Reflecting on the founding, growth and maturing of Overleaf

“Naively optimistic” is a phrase I often find myself using to describe the necessary mindset of a first-time start-up founder. It’s one that certainly applied – and probably still applies – to me.

It’s a variation of the oft-repeated comment that if you could see all the work ahead, and the effort required, you’d quit now (or never get started).

Looking back at the journey we’ve taken to grow Overleaf from a weekend-project into a fully-fledged SaaS (‘Software as a Service’) platform, it’s hard to believe how much work and effort has gone into it – certainly more than I would have been able to face if you’d listed it all out right at the beginning.

As a first-time founder, you really won’t know what’s coming up – the chance of you accurately predicting all the real challenges ahead is effectively zero – and if you focus on all the challenges ahead (whether real or most likely perceived ones), you’ll never get anything done. By limiting your horizon to the immediate future, the problems become tractable and achievable. As Y-Combinator (YC) partner and investor Paul Buchheit said in a recent interview, “I also like people who get things done. If you’ve been working on your startup for two years and have nothing to show, that’s a pretty bad sign. I’ve discovered that most people are really good at finding obstacles. I don’t fund these people.”

I should note that I speak with all the bias of someone describing their own particular journey, and everyone’s experiences will be different. However, a focus on ‘getting things done’ is a mindset I’ve also seen in other successful founding teams, having been part of the mentor networks at Bethnal Green Ventures (BGV) and within the scientific start-up community for the past few years. Perhaps more tellingly, those founding teams without this type of approach are the ones that often seemed to struggle.

Of course, you’ve always got to have an idea of where you’re aiming for in the long run, and I’m certainly not suggesting you limit the vision for your company; you should always keep sight of what you’re hoping to achieve, and why it matters. But so much is fluid in the early days of a startup that detailed long-term planning is a waste of time, and a distraction you can ill afford. By all means, come up with a 5-year business plan – at a high level it’s often a useful sense check of your assumptions – but put 95% of your effort into the current year, and most of that effort into what you’re going to do in the next three to six months.

Our vision was to make science and research faster and more transparent, and hence fuel scientific and technological innovation, by letting scientists collaborate more easily and more effectively – and it’s that vision which you have to project outward like we did in our BGV Demo Day pitch back in September 2013, after spending three months on their startup accelerator programme.

Not only did we have a compelling vision, we’d also built something and had traction; we’d grown from 8,000 users when we joined BGV to over 25,000 a few months later. Putting both together gave us a strong pitch, which lead to our first investment round with Digital Science, a strategic investor in the research space who’ve been a fantastic supporter of and partner to Overleaf ever since.

Four years further on we’ve just announced our biggest news to date – we’re joining forces with ShareLaTeX to create a community of over two million authors as we continue to build the best tools for collaborative writing. I’m hugely grateful to the BGV team for their input over those three months in 2013 which helped us join the dots together much faster than we would have done on our own.

Contrast this with our trip to San Francisco for our Y-Combinator interview in late 2012 – this was hands-down my worst trip abroad, ever. We went from being cautiously optimistic to having what felt to me to be the worst 10-minute interview possible, and after getting the rejection the next day, spent the next few days of the trip working out where to go from there.

Why such a terrible interview? Firstly, we’d not listened to those friends and colleagues in business who were detached from our niche of scientific writing and who responded with a “Huh, what? How does that make any money?” when we explained the concept of WriteLaTeX (as it was called then) to them. Rather than listening to their confusion, we assumed that the YC interviewers would just ‘get it’ – because to us it was obvious. The opening question (from Paul B if we remember correctly) of “Aren’t you just multiplying two small markets together [into a much smaller one]?” was the perfect opportunity for us to talk about the wider vision, and we never really recovered from being unable to articulate it (because we hadn’t really figured it out at that point).

We’d also spent the last few weeks before the interview launching a new homepage that looked much, much worse than the previous one; although there was more content, we picked a horrible new colour palette that made it looked like we’d designed a website from the 90s. Needless to say, our demo in the interview kept their interest for about 5 seconds, and it actually took away from the underlying technology which was pretty impressive.

Thankfully we picked ourselves up (that naive optimism kicked back in), got onto the BGV programme, and have carried that momentum forward ever since. But if we’d spent too long worrying about all the challenges, that lay ahead – especially after such a demoralising interview – we might have quit before we’d even got started.

 

To find out more about the beginnings of Overleaf, check out this Reddit AMA where I talk about my transition from academia to industry to startup founder, or just head to www.overleaf.com and explore from there!

John Hammersley

Mathematician; physicist; dancer. Entrepreneur; founder; CEO. Co-founder of Overleaf. Did driverless taxis at UltraPRT. Forever learning :)

3 thoughts on “Reflecting on the founding, growth and maturing of Overleaf

  • August 2, 2017 at 17:03
    Permalink

    One thing I’ve often wondered about start-ups is “What is the right time to reveal your idea to other people?” I am guessing that as an engineer you had real-life examples of the frustrations of working with LaTeX which you could discuss with your colleagues, find out that they shared your annoyances and thus were convinced that there was a market for a solution. Not everyone who starts a company (not even everyone who starts a successful company, I am guessing) has that luxury, yet we are told constantly that there is rarely a “Eureka!” moment in any invention, and that is far more commonplace for good ideas to take shape in an iterative way, by consultation over time.

    Reply
  • August 2, 2017 at 23:08
    Permalink

    Hi Mark,

    In my view, the best time to reveal your idea to other people is as soon as it naturally would come up in conversation. If you’re needing to force an idea out there, it’ll probably won’t have the right context to get a useful reaction. If the natural point never comes up, then you have to ask yourself why not, as if it’s a useful idea that you’re enthusiastic about it would be worrying to never feel like talking about it.

    I also echo advice heard elsewhere that you shouldn’t keep an idea to yourself for fear of being scooped. Early on you need positive reaction (for motivation), early adoption (for validation), and critical feedback (for development), and you don’t get any of those if you keep the idea to yourself. Overleaf (then called WriteLaTeX) wouldn’t have got off the ground if it hadn’t been available publicly for people to use from a very early stage. Those early adopters (who mainly found it through word of mouth and through searching for things like “How do I write in LaTeX?”) shared it with more people, and it showed us that hey, maybe we aren’t just solving a problem we have, but one that many people have, and it convinced us to start working on it full time. Through those early users we were featured on Hacker News a couple of times, and we got a lot of early traction from it being online early.

    Hope that helps!
    John

    Reply
  • Pingback:Five years on from flunking our YC interview – John Hammersley

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.