Inside folk’s journey to 5x year-on-year growth

Co-founder Simo Lemhandez on four years of building a CRM startup and the path to 100,000 users

May 15, 2024

👋 Hi, it’s Kyle Poyar and welcome to Growth Unhinged, my weekly newsletter exploring the hidden playbooks behind the fastest-growing startups.

Today, I’m bringing you the 10th installment of my popular Zero to One series, unpacking the early growth journeys of promising tech startups. Up now: Accel-backed folk’s journey to 100,000 users and 5x year-on-year growth.

Simo Lemhandez started building folk, the next gen CRM, in April 2020 when he was just 25 along with co-founders Jean-Yves Poilleux and Thibaud Elziere. The Accel-backed startup now boasts:

Over 100,000 users
2,000+ paying customers
5x year-on-year revenue growth
Winning #1 product of the month on Product Hunt and two Golden Kitty awards after their 2022 and 2023 launches

I sat down with Simo to hear the behind-the-scenes story – folk’s four year journey from zero to 100,000 users. Here’s the TL;DR:

Attracted a waitlist of 10,000 people before writing a single line of code
Spent 18 months building before entering beta, and another six months before launching on Product Hunt
Doubled prices every six months to land on the right pricing strategy
Turned to influencers and affiliates to stand out in the most competitive software market on the planet
Manually onboarded hundreds of users (Superhuman-style) before moving to self-serve
Doubled free-to-paid conversion after experimenting with a new user onboarding flow

The early years: There’s no MVP here

The first thing Simo and team did when launching folk wasn’t creating an AWS account or writing code. It was building a landing page.

“It was like starting at the end and not at the beginning,” Simo told me. “Fake it until you make it.”

He wanted to see if folk’s idea resonated with an audience. It was an opportunity for the team to explain the pain they wanted to address and how they wanted to address it, and meet with potential users to go deeper.

The early landing page attracted significant interest – generating a waitlist of 10,000 people. This audience came organically from social media and word-of-mouth. Each waitlist signup provided an opportunity for user research. The folk founders would email them in a way that felt handwritten and personal, thanking users for signing up and asking them to reply with an explanation of what they were expecting from folk. While Simo and team weren’t planning to build exactly what prospects asked for, these replies provided insights about the exact customer pain that folk could solve.

“We learned that the market was massive, certainly the largest software category, and the NPS was very low. But it turned out that expectations were very high, too. Given that CRM is a mature category people take for granted that you offer all the features like analytics, audit trail, pipeline management, leads integrations, and so on.”

Put differently, there wasn’t a clear minimum viable product (MVP) here. The folk team would need to build – a lot. And they spent 18 months doing so.

“In a crowded market, it matters even more to show you're different than better,” Simo emphasized. folk planned to be different in two ways:

People-first rather than sales process-first: folk invested in non-standard CRM capabilities like a Chrome extension for adding contacts, conversation templates for Gmail and LinkedIn, and mail merging.
Make CRM more ‘Notion-like’: folk wanted to reset the expectations for CRMs following modern software practices like collaboration, integrations, ease-of-use, templates to start, and a light UI.

Only then did they start progressively opening the app to more users. They kept a close eye on engagement metrics like daily active users to monthly active users (DAU:MAU), weekly active users to monthly active users (WAU:MAU) and churn to assess product-market fit. And they complemented this with qualitative indicators that users are getting strong value from the product such as: whenever they release something, how quickly do users adopt it? If there’s a bug, do people report it instantly?

Launch: Manually onboarded hundreds of customers, doubled prices every six months

After six months of beta testing, Simo concluded folk was ready for prime time – a Product Hunt launch. The team concentrated attention on making as big of an echo as possible by spending significant time polishing materials, coordinating with their user base, and pulling in angel investors.

Spoiler: it worked. The April 2022 launch won Product of the Month and was nominated for Product of the Year.

At the same time folk began monetizing. “It was a very manual monetization process at the beginning,” Simo recalled. “We were sending Stripe payment links one by one to each customer.”

folk required customers to pay even to see the app. Inspired by Superhuman’s early approach of white-glove onboarding, Simo and team personally onboarded every single customer. During onboarding calls they’d go through the same discovery process; after hundreds of these, the founders developed an instinctive understanding of customers’ pain points, goals, and alternative solutions.

