Growth · Featured Debate

9 guests 12 episodes 4,270 words

The Growth Motion Trap: Why Choosing Between PLG and Sales-Led Is the Wrong Question

Should you grow through product-led growth or a sales-led motion?

A company doing $5 million in ARR, growing over 100 percent. The founders fire the sales team to go product-led. Eighteen months later, revenue has collapsed to under $1 million.

That story, told by Jason Lemkin on Lenny's Podcast, sits alongside an equally common failure mode: the enterprise startup that bolts on a free tier, calls it "PLG," and quietly sunsets the experiment six months later when engagement tanks and revenue stalls.

Nine go-to-market operators -- people who built the growth motions at Slack, Figma, Atlassian, Snyk, HubSpot, Stripe, and Confluent -- converge on a surprising pattern. The question is not which motion to pick. It is what order to layer them, and the answer depends on a variable most founders misjudge: how their first customers actually buy.

The 4 Positions

1 / 4

Start with Product-Led Retention

Elena Verna (Lovable · Miro · Amplitude · Dropbox) · Ben Williams · Claire Butler

Advocates: Elena Verna (Lovable, Miro, Amplitude, Dropbox), Ben Williams (Snyk), Claire Butler (Figma)

Elena Verna has led or advised growth at six major B2B companies including Miro, Amplitude, SurveyMonkey, and Dropbox, making her one of the most experienced practitioners of product-led growth in the industry. Her PLG sequencing framework has become the dominant mental model in the space -- and it is far more nuanced than most people realize.

Her argument: retention must come before acquisition, and product-led retention is the foundation that makes everything else work. Retention breaks into two KPIs: activation and engagement. If your product cannot activate and engage users through habitual loops, you have no chance of hooking an acquisition engine into it. Product-led acquisition means users invite, refer, or create content that attracts other users -- but all of that requires habitual usage first.

"I think every single company has to first focus on being product led and retention, period. The only way that you will ever have any chance of acquisition being product led is if you nail your product led retention."
Elena Verna
▶ 00:34:00

The sequence she prescribes: activation, engagement, retention, then product-led acquisition. Only after that foundation is solid should you layer on additional growth motions. But Elena draws a critical distinction most people miss. The hiring order depends on how money is collected, not on philosophy.

"If you're starting with self-serve monetization, growth hire should be frankly first before sales. Your sales are going to be more opportunistic in nature versus if you sales led company, go ahead, hire sales, wait for growth until you are ready to overlay product-led growth on top of your sales motion."
Elena Verna
▶ 00:13:23

She is emphatic that failing to layer every growth motion is a huge mistake. Growth models have a shelf life -- most spin out their ability to produce meaningful results within five to seven years. She recommends introducing new growth experiments every 18 months, and every five years something substantial must take a meaningful share of volume. Even Dropbox's sharing loop, which she calls an anomaly that has fired for 17 years, had its own dedicated growth pod.

"Every single company starts with their growth efforts focused either on product-led growth... or on sales and marketing... And that's good and fine for a while depending on how long that while is. But not overlaying every single way that you can grow through product, through marketing and through sales as an evolution is a huge mistake."
Elena Verna
▶ 00:50:58

Ben Williams saw this play out at Snyk. As the VP of Product who built Snyk's growth organization from the ground up, he scaled it from a niche Node.js security tool to an $8.6 billion company with millions of developers. Snyk's strategy was a fundamental disruption: security had always been sold top-down to CISOs, but Snyk bet on a developer-first approach. They started with a laser-narrow focus -- Node.js developers worried about open source vulnerabilities -- and were free from day one. The founders waited until roughly 5,000 free users before attempting any monetization.

"The macro structure, it's changed over time to enable us to focus on the biggest constraints in our growth model. At the beginning we just focused on acquisition and activation, intentionally deferring any investment into specific monetization initiatives around the self-serve."
Ben Williams
▶ Podcast

The activation metric was team-based: "F30D" -- fix in 30 days. Teams that fixed a vulnerability using Snyk within 30 days had strong retention and higher propensity to purchase. This team-based metric naturally created expansion opportunities, feeding the sales pipeline from the bottom up. Snyk also built "sidecar products" -- ungated resources like the Snyk Advisor and vulnerability database -- that attracted developers through SEO, then funneled them into the free product.

