$30,000 In Debt To Building A $4 Billion Company: The Story Of How Three Cofounders Beat Impossible Odds At Webflow
The making of a Silicon Valley unicorn is usually a function of brilliant early-stage execution, calculated resource allocation, meticulous go-to-market strategy, and rapid growth. In that lens, Webflow, now worth $4 billion, beat impossible odds and its journey to build a no-code website builder for the masses riddled with fierce competition.
The company’s journey began just like many others: in 2012, co-founder Vlad Magdalin worked on Webflow after coming home from his software engineering role at Intuit, but ultimately left to focus on building an MVP.
The difficulties were apparent early–and got serious quickly. With $25,000 saved up between him and his wife, two kids (ages 1 and 3), no streams of income, and the challenges of raising venture capital in the early days, Vlad racked up tens of thousands of credit card debt while trying to get Webflow to market with almost no signs of traction.
Fast forward 10 years, Webflow serves over 200,000 customers across freelancers, agencies, startups, and enterprises, and is about to hit $100 million in annual revenue.
I recently spoke with Vlad, who leads Webflow’s team of over 400, to learn more about the company’s growth story. In our conversation, he shares a candid account of the company’s earliest days, how the team executed on supporting individual- and enterprise-level customers, the two biggest risks he’s taken as CEO, and much more.
Steven Li: Webflow’s founding year, 2012, seemed like a really tough one. At the time, you racked up $30,000 in credit card debt trying to get the first iteration of the product to market and were the primary breadwinner in your family. What did these early days look like from a founder’s perspective and how did you think about resource allocation?
Vlad Magdalin: Our strategy for resource allocation at the time was “try to survive as long as possible and devote as much time to figuring out a working product as possible.” Initially, it was just me working on Webflow starting in early 2012, and I tried to carve out “moonlighting” time to try to make progress. I still had my day job at Intuit, and I shifted my schedule to start work at 7 am, skip lunch, and clock out at 2 pm. I would then head over to Red Rock Coffee in downtown Mountain View, find a spot to settle in, and design/code until 8 pm every day.
Throughout the summer of 2012, I realized that there was no way that I would be able to make meaningful progress while moonlighting, and decided to start working on Webflow full-time around July. Luckily, I had a retention bonus (which Intuit had awarded me when I first tried to quit to start Webflow earlier that year) that just came due, and I was able to net about $10K after taxes in addition to the $15K or so my wife and I had in our savings account.
A few weeks after quitting, it became obvious that working on this by myself was going to be tediously slow. In desperation, I reached out to my brother Sergie, who I had done a lot of freelance website designs with in the past, and asked him to help me out with some contracting work. He had recently graduated from UCSD, and was working at a local skate shop doing various graphic and web design for them. That conversation quickly turned into the idea of him joining as Webflow’s co-founder, and by early August he was making plans to move up from San Diego to Mountain View.
Once Sergie moved into our family’s small condo (in order to avoid additional rent expenses), we started working effectively 7 am to midnight daily – with the exception of Sunday mornings until noon, which my wife had wisely claimed as non-negotiable family time. It’s hard to remember exactly how we allocated the time, other than knowing that you could always find us sitting side by side at the Hacker Dojo – one of us designing, one of us coding.
In terms of money, we were honestly pretty naive in allocating it correctly at first. Out of the total $25K that my wife and I had in savings, $5K immediately went towards incorporation, another $5K towards new laptops (which we totally could have saved money on), and more than $10K towards the production of a Kickstarter video that ultimately never launched. It was only when we were basically out of cash that we started budgeting religiously – including eating the same exact dinner for months on end (a single fajita plate from Una Mas that cost ~$9.50 and could be split to make two separate “burritos”).
In early 2013, we were able to convince Bryant to join our team as our 3rd co-founder. When he joined, we had not yet built the product and were pretty deep in debt – in fact, if memory serves correctly, our company bank account had less than $500 in it on his first day. Thankfully, he was able to bail us out by injecting some of his own savings into the company, which kept us afloat until we got our first infusion of funds from Y Combinator.
In short, we were very lucky to survive those early days, and in hindsight I can honestly say that the biggest factor in helping us keep going was an unshakeable–or, borderline delusional–sense of optimism that it was all going to work out somehow.