Simo emphasized that this wasn’t as challenging as it might sound. “It wasn’t an effort for us to have customer calls booked. Given that we set up email automations, customer calls automatically arrived on our calendars. The routine of talking to customers was effortless.”

Back then pricing was “super low” – only $5 per month. Folk has since updated pricing plans every five to six months. This could mean new packages or new limits, different value metrics, or changes to the price points themselves.

“We usually doubled the price. Each time we doubled the price, we were reluctant to do it. But we realized it wasn’t changing conversion to paid and was obviously increasing ACV, so we kept doing it.”

Interestingly, Simo also found that a higher price can anchor the product as more valuable in people’s mind, and therefore make them more likely to invest into it. This initial investment is necessary for success with high-friction tools like a CRM.

Simo believes that revisiting pricing every six months is roughly the right cadence for a startup. It takes time to collect data and reflect on the learnings from the last pricing changes – conversion, churn, upgrades, etc. – and incorporate those learnings in the new pricing (but don’t wait too long before the next iteration).

Acquisition: Experimented to find scalable growth channels

In September 2022, six months post-launch, folk finally integrated pricing in-app. It was time to look for scalable growth channels. Simo and team ran a rapid experimentation approach to test growth channels, systematically scoring each channel based on five criteria:

Volume of signups
CAC payback period
Level of effort
Overall revenue generated by the channel
Overall churn generated by the channel

A major channel folk tested was LinkedIn, which is where customers were spending their time and where there was a tie-in to one of folk’s premium features (a LinkedIn plug-in). They started with a few B2B micro-influencers, initially helping these influencers with the content itself–suggesting angles, images, etc. while leaving the writing to the influencers. Not only did this work, it created a flywheel where it attracted stronger organic user love as well.

Rigorous attribution on LinkedIn will be a pain, Simo warned. “Even if you use UTM links, it’ll be a nightmare in terms of operations and it might make the audience feel like it’s fake.” He found the best approach is to maintain existing tracking and to observe differences in signups based on days with versus days without LinkedIn posts. If there’s a significant signup change, you can infer that it’s working even without bulletproof attribution. (Casey Hill walked through ActiveCampaign’s LinkedIn attribution approach in a prior story.)

Interestingly, the biggest influencers weren’t the biggest hit in terms of attracting users. “We ended up with way better CAC payback and conversion when we had a niche influencer with a very targeted audience that matched the type of content our audience naturally interacts with.”

From there the team turned to other low-cost channels like affiliates and SEO. On the affiliate side, Simo admitted that they were “dragged into it” – users were sharing folk and asking for affiliate links. While a traditional referral program didn’t work well for the company (most people didn’t care about getting a discount to save their company money), an affiliate program did.

Simo observed that there’s a “power law” with this channel where the top 5-10% of affiliates bring in 80% of the revenue. He thinks that affiliate programs work best for products with strong organic love, but it takes time to find your best affiliate persona.

The third pillar in folk’s acquisition is SEO. “This is the channel which you know you have to invest in for the long term and you know it takes time. The sooner you start, the better.”

Simo began by pulling a long list of keywords that had enough volume, acceptable competition, and – most importantly – good signs of intent (i.e. the people who are searching for this keyword are likely to pay for a subscription). Then, it came down to consistent testing. folk heavily leverages AI for this part; they look at what happens if they add a video, a comparison page, etc. and optimize accordingly.

Product growth: Adjusted onboarding to 2x conversion

From day one, the folk team recorded and categorized every single piece of feedback they got. “That helps for the feedback loop, it helps make your customer feel valued, and it helps even more to create a shared understanding of customer pains.” While Simo doesn’t believe that folk should ever build exactly what customers say they want, the feedback helped hone on the right problems to solve.

Two product surface areas emerged from this feedback and have seen lots of iteration: onboarding and payment flows. This iteration has had a massive impact, too, doubling free-to-paid conversion.

“We are always looking to understand, in a granular way, the key actions that are the best predictors for long term retention in the product. Then we set up onboarding in a way that maximizes these for users. Good friction isn’t necessarily a bad thing during onboarding; a few additional steps that make the user well set up can make a big difference.”