Claire Butler's experience at Figma is the purest expression of bottom-up growth. As Figma's first GTM hire and 10th employee, she built a growth engine that relied entirely on practitioners falling in love with the product. Figma had no sales team for the first three years. All revenue was paid but entirely self-serve.

"We actually didn't have a sales team for the first three years. So all of our revenue, it was paid, but it was all self-serve. And so we'd work with these... The individual contributor or maybe the manager would just put them on their credit card."
Claire Butler
▶ 00:15:27

The Microsoft story reveals the mechanism. Claire noticed people within Microsoft signing up for Figma. Over time, adoption spread through organic node graphs -- small pockets of users jumping to adjacent teams. Eventually Microsoft had a massive network of Figma users, all on credit cards, with no enterprise product to buy. Microsoft came to Figma saying: we need security, procurement, account management. The enterprise customer pulled Figma's sales organization into existence.

"They actually come to us and they're like, 'Hey, I work at this company. I really want to get my whole company to use it, but security's not letting me. Can you help me unblock it with them?'"
Claire Butler
▶ 00:16:04

When This Applies

Your first customers needed demos, onboarding, or security reviews to buy (Lemkin's DNA test)
The buyer is an executive, not the end user
Deal sizes are $50K+ from early on (Jeanne's Sierra example: eight-figure deals from day one)
*Then begin building PLG when you want non-linear growth decoupled from headcount (Christopher Miller's HubSpot insight) and have data infrastructure to learn from free users (Hila Qu's DLG prerequisite)*

Be Honest About Your DNA

Jason Lemkin (SaaStr · EchoSign/Adobe)

Advocate: Jason Lemkin (SaaStr, EchoSign/Adobe)

Jason Lemkin has advised or invested in thousands of SaaS companies through SaaStr and personally built EchoSign from zero to acquisition by Adobe. His counter-argument begins with a deceptively simple principle: look at how your first 10 to 20 unaffiliated customers actually find and buy your product. Not how you wish they would. How they do.

"The most important thing though I've found is however you get those first 10, 15, 20 customers, be honest, be honest, and if you've talked to them as a founder and you know that they need a sales type motion, they need effort to deploy it, they have questions about security, they have questions about competition, they have onboarding requirements, and you say, 'Hey, I don't like sales, so I'm going to do a PLG motion,' you'll fail."
Jason Lemkin
▶ 00:07:06

Producty founders often have an emotional aversion to sales. They see PLG as an escape from human selling. Lemkin has watched this instinct destroy companies. He describes a portfolio company doing $5 million in ARR, growing over 100 percent, that fired its entire sales team because the founders found sales "icky."

"I'll tell you a company I invested in 18 months ago... they were doing 5 million in revenue, growing over 100% with the sales like motion. Things got harder... and they decided to fire the whole sales team, and now they're doing less than one [million] from 5 million... because the founders didn't like sales, they thought it was icky."
Jason Lemkin
▶ 00:08:00

But Lemkin is not anti-PLG. His actual position is that most companies are hybrids. He points to Canva (deferred sales until north of $500 million in revenue), Slack (started self-serve, majority enterprise sales by IPO), Notion (first salesperson around $10 million ARR), and Asana (a third of revenue still from self-serve). The spectrum is enormous, but the destination is the same.

"Most companies are a hybrid. Even there's weird hybrids like there's folks like Mongo and Snowflake... There's all different ways to combine a product-led or self-serve motion with sales-led. Too many founders in the beginning think it's either or. It's not either or."
Jason Lemkin
▶ 00:10:40

His most underrated insight: even founders who hate sales are "A+ middlers." Customers love talking to the CEO. Founders naturally excel at explaining their product once someone is in the conversation -- the problem is they do not know how to do outbound or closing, and so they avoid the entire function. But a worse version of this avoidance is the founder who has 5 customers paying $19/month through self-serve and 5 willing to pay $5,000/month through sales -- and chooses to focus on the $19 customers.

