ExpertsFMAn 86-minute guide to finding PMF | Rob Snyder (Author, The Power of Pull)
Lenny's Podcast: Product | Career | Growth · October 25, 2026
5 Selected Lessons:
heuristic · 3 framework · 2
Structured sales-call discovery-to-pitch transition
Source
Lenny's Podcast: Product | Career | Growth — An 86-minute guide to finding PMF | Rob Snyder (Author, The Power of Pull) (October 25, 2026) · at 40:37 ▸ · at 43:34 ▸
When this applies
when running any discovery conversation where you need to earn permission to probe before pitching anything
The takeaway
Let the counterpart self-reveal their need before you ask structured questions, and get explicit buy-in on the conversation's agenda before probing deeper, so questioning doesn't feel like interrogation.
Why it works
Opening with 'why did you take this call' invites the prospect to volunteer their own pull framework elements unprompted, which is more credible than extracting it via interrogation; agreeing on an agenda first makes subsequent probing feel consensual rather than invasive.
Where it stops holding
This stops working when the prospect has no self-awareness of why they took the call, or when time pressure doesn't allow for this two-step framing.
Detail
procedure:
- Open with 'why did you take the call with me today?' to let them reveal pull unprompted
- Reflect what they said back as the call's agenda and get their explicit agreement
- Ask permission to ask a few context questions before probing ('do you mind if I ask a few questions...')
- Keep calls to 30 minutes (smaller blocks often get skipped)
- Skip or keep demo very short unless this is enterprise sales, where demo timing differs
benchmarks: {"call_length":"30 minutes","assessment_checkpoint":"after 5 calls, review whether pull was found at all"}
rationale: This structure surfaces genuine pull data rather than persuasion theater, and if you learn nothing after five calls with everyone 'leaned back,' something is wrong with the hypothesis.
In their words
Hey, thanks so much for taking the time. I'm curious, I'm sure you get a bunch of people reaching out to you. Like, why did you take the call with me today?
Five-call checkpoint and goblin mode escalation
Source
Lenny's Podcast: Product | Career | Growth — An 86-minute guide to finding PMF | Rob Snyder (Author, The Power of Pull) (October 25, 2026) · at 43:34 ▸ · at 1:02:25 ▸ · at 1:06:03 ▸ · at 1:19:27 ▸
When this applies
when testing any hypothesis iteratively and needing to decide how much evidence to gather before updating or abandoning it
The takeaway
Set small, fixed checkpoints to evaluate hypotheses quickly, and reserve high-intensity, unconventional effort as an escape valve when normal iteration isn't generating signal.
Why it works
A small batch (five) is enough to detect a signal or its absence without wasting excessive time on a dead hypothesis, while also not jumping to conclusions from a single data point; when conventional iteration stalls, deliberately extreme actions (goblin mode) generate new information fast.
One level up
Time in an uncertain, stuck state is the enemy — bias toward fast cheap tests and occasional extreme bursts of effort to break stagnation.
Where it stops holding
This stops working when sample sizes of five are too small for the specific market (e.g., very large enterprise sales cycles) to reveal any signal at all.
Detail
procedure:
- Run roughly 5 sales calls against a pull hypothesis
- Assess: did anyone show real pull-shaped action?
- If no, ask the Hail Mary question across those calls, generate a new hypothesis, repeat
- If stuck entirely, escalate to 'goblin mode': extreme, unconventional actions (flying to a conference, going door-to-door) purely to get in front of customers and generate new leads
benchmarks: {"checkpoint_size":"~5 calls per hypothesis iteration"}
rationale: Doing hundreds of calls without checking in wastes time; five is enough to tell you something is probably wrong and prompts a pivot.
In their words
I don't have a hard, like what I tend to do is if you have five calls and everybody's leaned back, and you learn nothing, then there's a sign that something's wrong.
Pull belongs to a person, not a company
Source
Lenny's Podcast: Product | Career | Growth — An 86-minute guide to finding PMF | Rob Snyder (Author, The Power of Pull) (October 25, 2026) · at 0:00 ▸ · at 16:10 ▸
When this applies
when pitching any large organization, institution, or committee rather than an individual decision-maker
The takeaway
Locate and sell to the specific individual who personally owns the unmet need, not the abstract entity they belong to.