User onboarding screens — Early user onboarding at folk

Previously, folk kept onboarding to a few screens in order to minimize friction (see the images above):

Screen 1: Name your network
Screen 2: Pick your groups of contacts to store in your network
Screen 3: Tell us about yourself
Screen 4: Invite your teammates

Folk has steadily evolved onboarding since then, testing multiple versions before landing on the current approach. The winning flow follows several design principles:

Show, don’t tell
Very simple flows that tell a story linked to the benefits the users are looking for (e.g. sell more, build partnerships, recruit candidates)
Be opinionated about the best next action (folk’s opinions came from measuring the actions that are linked to long-term retention and making sure users adopt them)
Make the users import their data for the action that’s linked with long term retention, actions like source sync, template adoption and the Chrome extension
Show them what success look like to avoid the blank page problem
Personalize the experience based on what users tell us (their name, company, etc.)
Make users feel folk built the experience for them to be successful rather than for the company to simply gather data
Be aware of the way users feel during onboarding. For example, when sharing sensitive data folk should reassure them. When folk does something great for users, the product should say it loudly.

Winning onboarding screens — New and improved user onboarding at folk

You can notice these improvements in the revised onboarding flow (see the carousel above), which has value-centric messaging, drives users toward key milestones, and ensure the app gets properly set up.

Org design: Kept a lean team to stay efficient

folks has grown to 2,000 paying customers and 100,000 users with a team of 25 people. About half of the 25 are engineers and another four are part of onboarding (the manual onboarding covered earlier). There’s only one growth person, marketer and SDR on the team. (folk still hasn’t hired any product managers, by the way; product remains a distributed role led by engineering.)

Simo believes that early decisions on team structure have a big impact on a company’s future; the unit economics essentially get ingrained as a company scales. That’s influenced him to keep folk extremely lean (high ARR per FTE).

“We try to be slightly stretched so that we either automate stuff or solve it in a product-first way rather than thinking that hiring a new person will always solve the problem,” Simo said. “There’s a strong constraint we impose on ourselves not to hire too fast and to not consider hiring as the only solution to solving a problem.”

As much as Simo loves efficiency and the magic of self-serve, he doesn’t see the personalized onboarding role going away. “With self-serve, you put more signups in the funnel and they end up converting without additional costs in the machine. But when you think about it, the costs that you aren’t putting in the machine, the customers are actually paying for [with their time]. That’s why it’s a balance.”

Simo measures the expected revenue of every signup – using an internal-built lead scoring algorithm that combines expected conversion and expected annual contract value (ACV) – and only assigns an onboarding manager if the expected revenue exceeds a certain threshold. This was possible because folk invested in a data warehouse where they centralize data from multiple sources (like product data, Segment, acquisition, Stripe), compute the lead score, and then re-send this new data back to their other tools (like Intercom).

He’s discovered that having a manual onboarding session with these customers improves the chances of conversion by 4x. At that point it becomes a question of unit economics; what’s the threshold of expected revenue in order to make the model work at scale?

It’s a virtuous cycle, Simo highlighted. The self-service focus makes the product better for all users. And the option of white glove onboarding is a benefit that collects richer feedback from high value customers.

The TL;DR: Learnings from folk’s growth journey

Attracted a waitlist of 10,000 people before writing a single line of code. Each waitlist signup provided an opportunity for user research.
Spent 18 months building before entering beta, and another 6 months before launching on Product Hunt. In a highly competitive market, it was important to validate user engagement (DAU:MAU) and churn in order to assess product-market fit.
Doubled prices every six months to land on the right pricing strategy. Each time folk doubled the price, they were reluctant to do it. But it didn’t hurt conversion and did meaningfully improve ACV.
Turned to influencers and affiliates to stand out in the most competitive software market on the planet. They observed a “power law” with this channel where the top 5-10% of affiliates bring in 80% of the revenue.
Manually onboarded hundreds of users (Superhuman-style) before moving to self-serve. Even today, manual onboarding has a 4x impact on conversion rates.
Doubled free-to-paid conversion after experimenting with a new user onboarding flow. Good friction isn’t necessarily a bad thing during onboarding; a few extra steps can set the user up for success.