
Playbook
Tasteful Skills
“Tasteful Skills” argues that the best agent skills are not documentation or best-practice lists.
Influence isn’t politics — it’s how good ideas survive — Jessica Fain frames influence as increasing the odds that strong product ideas get funded and shipped, not manipulating people for personal gain.
Most PMs fail with execs because they forget empathy — Fain says product managers obsess over understanding users, then lose that curiosity when talking to leaders, even though execs are just as important to understand as any customer.
Exec calendars are a ‘strobe light,’ so context-setting is a power move — because leaders jump from finance to legal to hiring to product reviews all day, she recommends spending 30-60 seconds at the top of meetings clarifying why you’re there, where things left off, the goal, and whether there’s anything else they hoped to cover.
The best tactical question in the episode: ‘That’s so interesting — what led you to believe that?’ — Fain uses this to unpack confident executive opinions, uncover pressures like board demands or recent conversations, and turn disagreement into co-creation instead of silent resentment.
Senior operators build trust by killing things, not just asking for more — one of Fain’s strongest points is that deprioritizing weak ideas, defining failure criteria, and returning on a specific date with results signals executive-level judgment far more than endless feature advocacy.
AI makes influence more valuable, not less — as coding, prototyping, note-taking, and analysis get automated, Fain argues the differentiator shifts to deciding what should exist, getting buy-in for V2/V3 investment, and giving teams and agents enough strategic clarity to build the right thing.
Lenny opens by calling influence one of the highest-leverage skills outside AI, and Jessica Fain immediately agrees: if you can’t build momentum and buy-in, you can’t build great products. The big trap, she says, is thinking great work will “show itself” while other people quietly get promoted because they know how to get ideas backed.
Fain tells the story of being an IC PM at Slack, confused why some ideas got funded while others “died on the vine,” even when she deeply understood users. Eight-and-a-half months pregnant, she asked newly returned CPO April Underwood if she could be her chief of staff, then later served Tamar Yehoshua too — and that’s where she learned people wildly misunderstand how executives actually make decisions.
Her most memorable analogy: an executive’s day is like “a strobe light going off.” They’re context-switching between budget, hiring, legal, people problems, and product, while the PM walking in has been thinking about one idea for weeks — so the PM has to supply the context the exec simply hasn’t had time to hold.
Fain’s core reframe is that PMs should use the same curiosity and empathy with executives that they use with customers. Don’t walk in looking for a rubber stamp; walk in asking how to strengthen the idea with the leader’s domain knowledge — and if you truly think they have nothing useful to offer, she says bluntly, you should probably quit.
She pushes back hard on the idea that influencing up means agreeing with everything. You’re paid to be the domain expert, and the move when you hear something that sounds wrong is not to sulk but to ask, “That’s so interesting. What led you to believe that?” — a line she praises because it surfaces the hidden board pressure, recent meeting, or broader company context behind a strong executive opinion.
One of the most practical parts of the conversation is her advice to use chiefs of staff, EAs, and past successful presenters to learn what matters to a leader right now. At Webflow, she says a peer trained a GPT on past product reviews so PMs could test docs against likely pushback; she also describes Slack’s “Hey Stuart, what do you think?” sessions, where teams asked Stewart Butterfield for early raw thinking before showing polished designs he might hate.
Fain admits she once overloaded a review with too much evidence and watched the person she most needed glaze over and grab their phone. Her advice: lead with the recommendation, have the supporting detail in the appendix, and show multiple options so the exec sees you didn’t miss the obvious alternatives — she calls one old Slack rule “Stuart plus two more,” meaning do the version he asked for and bring two additional takes worth debating.
A huge part of influencing execs is tying your work to their success criteria — what the board is pushing on, what metric they own, what company bet they’re making. Fain gives the example of Slack’s “customer love sprint,” where the team paused normal work, let engineers choose fixes, shipped 65 improvements, and reconnected with leadership’s belief in product craft; later, she says the deepest trust-builder is being willing to kill projects, shrink risky bets into experiments, and come back on a clear date with evidence.
In the final stretch, Fain argues we’re entering a golden age of product management, even if not of product managers. As AI handles more execution, the job shifts toward deciding what survives, aligning teams, codifying strategy clearly enough for humans and agents to run, and using AI as a smart colleague to poke holes in ideas — while remembering that judgment, trust, context, and buy-in are still deeply human work.
Share
Keep Reading
The Weekly Echo. The inbox-shaped summary of what mattered.
New editorials announced here.

Playbook
“Tasteful Skills” argues that the best agent skills are not documentation or best-practice lists.

Playbook
Learn how tasteful prompting helps you move beyond generic AI output by shaping context, style, and judgment from the start.

Playbook
OpenAI shipped /goal for the Codex CLI. It turns a prompt into a persisted, self-continuing contract.