Budgets are tighter, buying groups are pickier, and AI ships faster than ops can absorb it. AI is flooding GTM with “smart” everything—scoring, routing, outreach, pricing. But not every smart makes money. These are the six traps I keep running into, plus the moves I ask teams to make this quarter.

When shipping software was harder, teams had to live with the problem before writing code. Now GenAI can spin up a feature by lunch—and the patience for real discovery evaporates. I see teams building what they can build instead of what customers hire them to do.

Leader’s Move: Institute a 1 business week pre-code pause: 5 customer conversations, 1 job story (“When ___, I need ___, so I can ___”), and 1 success metric (time saved, risk reduced, dollars realized) before any prompt engineering. No job, no ship.

It’s never been easier to mint new GTM metrics—LLMs happily synthesize clever ratios and dashboards. The result: metric inflation, decision paralysis, and instinct wrapped in charts.

Leader’s Move: Decision minimalism + metric contracts. Name your 5 recurring GTM decisions and bind each to one directional metric. Every other metric must have a written contract: “A +Δ in this metric predicts ≥X% uplift in EBITDA (or EGP) within Y days.” Example: “+10% in Qualified Demo Rate predicts ≥3% EBITDA uplift in 2 quarters via +2 pts win rate. No contract, no dashboard.

AI is accelerating product development—and shrinking the shelf life of “winning playbooks.” What worked last quarter decays in weeks as competitors and customers adapt.

Leader’s Move: Pair performance models with change detectors and review monthly (not quarterly). Examples: segment drift (ICP firmographics mix), win/loss language shifts (top n-grams), channel CPM inflections (>20% MoM). Ownership: RevOps + Product Marketing. Sunset plays fast; fund small exploration cells to seed the next playbook.

If a rep misses, we coach; if an AI misses, we cancel the program. That double standard freezes adoption and strands ROI. I see it all the time.

Leader’s Move: Set defect parity: AI must beat human baselines by ≥10–20% on agreed metrics (precision/recall, handle time, margin) before scale. Ship with a one-click AI-off fallback and publish an error budget (e.g., 1.5% monthly) alongside uptime.

Tools like Microsoft Copilot Studio lowered the barrier for “good enough” apps. Innovation is up—but so is sprawl. I often find 7–10 prospecting tools in one org; nobody knows which path to trust.

Leader’s Move: Define a golden path (the blessed stack for prospecting, pricing, proposals) and a kill list. Centralize work in one owned surface (your CRM workspace, e.g., Salesforce console), enforce an app registry, and ruthlessly deprecate duplicative flows. Policy: duplicate tools must show ≥15% outcome lift or get deprecated within 30 days. Registry owned by Ops/IT; each app needs an owner, outcome metric, and review date.

GenAI makes it trivial to create adaptive metrics—and then optimize to them. I’ve seen reply rates hit records while EBITDA sagged. What you reward gets gamed.

Leader’s Move: Make proxies second-class citizens. Optimize to primaries: EBITDA, EGP, or LTV:CAC, and gate launches on revenue-proximate metrics: stage conversion, stage velocity, forecast accuracy, and gross margin. Keep a pyramid dashboard:

• Primary: EBITDA / EGP / LTV:CAC
• Revenue-proximate: stage conversion, stage velocity, forecast accuracy, gross margin
• Proxies (read-only): opens, replies, meetings set Rollback rule: If proxies rise for 2 sprints while primaries don’t, roll back.

• Which KPI could someone maximize while hurting EBITDA? If that list is long, you’re proxy-led.
• Do sellers know the golden path for their top three workflows?
• What are your change detectors, and when did they last force you to retire a “winning” play?
• Which AI-assisted workflow has an explicit error budget and defect parity target?

Bottom line: GenAI didn’t remove the need for judgment—it raised the bar for it. Start with decision minimalism, anchor on EBITDA/EGP, set defect parity, and make the golden path unmistakable. Teams that do this will compound advantage while everyone else is still tuning dashboards.

P.S. I’m early in this journey myself. I’m working to make these Leader’s Moves stick—but first I had to name the traps. Noticing them has been the first step toward better decisions.