it the hard way.

The rule still applies to AI startups, and it stays worth taking seriously. What has changed is the difficulty setting. Thinner AI gross margins mean the same score of 40 asks more of your growth engine, so the founders who win are the ones who treat margin as a product decision from the very start and make the math winnable before the numbers ever reach a board slide.

Preguntas frecuentes

Does the Rule of 40 apply to AI startups?

Yes. The Rule of 40 applies to AI startups the same way it applies to any software business. It just tends to be harder to clear, because AI gross margins are structurally lower than classic software margins, which means growth has to carry more of the combined score.

How is the Rule of 40 calculated?

Add your year-over-year revenue growth rate to your profit margin, each expressed as a percentage. If the two together reach 40 or more, you pass. The profit margin can be EBITDA, operating, or free cash flow margin, as long as you use the same definition consistently.

Why do AI startups have lower gross margins than SaaS?

Because serving each customer costs real money. Inference, GPUs, third-party model APIs, and human review are variable costs that belong in cost of goods sold. Andreessen Horowitz reported AI gross margins around 50 to 60 percent versus 60 to 80 percent or higher for comparable SaaS, and that gap is what pushes the growth requirement up.

What is a good Rule of 40 score for an AI startup?

The bar is the same 40, but the path there is different. Because margins start lower, an AI company usually needs a higher growth rate to reach 40 than a SaaS company would. Track the trend as you scale, and defend gross margin actively so the growth you need stays realistic.