Why it works
Organizations don't have to-do lists or urgency — individual humans inside them do; framing demand at the org level obscures who actually experiences the stuck feeling that drives buying.
One level up
Abstraction hides the actual locus of motivation and decision; always decompose a system down to the individual whose behavior you need to change.
Where it stops holding
This stops working when the purchase truly is a committee-level, no-single-owner decision with no individual champion feeling the pain.
Detail
rule: Always find the specific person inside the organization who has the project and is personally stuck — not the company as an abstraction.
threshold: If you catch yourself saying 'we sell to mid-market PE firms' instead of naming the specific role/person and their specific task, you've made this mistake.
rationale: You're not selling to SpaceX, you're selling to the head of legal at SpaceX who has a project she can't do herself.
In their words
You're not selling to SpaceX, selling to the head of legal at SpaceX who has a project on her to-do list that they can't do themselves.
Founder-magic cold outreach to land first sales calls
Source
Lenny's Podcast: Product | Career | Growth — An 86-minute guide to finding PMF | Rob Snyder (Author, The Power of Pull) (October 25, 2026) · at 34:39 ▸ · at 38:12 ▸
When this applies
when trying to get attention from strangers in any cold outreach context without yet having proof or a pitch worth making
The takeaway
At the attention-getting stage, optimize the message for curiosity about the sender, not for persuading the recipient of value — because pull, if it exists, is independent of your message and will surface once you're talking.
Why it works
A generic value-pitch message triggers pattern-recognition as spam and gets ignored; a short, personally interesting message (about the sender, not a sales pitch) triggers curiosity because it doesn't ask for anything and signals a real human worth talking to.
Where it stops holding
This stops working when your target audience doesn't respond to personality-driven hooks (e.g., highly formal industries) or when the hook is irrelevant/incomprehensible to them (e.g., name-dropping Silicon Valley figures to HVAC contractors).
Detail
procedure:
- Keep message to one or two sentences
- Include something genuinely interesting about the founder/story, tailored to what that audience would find interesting (not necessarily personalized to each recipient)
- Never pitch value or pain points in the message itself
- Goal is only to get a conversation, not to sell
benchmarks: {}
rationale: No template pitch gets responses except 'leave me alone'; the Hail Mary example: 'Snoop Dogg invested in us, and I thought you'd find it interesting too.'
In their words
The way I typically recommend founders start is to just reach out to your potential customers on LinkedIn or via email. And it's with a very specific kind of message that is not a, hey, do you have XYZ pain points?
Service-shaped delivery as a validation-and-build bridge
Source
Lenny's Podcast: Product | Career | Growth — An 86-minute guide to finding PMF | Rob Snyder (Author, The Power of Pull) (October 25, 2026) · at 59:16 ▸
When this applies
when any team has proven demand for an outcome but hasn't yet built the automated system to deliver it at scale
The takeaway
When product maturity lags validated demand, sell the outcome as a service first, deliver manually, and progressively automate — trading margin now for learning and iteration speed.
Why it works
Framing the offer as a service (delivered manually behind the scenes) rather than a finished platform resets buyer expectations appropriately while letting the team learn the real requirements through hands-on delivery, gradually automating as patterns emerge.
Where it stops holding
This stops working when the team gets stuck in the service model permanently, unable to transition to productized automation, ending up as a low-margin consulting business.
Detail
rule: Sell as 'an AI [role] who can do X, Y, Z' delivered manually for now while automation is built behind the scenes, rather than promising an enterprise-grade platform immediately.
threshold: Acceptable tradeoff despite risk because 'everything has downsides'; the goal is validating real pull and learning what to build while delivering value.
rationale: You validate real pull exists and learn what to actually build while delivering a mix of manual and automated work.
In their words
You can sell something that's like, you know, you might not have a full platform ready. You can sell something that's more service shaped early on.