Each Monday, a short list of the sharpest sales + GTM thinking from the past week. Not summaries — read the originals, they're better. Just what's worth your time.
Four pieces this week.
1. Steve Lucas — "AI is about to flip from influencing 20% of decisions to making 90% of them"
Steve Lucas (Chair and CEO, Boomi) sat down with GTMnow at the HumanX conference in San Francisco and made a prediction worth taking seriously: AI currently influences roughly 20% of executive decisions and will make 90% of them within two years. His frame for 2026 is "pilot-to-production year" — the inflection point when experimentation ends and AI moves from proof-of-concept into operational architecture. The implication he draws: leaders who are still running AI pilots at the end of this year will have organizational decisions made for them, not by them.
→ Full GTMnow conversation with Steve Lucas (podcast + show notes, May 2026)
2. Sam Senior — "By the time the buyer calls you, 70-80% of the decision is already made"
Sam Senior (Founder & CEO, TestBox) made an argument on GTM 186 that tracks a shift many have been watching build: enterprise buyers are arriving to first calls with most of the decision already locked in. The shortlist has collapsed from three or four vendors to one or two. The discovery call is now a validation call — the seller's job is to validate or correct a picture the buyer already holds, not build one together. Sam runs 15 AI experiments per week at TestBox and ties token usage to performance reviews, not because it's a stunt, but because the AI literacy gap across a team compounds into a competitive disadvantage faster than people expect.
→ GTM 186 on GTMnow (podcast + show notes, April 2026)
3. Auren Hoffman — "Every software moat is gone, and your customers won't sign a year-long contract to prove it"
Auren Hoffman (GP, Flex Capital; founder, LiveRamp) made three connected arguments on GTMnow that most GTM leaders are underweighting. First: the era of "good enough" SaaS is structurally over — if you're not improving your product materially every month, you will lose customers. Salesforce, LinkedIn, DocuSign — none of them are exempt. Second: annual SaaS contracts are becoming un-signable. Buyers won't commit to a year when the product category may be obsolete in three months. Third: by the end of 2026, the first VC meeting at Flex Capital will be fully agent-to-agent — their AI talking to the founder's AI before any human is in the room. He's already running 500+ AI agents for deal sourcing; 53 investments in 2025 at roughly one per week.
→ Full GTMnow conversation with Auren Hoffman (podcast + show notes, April 2026)
4. Jon Addison — "91% of enterprises are deploying AI agents. 10% have a security strategy. That gap is a product launch."
Jon Addison (CRO, Okta) walked GTMnow through the most concrete CRO operating playbook I've read this quarter. Okta went from nearly $850M in operating losses to over $760M in operating income — and the key mechanism was identifying a structural gap the market had surfaced: 91% of enterprises are already running AI agents, but only 10% have any security strategy for them. That "AI governance gap" became the foundation for Okta's biggest product launch in years, Okta for AI Agents. The GTM result: deals that included the new AI products carried 40% higher average contract value than deals that didn't, and 30% of all Q4 revenue came from those new products. The path from $3B to $5B ARR runs through the gap.
→ Full GTMnow conversation with Jon Addison (podcast + show notes, April 2026)
Common thread this week. Lucas says AI decisions are compounding whether you plan for them or not. Senior says buyers are compounding their own research before you ever talk to them. Hoffman says the moat is gone if your product isn't compounding month over month. Addison says the product that compounds into existing customer architecture commands 40% more. The argument isn't about AI adoption anymore — it's about what you're compounding into, and whether your architecture can write it all down.
Next Monday: a new four.
If you've read something this week worth flagging in next week's list — hello@mallin.io.