"Even if you hate sales, even if you think it's icky, even if you don't like it, as a founder you've, 95 times out of 100, you've got to find a way to do it."
Jason Lemkin
▶ 00:11:31

When This Applies

You are in a market where some segments self-serve and others need sales (Shaun Clowes's Confluent model)
Your product has a natural wedge (a simple use case that self-serves) and an expansion path (complex use cases that require sales)

PLG Has a Ceiling -- Plan for It

Jeanne DeWitt Grosser (Vercel · Stripe) · Merci Grace

Advocates: Jeanne DeWitt Grosser (Vercel, Stripe), Merci Grace (Slack)

Jeanne DeWitt Grosser built the early sales organization at Stripe from the ground up and now runs all go-to-market functions as COO at Vercel. She deeply respects PLG -- it is a major driver of Vercel's growth and was big at Stripe -- and she knows exactly where it breaks down.

PLG makes sense for most companies at the outset, unless you are explicitly building for enterprise from day one. Sierra, for example, is going after Global 2000 with eight-figure deals -- PLG would not be useful there. But for everyone else, the real danger is not the motion itself. It is the ceiling.

"The thing that folks get wrong is it does typically have a ceiling. So people are generally not going to give you $1 million via self-serve flow. So at some point if you want to sustain growth rates, you're going to have to have your deal sizes get bigger and bigger."
Jeanne DeWitt Grosser
▶ 01:14:28

The critical mistake is not adding sales -- it is adding sales too late. Building a replicable sales process takes quarters or years. You start with inbound, then need to add outbound, and turning outbound into a predictable engine is not a switch you flip. If you wait until PLG plateaus to start, you create a growth valley of death while the sales machine is being assembled.

"Where I think folks get stuck is waiting too long on PLG because it does take a while to build a replicable sales process and a sales process, which often you're getting fed by inbound at the beginning and then you got to add outbound. It takes a while actually to turn outbound into a predictable engine."
Jeanne DeWitt Grosser
▶ 01:14:28

At Stripe, Jeanne experienced the challenge of building sales within a PLG culture. With only 4 SDRs where other companies had 30 (Stripe's "efficiency is leverage" principle), she built Project Rosland -- a massive company universe database enabling hyper-personalized outbound. Her litmus test for PLG-native sales: put an AE in front of 10 engineers, and it should take them 10 minutes to figure out the AE is not a product manager.

"The litmus test I have always given my sales team is if you are an account executive in my org and I put you in front of 10 engineers at our company, it should take them 10 minutes to figure out you aren't a product manager."
Jeanne DeWitt Grosser
▶ 00:01:23

She delivers the most provocative observation in the entire debate: she cannot think of a single $100 billion company that is PLG-only. Even Atlassian eventually added sales.

Merci Grace's experience at Slack reinforces this from the inside. As one of Slack's earliest growth hires, she was brought in because of her game design background -- Stewart Butterfield wanted someone who understood player psychology and motivation, not someone with a growth playbook. Her role was not even called "growth" at first; it was "new user experience."

"Early on, I mean, I definitely got the job that I got at Slack because I had been a game designer. And Stewart Butterfield, the CEO and founder... knew, and this was part of our discussion, that I would bring a familiar sensibility to the role I was hired to, which wasn't called growth at first, it was new user experience."
Merci Grace
▶ 00:15:37

Slack's approach was rooted in curiosity. April Underwood, Slack's former CPO, had a line: "No one has built Slack before." This prevented the team from applying cookie-cutter tactics. They observed competitors copying Slack's animated onboarding circles -- design elements the team knew were not working. Users would still have them visible six months in.