Li: Despite these early setbacks, Webflow had a successful Hacker News launch that netted the team over 20,000 signups. That was a watershed moment for your company at the time–and it certainly played a role in helping the team get into YC and raise its first significant capital. What preparation went into the product launch behind the scenes and what lessons did the team learn from it?
Magdalin: The Hacker News launch was, in fact, a last-ditch attempt after other places like Digg and Reddit didn’t get any traction – something we were really disappointed by because those were the places we knew that many more designers hung out than at HackerNews. (I told a bit of that story on HN a while ago.)
When Digg and Reddit didn’t get any traction earlier, we were pretty dejected and decided that the only way that we were going to get any interest was to try to be as loud as possible on social media. But none of us had any meaningful number of followers, and could really only share with our friends on Facebook. Luckily, we ran into a new product called KickoffLabs around that time which allowed us to add a way for people to join a waitlist on our landing page and get a unique referral link – if they got others to sign up for the waitlist using their link, we promised to get them earlier access to Webflow when it was ready.
Long story short, that mechanism didn’t help us go viral as we had hoped, and kind of fizzled out over a few days. But having it in place ended up being absolutely crucial once we decided to post to Hacker News as a last resort. We honestly thought it was going to fall flat on the coder-centric site since our product was all about auto-generating code to build websites.
But despite all of our assumptions, the post took off like wildfire, and we had over 5,000 people join our waiting list just from that HN post. And crucially, over the next few days, the KickoffLabs viral mechanism also started working really well – many of the people who signed up wanted earlier access, and we sent out several emails encouraging them to share with their friends to move up in line. This led to over 20,000 more signups over the next few weeks, primarily fueled by people posting their referral links to their social networks.
So I’d say there were two primary factors – the viral referral mechanism facilitated by KickoffLabs, combined with what I believe to be sheer luck of HN being so accepting of a novel idea even though our product was not targeted towards programmers.
Li: I learned that out of the 20,000 initial signups there were only about 50 paying members. How important was it to focus on these users in the early days, what did the team do to pay more attention to them, and what were the long-term implications of deciding to focus on this small but passionate user base?
Magdalin: It was a very disappointing conversion rate at the time. For the people who wanted to pay for it, though, it became clear that it was a product that was very important to them – with some using it for over 10 hours a day. When we had such a big waitlist, I think we had too much false confidence in the “if we build it, they will come” mentality – and at first we weren’t worried about spending a lot of time talking to users.
But once we saw how hard it was to get people to pay for our product, we realized pretty quickly that we absolutely had to do everything in our power to make these paying users happy – and we focused nearly all of our non-building time and attention to make sure they were successful. We didn’t intentionally decide to focus on this smaller audience but rather had to come to terms with the reality that if we couldn’t retain these early adopters then we wouldn’t have anything resembling a working business.
The two most crucial things we did to focus on these early users were to set up a public community forum (where we’d each spend at least several hours a day answering questions, talking about feature requests, and investigating bug reports) and to make sure that all three of us were present in our support queue to answer support requests quickly.
This made sure that people who trusted us enough to use a limited–and often buggy–product felt like the Webflow founders were always there, and in many ways, it felt like co-building a product with our community of early power users at the time.
In hindsight, it was the absolutely correct investment of our time and attention, because as that handful of early users became successful, they then became active helpers in the forum themselves, which formed the early foundations of our vibrant community.
Li: Coming off of the momentum from Webflow’s Hacker News launch, you, Bryant, and Sergie got into YC; those three months ended up being crucial for Webflow. Can you help me understand what responsibilities each co-founder had during that time and provide some context around the progress the team made during the program?
Magdalin: When we started YC, all of our time was focused on actually building the first version of the working product. We started the program with only a frontend demo, with zero backend code, and no ability to even sign up as a user.
Sergie focused on constantly iterating our UI, creating tutorials, and getting all of our marketing ready for launch. Bryant built the entire backend, hosting infrastructure, and all the dashboard parts of our application in those summer months. I focused on the frontend work in our core Designer tool, primarily working out massive performance issues that we had with our original frontend demo.
Near the end of the summer, as Demo Day was approaching in August, we seriously didn’t feel anywhere near comfortable launching – and felt bad trying to charge for what would be a very limited product. When YC first started, we thought we’d build a product that had feature parity with WordPress – including blogging, dynamic content, and plugins.
But by the end of summer, we only built the very basic parts of Webflow, and it was only possible to build one-page websites that had limited utility. We were tempted to not launch at all until blogging was ready (it ended up taking another two years to build!), but the partners at YC convinced us to launch early anyway even though we were really embarrassed by how limited Webflow was.
