AI Is Eating Your Business Model. You Are Still Talking About Productivity.
Every technology wave destroyed one way of making money and built another. AI is not changing how you work. It is making your current revenue model obsolete. Zendesk already changed their invoice. Your competitor is studying this. You are in a productivity meeting.
Every wave kills one business model and builds another on the grave. The PC killed the mainframe rental. The internet killed the print classified. AI is not the next wave. It is the demolition crew. And most CEOs are still selling subscriptions to a world that has already decided to pay per result.
AI agent software spend is more than quadrupling between 2025 and 2027, according to Gartner. That is not market growth. That is a category being invented in real time. The businesses that price for seats will watch the businesses that price for results eat their margins whole.
Every wave makes three moves. First it gives you better tools. Then it hands the distribution to someone else. Then it rewrites the invoice. We just finished move one. Move two is happening now. The companies that own move three will own the decade. The rest will explain at board meetings why growth slowed.
Zendesk now charges per resolved ticket — not per agent seat. $1.50 every time AI closes a support issue without a human. Their 20,000 AI customers are not paying for access anymore. They are paying for results. Every support software vendor still charging per seat just lost the pricing argument. Ask yourself: who in your sector just quietly changed their invoice?
The Straight Take
You are using AI to do what you already did faster. That is efficiency. Efficiency is not a moat. It is a grace period.
In 1994 every business leader thought the internet was a faster fax machine. It was not. It did not speed up classified advertising. It deleted it. Travel agents did not get a website. They got replaced by one. Yellow Pages did not go digital. It ceased to matter. The internet did not improve the old model. It wrote a new one and left the old one on the floor.
Nobody announced it. There was no press release that said “the economic rules have changed.” One day the invoice just looked different. And the companies that had been building for the old invoice had no answer.
I have watched this pattern twice in my career. The tools arrive and everyone celebrates. Then the platforms arrive and everyone scrambles. Then the economics arrive and most people are too late. AI is running the same playbook. We just left Phase 1. The celebration is over.
You Already Know This. You Just Haven’t Named It Yet.
That smile. I know that smile. The newspaper editor had it in 1997 when Craigslist launched. Ugly website. Free listings. Clearly a hobby project. Until it collapsed classified ad revenue by 75% in twelve years. The newspaper did not lose because it got worse. It lost because the model underneath it became irrelevant. The content was fine. The invoice was wrong.
AI does not compete with your model. It makes your model a question nobody needs an answer to anymore.
Every Wave Makes Three Moves. You Are Inside Move Two.
I have seen this before. The tools come first and everyone feels smart for adopting them. Then the platform layer forms and the distribution shifts away from the incumbents while they are still congratulating themselves on move one. Then the pricing model breaks and the companies that did not see move three coming spend the next decade in decline. Same playbook. Every single time.
Your Competitor Is Not Improving Your Model. They Are Replacing It.
Every incumbent makes the same mistake. They see the new technology and ask: how do I use this to do what I already do better? That is the wrong question. It is the question that killed Oracle’s dominance when Salesforce launched. It is the question Kodak kept asking until they went bankrupt. Improve the existing model. Cut costs. Keep the structure.
Salesforce did not improve the software box. They removed it. Instagram did not improve the film roll. They made the film roll a question nobody asked anymore.
Zendesk changed their invoice in August 2024. They stopped charging per agent seat. They started charging $1.50 per resolved ticket — only when the AI actually closes the issue without a human. By end of 2025 they had nearly 20,000 AI customers and $200 million in AI annual recurring revenue. Their Voice AI agents claim to resolve 80% of tickets without human intervention. That is not a product story. That is a business model story. And every support software company still selling seat licences just walked into a pricing fight they are structurally unequipped to win.
McKinsey said it plainly in September 2025: AI is moving from enabling work to performing work. That one sentence should end every board discussion about AI adoption. You are not buying a better tool. You are buying a workforce that does not show up on your headcount. The moment your competitor understood that, your pricing model had a clock on it.