"We really started with curiosity first, and we weren't like, 'Okay, here are all of our growth levers...'"
Merci Grace
▶ 00:17:13

But the most important lesson: by the time Slack went public, the majority of revenue came from enterprise sales, not self-serve. Even the poster child of B2B PLG became a sales-driven revenue business. PLG was the beginning of the story, not the ending.

When This Applies

Your first 10-20 customers found and adopted your product without human help (Lemkin's DNA test)
End users, not executives, are the primary buyers (Figma's designers, Snyk's developers)
Your product delivers value in minutes or hours, not weeks (Hila Qu's time-to-value criterion)
Revenue is collected via self-serve: credit cards, usage-based billing (Elena Verna's hiring sequence)
*Then begin building sales when users start requesting help with procurement, security, or org-wide deployment -- as Microsoft did with Figma*

Make the Motions Feed Each Other

Shaun Clowes (Confluent · Atlassian) · Christopher Miller · Hila Qu (Reforge · GitLab)

Advocates: Shaun Clowes (Confluent, Atlassian), Christopher Miller (HubSpot), Hila Qu (Reforge, GitLab)

Shaun Clowes occupies a unique position: he built the first ever B2B growth team at Atlassian and now runs both PLG and enterprise sales as CPO at Confluent. At Atlassian, what they were building was not even called PLG yet.

"Basically it's now called PLG, but yeah, at that time we didn't even know what to call it exactly."
Shaun Clowes
▶ 00:47:38

Today at Confluent, his insight is about the flywheel, not the funnel. Self-serve users who hit the free tier's limits become sales-qualified leads. Sales prospects who are not ready get routed into self-serve to build familiarity. Each motion feeds the other.

"If you can get a PLG motion work to feed your sales team and a sales team motion work to feed your PLG funnel when the sales leads aren't ready yet and you can get those motions into playing with each other, you can make a lot of money."
Shaun Clowes
▶ 00:53:00

Christopher Miller saw the transformation from the other direction. He joined HubSpot in 2016, about a year after they launched their free CRM -- a massive strategic bet without a fully formed plan. As VP of Product for Growth and AI, he helped create the growth team that transformed HubSpot from a legendary sales organization into one of the most successful PLG businesses in B2B SaaS history.

"When I joined HubSpot in 2016, it was maybe like a year or so after HubSpot had launched their free CRM, which was a big strategic play for them and for us at the time, and it was meant to be disruptive, but I don't think that there was a fully formed perspective on what was going to happen after that."
Christopher Miller
▶ 00:05:58

His insight about why even sales-led companies need PLG is structural: when revenue is tied to headcount, scaling gets harder the bigger you get. PLG breaks that linear relationship.

"I think any company is probably searching for ways to operate more efficiently. And if your revenue is so tied to go-to-market headcount, it gets really hard to scale the bigger you get. And so I think there's an innate desire to want to be more non-linear in our growth."
Christopher Miller
▶ 00:54:01

But his most practical warning: the number one mistake is under-resourcing. Every PLG veteran has a joke about the head of growth handed no tooling, no engineering, no data, and a scary number. The second mistake is expecting quick turnaround. PLG is R&D -- planting seeds for durable, efficient growth, not a direct revenue play.

"The number one mistake is hiring a head of growth, giving them no resources and expecting them to pull rabbit out of their hat."
Christopher Miller
▶ 01:04:55

Hila Qu launched GitLab's PLG motion on top of an established sales organization and now advises dozens of companies on the same transition through Reforge. Her most important reframing: PLG is fundamentally "DLG" -- data-led growth. When you give away a free product, you are making a trade. You get broader reach and usage data. Without the data infrastructure to analyze that data, you are giving away your product for nothing.

"PLG I always say is actually fundamentally DLG, data led growth. So when you give away your free product, what you want to get in exchange are two things. One is the broader reach because free product spread itself is lower barrier to entry. Two, you want to understand the usage behavior of those free users... If you don't have a foundation of data and understanding of how to analyze those data, you are giving away a free product for nothing."
Hila Qu
▶ 00:14:55

She warns that PLG requires a 1-2 year committed roadmap, a dedicated team, and organizational process changes. Companies that assign one person to "figure it out" are asking that person to be a magician. And the biggest red flag: thinking PLG equals "launch a free version."

