HubSpot Just Priced the Unpriceable
The one thing you need to know in AI today | AI-Ready CMO
HubSpot rolled out new pricing for its Breeze AI agents today: $0.50 per resolved customer support conversation, $1 per qualified sales lead. The old model charged per conversation handled or per contact enrolled, regardless of whether anything actually got resolved or qualified. Now you pay only when the work gets done.
On the surface, this reads as a routine pricing update. We think it’s bigger than that.
For decades, marketing and sales operations have struggled to put a clean price tag on knowledge work. We all know roughly what an MQL “costs” — on paper. In reality, that number rarely captures the SDR’s salary, their manager’s time, the CRM licenses, the training, the turnover cost when they leave in 8.5 months. Customer support is worse. Productivity in a support queue depends on ticket complexity, team morale, tooling, and a dozen other variables nobody wants to (or can) model in a spreadsheet. Executives have been making staffing calls based on fuzzy averages for as long as these functions have existed.
Now, a qualified lead has a price: one dollar. A resolved ticket has a price: fifty cents.
Whether those numbers are accurate for your business is a separate conversation, and the definitions of “resolved” and “qualified” deserve real scrutiny before anyone signs a contract. But the psychological shift at the executive level is enormous. Cheapness is part of it — fifty cents a ticket is hard to argue with. The bigger thing is legibility.
AI work shows up on the P&L with a clean unit cost, no footnote required, and you can forecast against it like you forecast cloud spend. Predictability beats cheapness, nine times out of ten.
The interesting question is where this goes next. Image generation already has a per-output price. Video generation does too. Today, those numbers sit in the “cost of experimentation” column on most budgets, often tucked quietly into an agency invoice somewhere. We’re probably one or two product cycles away from somebody — Canva, Adobe, a challenger we haven’t heard of yet — charging per approved creative asset, per campaign variant that survived brand review. When that happens, the last great unpriced domain in marketing finally gets a price tag.
The conversations we’re having today about AI in support and sales will feel like the warm-up act.
— Torsten and Peter
Share with a colleague. Earn paid membership months.
Everything we know about AI marketing. Free.
Courses, workshops, frameworks, tool reviews, and implementation guides — all free, all on-demand. Built for senior marketers who don’t have time for theory.
3 AI Marketing Tools To Try Today
Brandwell
Brandwell is an AI long-form content engine that generates research-backed, SEO-optimized blog posts designed to bypass AI detection. Use it to scale organic content production without sacrificing quality or search rankability.
SitesGPT
SitesGPT uses AI to generate fully responsive websites in minutes by analyzing your business information and applying industry best practices. Use it to spin up professional landing pages and microsites without design or dev resources.
Birch
Birch is an AI-powered ad automation platform that manages, optimizes, and scales paid campaigns across Meta, Google, TikTok, and Snap with real-time KPI monitoring. Use it to stop babysitting ad accounts and let automation handle bid adjustments, pausing, and scaling.
Paid members, your toolkit and this month’s report are below.