I truly believe this was one of the most pivotal decisions in our history, and I’m convinced that if we didn’t have the healthy pressure that YC provided we might have died as a company before we launched a single thing.
Li: In order to not raise more money for another four years after the seed, Webflow must’ve had to focus heavily on making the unit economics work. Can you share the breakdown of how the team invested its capital, including the funding it raised as well as its profits generated?
Magdalin: We eventually got better at analyzing our revenue metrics as the years went on, but for the first few years I personally had a very naive way of judging the health of our business. I would log into our SVB bank account every morning, and see how much cash we had in the bank. I would then copy-paste the bank balance to a spreadsheet I kept that graphed our cash balance over time.
One morning around early 2015, I had a realization that at the pace we were burning cash, if we didn’t change anything we’d be “default dead” within a year. And at that moment, I decided that we must become “default alive” – or what I called at the time “in control of our own destiny” – meaning that we have to find some way to turn around the weekly decline in our cash balance. Our revenue was growing, but quite slowly compared to other early-stage companies in our peer group. So the lever we pulled was to significantly slow down our hiring pace (like most startups, our largest expense was payroll), to give more time for our revenue to catch up.
About a year later, we hit that magical moment when the slope of the graph of our cash balance stopped being negative and turned flat – meaning we were cash flow breakeven. At that moment, I had known that we were now on the path to being a truly sustainable business. For the next many years (up until our Series A in 2019), we would only hire new folks when revenue increased enough to be able to afford the added salary. We decided that we weren’t really looking to generate any profits, but rather to quickly convert any revenue that came in above expenses to a new job opening so that we could hire another person – which helped us invest faster into our product development.
Li: How was Webflow able to reach profitability so quickly while managing rapid growth? In comparison, for example, it took Squarespace and other competitors quite a long time to reach profitability.
Magdalin: Even though we rapidly grew the number of users we had in the first few months, a lot of that didn’t manifest in durable customer accounts as people were trying out the product. We grew actual paying customers and revenue pretty slowly (usually well below 7-8% a month even on a small revenue base) for many years, and because we wanted to be “default alive” we also hired slowly as revenue allowed.
We reached cash flow break-even several years in, but primarily out of necessity so that we wouldn’t die (since our business wasn’t growing fast enough to warrant a Series A financing round until way later in 2019) after that moment in 2015 that we were going to run out of cash.
In retrospect, we were really lucky that a major competitor didn’t emerge during those years (luckily so many other companies were focusing on mobile at the time), because it gave us the luxury of being able to slow down our cash burn and rate of hiring long enough to get to escape velocity.
Li: What is your mental model around growing individual customers vs. enterprise customers? I read that for the first five years Webflow focused exclusively on non-enterprise customers, but I imagine the enterprise customers would generate more revenue but might demand more features, which would also demand greater investment in product and engineering.
Magdalin: From our earliest YC days, we built a bit of an early aversion to seeking large enterprise customers, in large part because our first enterprise-scale deal (circa 2014) came with a lot of custom requests that even required us to share our entire codebase in order to run a copy of Webflow behind the customer’s firewall. It was a significant amount of revenue (over $100K when all of our other customers were paying us less than $50 per month), it showed us how much of a distraction those kinds of deals can be from focusing on our core user community.
In hindsight, however, I wish we had not written off a sales-assisted motion completely for so long. We started re-exploring that motion about a year and a half ago, and almost immediately saw how effective it was in helping larger companies onboard onto Webflow’s platform. Seemingly small things like having a person to talk to in order to better understand our product, or having a service level agreement that builds confidence in uptime for mission-critical websites, have led to really phenomenal growth (over 500% revenue growth last year) among customers where our Sales team helps bring them onboard.
Enterprise customers do indeed need more features tailored to larger teams, but thankfully our core visual development functionality is valued by companies of all sizes. The enterprise features tend to be more “around” the core product – e.g. additional authentication options, more robust collaboration workflows, and so on. The vast majority of the time, building a new feature for our self-serve customers also benefits all of our larger enterprise customers.
Our mission is to eventually empower everyone to create for the web, and we can’t genuinely do that if we only focus on the largest companies or only on individuals. So we always aim to build a product that serves companies both large and small, while still being empowering and accessible to any individual (including students) who wants to learn visual development and harness it to make a living.