"The red flags I've seen basically sometimes, first of all, they think about PLG equals launch of free version or launch of free trial. They made this assumption in their head, 'Oh, I already have a product. I already have a software that's working, customers are using. Now if I add a free version, if I open a free trial, that's it.'"
Hila Qu
▶ 00:16:27

When This Applies

You lack the data infrastructure to track and learn from free user behavior (Hila Qu's warning: "you are giving away a free product for nothing")
You equate "PLG" with "launch a free version" and have no roadmap for activation, retention, and conversion
Your product requires long time to value, high incremental cost to serve, and has no viral loops (Bobby Pinero's Equals lessons)
You plan to assign one person with no resources and call it a PLG initiative (Christopher Miller's "magician" warning)

Evidence from the Archive

SaaStr (founder); EchoSign/Adobe (former CEO)

Canva: deferred building a sales team until north of $500M in revenue because its self-serve motion was 'epic'

Portfolio company collapse: $5M ARR growing 100%+ to under $1M after firing the sales team because founders found sales 'icky'

Jason Lemkin · Founder of SaaStr, world's largest B2B/SaaS community ▶ 00:07:06

Snyk

Snyk had ~5,000 free users before any monetization attempts, deliberately deferring revenue to build adoption

F30D activation metric: teams that fixed a vulnerability within 30 days had strong retention and higher purchase propensity -- a team-based metric that drove expansion

Ben Williams · Former VP of Product at Snyk; PLG advisor ▶ Watch

Figma

Figma at Microsoft: Patient zero signed up organically; spread through node graphs of collaborator invitations...

Claire's scrappy early sales: 'I slid into the DMs of my friend's ex-boyfriend' to get the first meeting with Microsoft -- the ultimate bottom-up hustle

Claire Butler · First GTM hire at Figma (10th employee); leads bottom-up growth and community ▶ 00:15:27

Vercel (COO); formerly Stripe (CRO), Google

Sierra: Going after Global 2000 with eight-figure deals from day one -- PLG would not be useful here

Atlassian: The most famous PLG company eventually added sales, proving that PLG alone cannot sustain $100B+ scale

Jeanne DeWitt Grosser · COO at Vercel; built Stripe's early sales team ▶ 01:14:28

HubSpot

HubSpot's free CRM launch: a massive strategic bet that put an enormous piece of free software into the universe...

HubSpot's free CRM launch: a massive strategic bet that put an enormous piece of free software into the universe without a fully formed perspective on what would happen next

Christopher Miller · VP of Product, Growth and AI at HubSpot ▶ 00:05:58

Lovable (formerly Miro, Amplitude, SurveyMonkey, Dropbox)

Dropbox's sharing loop: a single product-led acquisition loop that has been firing for 17+ years, with its own...

Dropbox's sharing loop: a single product-led acquisition loop that has been firing for 17+ years, with its own dedicated growth pod optimizing both sender and recipient experiences

Elena Verna · Head of Growth at Lovable; legendary B2B growth advisor ▶ 00:34:00

The Synthesis

Everyone agrees on the destination. Every voice here -- from PLG advocate to sales champion -- acknowledges that successful B2B companies eventually run both motions. Elena Verna says failing to overlay every growth channel "is a huge mistake." Lemkin says "most companies are a hybrid." Jeanne cannot name a $100 billion PLG-only company. Hila Qu has witnessed many startups successfully running both motions from the early stage.

DNA Over Ideology

What single variable determines whether to lead with PLG or sales?

Layer Early

When should you start building the second growth motion?

Growth Model Clock

Why are you on a clock even if your growth motion is working?

PLG Friction Paradox

Can too little friction in onboarding actually hurt growth?