The new model charges for results, not access. It does not scale with your headcount. It scales with your ambition. And the cost of delivering one more unit of work is almost nothing. That last part is the one that changes everything.
And here is the part nobody wants to say out loud. Once your competitor has six months of outcome data in your market, you cannot buy it back. You cannot hire around it. You cannot outspend it. The first mover in outcome pricing does not get an advantage. They build a wall. You can see the wall. You just cannot get through it.
Everyone Is Measuring the Wrong Thing
Your board is asking about AI adoption. Wrong question.
Adoption tells you how many people are using the tool. It tells you nothing about whether your business model survives the decade. A newspaper in 2001 with 100% digital adoption was still a newspaper. The tool changed. The invoice did not. We know how that ended.
Putting AI on top of a broken model is not transformation. It is acceleration toward the same cliff. The companies that will define this decade are not the ones with the most AI usage. They are the ones that asked one different question: if my best competitor operates with no human labor cost next year, what does my pricing look like? That is a survival question. Not an adoption question. And most leadership teams have not asked it yet.
It Is Not Coming. It Is Here.
Legal services: AI contract review now charges per clause, not per seat. Hospitality: AI revenue tools charge per booking recovered, not per property. Recruiting: charge per qualified candidate, not per login. Insurance: charge per policy processed, not per licence.
Nobody made an announcement. Nobody held a press conference. They just sent a different invoice. And the incumbent in each of those markets had one conversation that went exactly like your scene box at the top of this article. “Aggressive pricing strategy. Won’t last.”
Count the sectors between those markets and yours. I have been doing this for thirty years. The distance is always shorter than it looks. And it always closes faster than the planning cycle.
Three Things Could Happen. Here Is What I Would Bet On.
Five Signs It Has Already Started in Your Market
You will not get a formal announcement. Watch for these instead.
The Straight Thought
The mainframe companies had ten years to adapt. Print media had five. Nokia had three. AI is moving faster than all of them. The cost of deploying AI dropped 280-fold in eighteen months. That is not a market cycle. That is physics. Speed is no longer a function of ambition. It is a function of economics. And the economics just changed permanently.
I have been in enterprise technology for nearly thirty years. I have watched companies survive technology waves and I have watched companies that did not. The ones that survived did not survive because they moved fastest. They survived because they understood what the wave was actually changing. Not the tool. The invoice.
Here is the board meeting that is coming for you. Two years from now. Your revenue retention is falling while the market is growing. Your CFO has four explanations. None of them are right. The right explanation is that a competitor changed their invoice eighteen months ago. You read about it. You called it aggressive. You kept your pricing model. You kept your headcount. You kept your contracts. And now you are sitting in that room trying to explain a number that has only one real cause.
I may be wrong about the timing. But I am not wrong about the direction. The invoice is changing. The only question is whether you change it or someone changes it for you.
That’s the straight thought.
Prasad Ballari · June 2026
Sources: Gartner Strategic Predictions 2026 — Oct 2025 press release; Gartner AI Agent Software Spend Forecast — May 2026 press release; Gartner Enterprise App AI Agent Integration Forecast — Aug 2025 press release; Gartner CxO AI Strategy Survey — Dec 2025 (n=197), cited in May 2026 press release; McKinsey “Upgrading Software Business Models to Thrive in the AI Era” — Sep 2025; McKinsey “AI Adjusts the Software Bill” — Sep 2025; Stanford HAI AI Index Report 2025 (covering 2024 data); Menlo Ventures “State of Generative AI in the Enterprise” — Mar 2026 (enterprise GenAI spend $11.5B 2024 to $37B 2025, 3.2x growth); Zendesk Relate 2025 conference announcements and Resolution Platform pricing (August 2024 launch, expanded 2025); Futurum Group market note on Zendesk outcome-based pricing (May 2025); CMSWire, Zendesk Relate 2026 coverage (May 2026); Flexprice SaaS usage-based pricing research (updated February 2026).
Read every issue at www.straightthoughts.com
For direct conversation: [email protected]
If this issue made you rethink one number in your business, forward it to someone who needs to see it.