Given Away: AI Features Are Shipping Faster Than Revenue Teams Will Price Them

For most of the last two years, the hard part of AI was building it. That is no longer true. Product teams can now ship a new AI capability in weeks, and many of them have. The commercial side has not kept pace. New research shows a widening gap between what software companies put into their products and what they actually charge for, and the revenue organization sits squarely in the middle of it: sometimes as the team asked to sell the value, and often as the team blocking the price change that would capture it.

Half Are Shipping AI Without Changing the Price

The inaugural 2026 B2B SaaS Price & Value Survey from advisory firm Pricing From the Start, fielded between June 20 and July 10, 2026, found that half of respondents had incorporated AI features while leaving their pricing model unchanged. Another 30% said they had seen no measurable shift in how customers responded to their pricing after launching AI. The sample is small, 30 B2B SaaS leaders, and the firm itself describes the results as directional rather than statistical. But the pattern it describes, which it calls a monetization log jam, matches what larger studies are finding.

More than 30% of respondents described some version of a build-first, price-second sequence, in which a premium feature exists because the product team built something technically impressive rather than because research showed buyers would pay for it. The result is AI that is expensive to run and free to use. Pricing ownership does not help: CEOs or founders own pricing 57% of the time, and the next most common answer, at 27%, was no single owner at all.

When planned pricing changes stall, the most common reason is not engineering. It is the people who sell and renew the product.

The Blocker Is the Revenue Team

The most uncomfortable finding for revenue leaders is why pricing changes stall. Asked why a planned pricing or packaging change gets delayed or cancelled, the top answer in the Pricing From the Start survey was sales or customer success pushback, at 31%, followed by disagreement across executive leadership. Product engineering and infrastructure limitations combined accounted for just 27%. The pressure is understandable. Reps worry that a new price line will slow deals already in flight, and customer success teams worry that asking renewing customers to pay for AI will introduce churn risk.

The survey also shows what fills the vacuum when price is not set from value. Competitor pricing was the single most cited reference point for setting price levels, at 30%, ahead of financial impact on the customer's business at 27% and cost-plus pricing at 20%. When every vendor anchors to every other vendor and most are giving AI away, the market price of AI drifts toward zero regardless of what it costs to deliver.

The Margin Math Has Changed

That cost is not trivial. Kyle Poyar's 2026 State of B2B Monetization report, based on a survey of more than 230 B2B software and AI companies in April and May 2026, found that the median target gross margin for AI capabilities is about 50%. Only 12% aim for 80% or higher. When pricing AI, 54% named internal costs and margins as their top factor, ahead of competitive pricing at 36% and customer productivity gains at 30%. Three in four software companies changed their pricing or packaging in the past year.

An analysis of the same survey by OpenMeter sets that 50% against the 70% to 80% gross margins that traditional SaaS businesses run on, and notes the consequence for packaging: a free tier that was a marketing expense at an 80% margin becomes a real cost line at 50%. Every AI feature bundled into an existing contract at no extra charge pulls the blended margin of that account down, and the revenue team owns the account when finance asks why.

Many companies do not yet know how large that cost is. DigitalRoute's State of AI Monetization 2026 report, which surveyed 631 senior finance, product and engineering leaders across six markets, found that only 8% are fully confident in what their AI features cost to deliver. Some 61% said forecasting AI usage and revenue has become harder over the past year, not easier, and 35% named margin protection as the single biggest driver of their 2026 AI commercial strategy. A year earlier, 80% had said they were ready to monetize AI.

Buyers add one more constraint. The Pricing From the Start survey found usage and volume metrics were the most common primary pricing metric, but argued that per-seat and flat-fee models persist because buyers can forecast them. Buyers can forecast headcount and business volume. They cannot yet forecast tokens. Any AI price that a customer cannot predict will meet resistance at the procurement stage, which is exactly where sales teams feel it first.

The CRO Playbook for Pricing AI

• Take a seat at the pricing table. With pricing often owned by a founder or by no one at all, the CRO should co-own AI packaging decisions rather than inherit them after launch. Sales input belongs before the feature ships, not after it is free.
• Turn pushback into evidence. If reps believe an AI price will cost deals, test it. Win-loss interviews show whether buyers actually rejected a price or whether the loss came down to something else entirely.
• Price for forecastability. Buyers who cannot predict an AI bill will treat it as risk. Package AI in units customers can budget against, such as tiers, bundles or committed credit pools, so the price helps the deal instead of stalling it.
• Use renewal as the re-pricing moment. AI added mid-term to an existing contract is the easiest place to give value away. Map which accounts received new AI capability and plan the renewal conversation around it well before the date.
• Anchor to customer impact, not competitors. Competitor pricing was the top reference point in the survey. Research what AI actually changes for your customers, then price against that outcome rather than a rival's price page.
• Ask finance for the cost line. With only 8% of leaders confident in their AI delivery costs, request a per-account view of AI consumption so discounting decisions reflect real margin, not the 80% the deal desk still assumes.

The companies that win the next phase of AI will not necessarily be the ones that ship the most features. They will be the ones whose revenue teams can explain what those features are worth and get paid for them. Right now, too many sales organizations are acting as the brake on that conversation. The opportunity for CROs is to become the people who lead it.