Li: What does the typical sales cycle look like for Webflow to win an enterprise customer?
Magdalin: Despite how large the no-code space has grown to be, we still see the vast majority of our competitors come from either non-consumption (e.g. traditional developers using code to run a marketing site) or more established CMS platforms like WordPress. For that reason, we don’t typically have to go head-to-head with other visual development platforms (primarily because none exist that are nearly as powerful as Webflow, especially for marketing sites), but rather have the task of convincing companies that there’s a much better way to power their marketing stack than code-based approaches.
The typical sales cycle usually starts with someone at a company already using Webflow actively for an existing project, or someone who already has deep familiarity with it in the past and has an individual account that they signed up for via our typical self-serve motion. They then “raise their hand” by contacting our Sales team (typically via our main enterprise marketing page which talks about the benefits of our Enterprise plan), which starts a back and forth conversation to closely understand their needs and requirements.
This usually involves multiple demos to key stakeholders, detailed discovery of what their team is trying to accomplish, and various working sessions to ensure that Webflow can indeed meet their needs. That’s followed by formal steps like security reviews, contract negotiations, and so on.
Li: As Webflow matured as a business, what have been the three most important business changes that were needed to sustain growth?
Magdalin: There are too many changes to list here as it seems that every day brings new learnings and shifts to adjust to new information, but I’d say the most impactful have been:
Defining and evolving our company’s Core Behaviors (values as seen through actual actions) as our team scaled into the hundreds to ensure we had a common understanding of our desired cultural and behavioral norms
Investing much more attention into marketing, since even with strong word-of-mouth adoption we were still not organically reaching all the people that could stand to benefit from Webflow’s platform
Deciding to focus not only on the design phase of creating a website, but very crucially on the importance of the value brought to customers via managed hosting – which makes a much more sticky offering, since folks are much less likely to churn if they have a production website running on a custom domain
Li: What were the biggest risks that you had to take as CEO of Webflow, when did they happen, and what were the results?
Magdalin: By far the biggest risk I ever took was going into really significant debt to help sustain my ability to keep working on Webflow over the first year when we didn’t have a product and didn’t have any revenue. It was especially difficult since I had two young kids at the time, and we had zero streams of income as a family.
Initially, I had assured my wife that we’d replace my income within 3 months of starting the company, but that assurance quickly evaporated as we went deeper and deeper into debt – and it wasn’t until our seed round a year later that we were able to stop borrowing money (mostly by way of credit card balance transfers, which luckily I had good enough credit to have access to at the time) to pay for normal family expenses like rent, groceries, etc.
The second biggest risk came in 2019, when I saw that our self-funded pace of movement towards our ambitious product vision was too slow, and decided to reconsider the idea of venture capital to accelerate our progress.
At the time, I was still very afraid that VCs would take control of the company and lead us in a direction that I wasn’t proud of, and it felt like a risky one-way door to take a lot of other people’s money to help fund our ambitions. Luckily, we found totally mission- and values-aligned partners who genuinely want us to succeed in building toward our long-term vision, and that risk turned out to be one of the best decisions we’ve made as a company.
Li: What does the next chapter of Webflow’s growth look like and how will you need to adapt as CEO in order to take Webflow to this next level?
Magdalin: The next chapter of Webflow’s growth is likely not going to be a result of any one feature, go-to-market motion, or business strategy. Instead, it’s going to be a combination of so many aspects of our product, organization, and ecosystem working together to keep expanding what is possible with no-code and visual development.
From major product expansions like our upcoming logic capabilities, to empowering visual developers to create very powerful membership experiences (even mini-SaaS products!), to closing the feature parity gap with mature platforms like WordPress, to bringing visual development education to every student, to harnessing the power of our community to turbocharge our ecosystem, to bringing developers into the fold to extend the power of our platform – it feels we’re still in the early phases of Webflow’s ultimate long-term potential to democratize software creation.
And in terms of how I’ll need to change as a CEO to help lead the company through this next stage of growth, I don’t think I can do a better job at describing that transformation than Ali Rowghani (our newest board partner) does in his essay about The Second Job of a Startup CEO.
Everything he describes – hiring a leadership team, creating purpose and alignment, and nurturing our company culture – comes much less naturally to me than sitting down and writing code. But it’s the most important work I can possibly be doing right now in order to maximize the odds that our mission and our company will reach its full potential.