The right sequence is determined not by philosophy but by a single observable variable: how your first customers actually acquire and pay for your product. If they found you independently and paid with a credit card, your DNA is product-led. If they required demos and procurement, your DNA is sales-led.

The synthesis is: read your DNA, start there, and begin building the other motion earlier than feels comfortable. Slack built enterprise sales while self-serve was roaring. Figma added sales three years in. They all started layering the second motion before the first hit its ceiling, not after.

Growth models exhaust themselves within five to seven years. Enterprises will not self-serve into million-dollar deals. Competitors will add PLG, stealing your end users. Revenue tied to headcount creates scaling problems only PLG can solve. Even success has a built-in expiration date.

Sometimes the right amount of friction outperforms its absence. One company launched freemium, 4x'd users, then watched engagement and revenue tank. Reverting to a credit-card-required trial immediately recovered performance. The goal of onboarding is not getting people in -- it is getting people to their first moments of value.

Which Approach Fits You?

Answer 3 questions about your situation. We'll match you to the right approach.

Question 1

Can users experience your product's core value without talking to a human?

Question 2

What is your company's DNA?

Question 3

Have you hit a growth ceiling with your current approach?

Notable Absences

The Bottom Line

Hila Qu's DLG reframing is the key that makes this actionable: you cannot layer PLG without data infrastructure. You cannot layer sales without a replicable process. Both take time. That is why the clock is always ticking. Ben Williams' framework from Lenny's newsletter quantifies the PLG side: top PLG products attract 40 percent of users through organic search and SEO, with CAC below $1 per unique visitor. If your numbers are nowhere near that, either invest in product-led marketing or accept that sales is your primary motion.

The synthesis is not "it depends." It is: **read your DNA, start there, and begin building the other motion earlier than feels comfortable.** Slack built enterprise sales while self-serve was roaring. HubSpot launched a free CRM while their sales team was legendary. Figma added sales three years in when self-serve was printing money. Snyk deferred monetization while pouring into acquisition and activation. They all started layering the second motion *before* the first hit its ceiling, not after.

Sources

Elena Verna — "Elena Verna on how B2B growth is changing, product-led growth, product-led sales, why you should go freemium not trial, what features to make free, and much more" — June 23, 2022
Elena Verna — "The new AI growth playbook for 2026 | How Lovable hit $200M ARR in one year" — December 18, 2025
Elena Verna — "Countdown of the top 10 episodes of the year" — December 29, 2022
Jason Lemkin — "We replaced our sales team with 20 AI agents -- here's what happened next | Jason Lemkin (SaaStr)" — January 1, 2026
Hila Qu — "The ultimate guide to adding a PLG motion | Hila Qu (Reforge, GitLab)" — April 2, 2023
Ben Williams — "How Snyk built a product-led growth juggernaut | Ben Williams (VP of Product at Snyk)" — November 6, 2022
Merci Grace — "Merci Grace (ex-Head of Growth at Slack) on PLG, interviewing, storytelling, building a diverse team, hiring salespeople, building a growth team, and much more" — July 11, 2022
Claire Butler — "An inside look at Figma's unique GTM motion | Claire Butler (first GTM hire)" — September 7, 2023
Christopher Miller — "Relentless curiosity, radical accountability, and HubSpot's winning growth formula | Christopher Miller (VP of Product, Growth and AI)" — August 10, 2023
Jeanne DeWitt Grosser — "What world-class GTM looks like in 2026 | Jeanne DeWitt Grosser (Vercel, Stripe, Google)" — November 30, 2025
Shaun Clowes — "Why great AI products are all about the data | Shaun Clowes (CPO Confluent, ex-Salesforce, Atlassian)" — December 29, 2024
Hila Qu — "Five steps to starting your product-led growth motion" — January 10, 2023
Elena Verna — "Freemium vs. trial" — March 8, 2022
Bobby Pinero — "Lessons from going freemium: a decision that broke our business" — November 28, 2023
Lenny Rachitsky — "Product-led marketing" — March 28, 2023