The 3 episodes below are this week's deep dives; the other 2 are written up in full in the Notion archive.
The Pipeline Efficiency Pyramid, from Union Square Consulting via GTM Science - a four-layer stack for fixing pipeline management in the right order, built on the premise that closing more of what you already generate beats generating more.
First step this week: audit how many open deals exceed 1.5-2x your average sales cycle. Measure success by whether your forecast starts matching reality within a quarter or two. Failure mode: skipping to amplification - AI on a broken foundation just produces confident garbage faster.
The strongest signal this week cuts against something most 2026 capacity plans have already baked in: that AI made sellers dramatically more productive, so quotas should rise to match. Tunguz on Topline puts the counter plainly - the supply side has not changed, enterprise budgets grew by a factor of ten, and that demand shift is what is driving the numbers people are crediting to their reps. The distinction matters because the two look identical on a bookings chart and imply opposite plans.
Holm on The Crew supplies the number that settles it. In the SaaS era he modeled "70 at 70", seventy percent of reps at seventy percent of quota, as the sign an org had cracked it. Today he puts the norm at 30-40% of reps making plan, with 50% best in class. Attainment has halved in the era supposedly making sellers multiples more effective. And his own recovery at LaunchDarkly, 18% to 54% over plan, came from value selling, pipeline-generation days, weekly S0/S1 instrumentation and spiffs: motion work, not tooling.
Reynolds on GTM Science and Larson on RevOps AF show the same reflex from the operator's seat: teams reaching for AI enrichment or a new CRM while stages sit undefined and lead routing leaks. The tooling instinct is the tell that fundamentals were skipped.
What to do with it: pin next year's capacity model to real budget signals rather than an assumed productivity multiple, and use participation rate as the gate on adding headcount. Get it wrong and you over-hire into split territories and a demoralized team, precisely the feast-or-famine Holm inherited.
The three episodes worth your time this week, with the mechanics behind the takeaways above.
Reynolds and Bueckert argue most revenue leaders reach for more pipeline when the real problem is a leaking one. Reynolds' opening questions to a new team: what's the average sales cycle, and how many open deals exceed 1.5-2x that? A deal past 2x the cycle has near-zero win odds; if that's half your pipeline, the forecast is untrustworthy. He also reads close-rate spread as a data-integrity tell - an 85% rep is sandbagging, a 5-15% rep is dumping every meeting into pipeline to look busy. Both make forecasting fiction.
The fix runs through a pyramid in order. Fundamentals means a real ICP reverse-engineered from wins and losses, including "fuzzy" criteria no tool can surface - a company that spent $500 on its website ten years ago won't buy a $24k marketing tool. Stages need explicit entry and exit criteria: don't give the custom demo until you've secured decision-maker access, because after the demo your champion has everything and you've lost all leverage.
Adoption gets enforced by keeping required fields minimal. He cites a client where finance mandated 35 fields per update; the entire sales team, plus the CFO and controller, got fired. The concrete optimization win: having SDRs book the meeting live on the call instead of emailing times afterward roughly doubled conversion overnight.
Holm's most portable material is on quota attainment as a scaling signal. At LaunchDarkly he inherited roughly 70-80 sellers with 18% over quota - "feast or famine," which he read as a sign the company hadn't cracked hiring profile, onboarding, or the outbound motion. The company had won on inbound PLG but was failing top-down outbound. His fixes: introduce value selling, run pipeline-generation days, instrument KPIs for S0/S1 opportunities weekly, and gamify with spiffs and quarterly competitions. Attainment climbed from 18% to 54%, and outbound-sourced deals went from under 20% to 70-80% of closed opportunities - which lifted deal size and cut ramp time by 30+ days because outbound targeting was intentional against ICP.
His benchmark: 30-40% of reps making plan is the current norm, 50%+ is best-in-class. The old "70 at 70" (70% of reps at 70% attainment) no longer holds because AI-native companies often skip quotas entirely - some pay base plus MBOs and don't track hours.
On hiring he uses ICE (intelligence, character, coachability, experience) and will hire outside dev-tools if the raw materials are there, since AI is new enough that everyone's learning on the job.
Kosoglow's frame: an AE sells promise to create pipeline; an AM sells proof, and value delivery is what creates pipeline. No delivered value, no expansion opportunity - full stop. From $100M to $250M ARR at Outreach, most growth came from the existing base, which is why he treats account management as its own discipline.
The account-management loop has four steps: understand the executive outcome (the target and metric), deliver the value proof by running the project with the CSM, communicate it back so the customer acknowledges it ("a tree that falls in the forest"), then earn the next outcome and repeat. Timing is driven by the customer's project rhythm, not calendar QBRs - he anchors touchpoints to their product launches and kickoffs to land in an actual buying window instead of hoping a quarterly QBR catches one.
Renewal is the best expansion moment - a commercial milestone where accumulated proof points earn the right to ask for more. His discovery-call structure: start with the trust factor ("this is what you asked us to do, here's the result"), run discovery, then bring back a plan built with the CSM. Only float a new-product hypothesis if delivery went well; if it's rocky, anchor on what they already paid for to rebuild trust first.
The whitespace map has three contract-value levers (new product, expanded scope, commercial mechanics) activated by four triggers (new use case, pricing/packaging modernization, consolidation, M&A/org change).
Full write-ups are in the Notion archive.
30 Minutes to Presidents Club, #605 - How to Build a Cleaner Sales Pipeline in the Age of AI (Stevie Case). It pairs directly with this week's GTM Science pyramid but comes from an operator's voice on the tactical edge of pipeline hygiene under AI, and the delivery carries specifics the digest can't compress.
Get the next one on Monday. Free, one email a week.