Thanks for reaching out. Be sure to check us out on LinkedIn for all of our current news and announcements.
Oops! Something went wrong while submitting the form.
Picture the scene. Your AI feature launches. Engagement is 10x anything else in the product. Your CEO is ecstatic. You're on every all-hands slide for a month.
Six months later, finance review. Gross margin has collapsed. Your power users cost more than they pay. Your best engagement metric is the one strangling your P&L.
This is happening quietly in products everywhere. The product metrics changed (we covered those in the last two editions). The bigger shock is what AI is doing to business models. Three shifts.
Shift 1: Cost-per-successful-task is your new CAC math.
Traditional SaaS: CAC, LTV, and gross margin were relatively stable per customer. Scale made things cheaper. The marginal cost of one more user was close to zero.
AI reality: your power users literally cost you money. Tokens are variable cost. Flat-rate subscription plus a heavy user equals negative margin on that account. The more someone uses your product, the worse your unit economics get. That's the exact inverse of what you want.
What to measure: gross margin per active user (not per paying user, per active user, big difference), cost-per-successful-task, model cost as a percentage of revenue, and the marginal cost of power users versus the marginal revenue from them.
Intercom's Fin nailed this. They didn't price per seat. They priced at $0.99 per successful resolution. You only pay when Fin actually solves the issue. That's outcome-based pricing, and it's mathematically honest about what AI products cost to run. If your pricing and metrics don't reflect variable compute cost, you're flying blind.
Shift 2: Pricing is a product decision now.
Usage-based and outcome-based pricing are still early. Hybrid models (a low monthly fee plus usage, with overage) are probably where most AI products land.
Here's what matters for corporate innovators and product managers building new things: pricing is a product decision now, not just a finance decision. The pricing model tells the user what success looks like, and it has to match the unit economics underneath.
Think about "unlimited AI queries for $20/month" versus "$0.99 per successful outcome." Those aren't two pricing models. They're two completely different products from the user's perspective. The first says "experiment freely, we'll eat the cost of your learning." The second says "we only win when you win." Both can work. Neither is neutral.
Pricing is now a core product design decision.
We’ve felt this directly building AI tools at Highline Beta. While building tools like Candor, we spend real time instrumenting the software to track AI usage and calculate cost. It's striking to see which features drive cost, then ask whether they drive commensurate value, then figure out how to price it all. Adding a SaaS feature used to be cheap to run. That's simply not true for AI features.
Shift 3: Experimentation isn't a vanity metric anymore.
Experiment count used to feel like a vanity number. In an AI era it's mission critical, and it earned that promotion through a specific mechanic.
AI lets you ship way more, way faster. That sounds like a pure win. It isn't.
If you're shipping faster but not running real experiments, you're vibe-stuffing your product. You're adding features because you can, not because you have evidence they'll create value. The product bloats, the codebase bloats, the cognitive load on users climbs.
Every AI feature carries an ongoing per-call cost, not a one-time build cost. So the bloat isn't just clutter. It's a tax that compounds with usage. You're paying tokens every time someone touches a feature you had no evidence for. Slow, expensive, and unproven, all at once. Pre-AI bloat was annoying. AI-era bloat is a margin killer.
Strong experimentation is the only defense, which is exactly why Lean Analytics goes up in value here, not down. A useful filter: for every experiment, write down the hypothesis and the decision criteria before you ship. If you can't, you're not running an experiment. You're running a release. Both have a place. Don't confuse them.
What to measure: experiments per quarter as a real metric, hypotheses written before launch, features sunset based on data (not just features added), and cost-per-feature in production.
Underneath all three shifts is one principle, and Ben Murray (the SaaS CFO) named it cleanly: "If SaaS is about margin efficiency, AI is about value density. You're optimizing for how much output, productivity, or labor you replace per dollar of compute."
That's measurable, and it's three ratios that move independently: cost-to-deliver per task, revenue captured per dollar of compute, and value delivered to the user per dollar of compute. A team can be strong on the first two and still build something nobody will pay more for. The diagnostic only works if you measure all three.
So, honest question: do you know your gross margin per active user right now? Or is your best engagement metric quietly the thing that's going to bite you?
