Hacker News new | past | comments | ask | show | jobs | submit login
Reflecting on the founding, growth and maturing of Overleaf (errantscience.com)
96 points by JohnHammersley on Aug 20, 2017 | hide | past | web | favorite | 32 comments

Super. Rather than taking rejection by YC as end-of-story they went on to success.

Remember when applying to YC and other batch based accelerators: you might not get in but in some cases that says more about the accelerator than it does about you and there are plenty of other roads to success.

If one rejection by YC can deter you from your ultimate destiny, don't call yourself an entrepreneur.

> If one rejection by YC can deter you from your ultimate destiny, don't call yourself an entrepreneur.

Bingo!! this is the ACID test for an entrepreneur/founding team .

Or it can just mean that you didn't present the idea optimally or you didn't have your best day in the interview.

It doesn't have to be a fault of the idea itself or the accelerator.

(Comment made with no experience with accelerators, but with experience asking for research grants, which I assume is not so different).

> Or it can just mean that you didn't present the idea optimally

The parent comment very specifically covered that appropriately with this:

"but in some cases"

It doesn't necessarily mean YC were wrong to reject either. It's a prediction game.

And there is a definite limit on the number of people who would ever consider using Overleaf. I detest MS Word (and word processors in general) and use Overleaf, but the market for TeX solutions is obviously not in the same league of scale as Dropbox.

Well, if I were YC and I'd be trying to improve my betting average the ones I bet on that were misses and the the ones I didn't bet on that were hits would be studied very carefully.

A couple points on this though:

1. Whether you're an incubator or a VC, batting average doesn't really matter. All that really matters (from a financial perspective anyway) is that you don't miss out on the super successful hits. I think it was one or two years back that pg said that something like 70% of YC's total return, since founding, was Dropbox and AirBnB. Thus, while Overleaf has become successful, at the end of the day it's not something that would really move the needle much for YC. 2. It's all about percentages, so of course you're bound to miss some winners. It's also possible that since Overleaf's addressable market (scientific paper writing) is not gigantic, YC might be less rigorous in finding "diamonds in the rough" if a startup's addressable market is not that big (pure speculation on my part). 3. Numbers 1 and 2 should highlight that the needs and goals of investors and founders are not 100% aligned. Investors want to find the few mega home runs, while I think most founders would be happy with a double or a triple. Kudos to Overleaf for their perseverance and success!

While the math here certainly works out, I don't think it's the whole truth, at least not with regards to YC. While they probably want to maximise the chances of betting on the next dropbox, they also fund plenty of companies that very clearly are not the next dropbox, probably because their mission is also to fund good, profitable ideas in general, and that each of these clearly-not-dropbox bets is quite cheap.

So I would guess their strategy is more like 'try to fund all potential future unicorns, and fund a bunch of others that seem cool and in a way so that those don't lose money on average'. That's certainly the strategy I would have chosen!

This is probably not true, though. Many of the people at YC have said, many times, that what they're trying to do is find the next Dropbox, and everything else doesn't matter. They pretty explicitly talk about how hard to do this is, but how their efforts should go towards that anyway.

The thing is, it's hard to to tell, especially from the outside, whether a company can viably become a billion dollar company. Sure, you've seen plenty of YC companies that ended up either not going for it, or not looking like it's possible for them to end up as unicorns, but this is in some cases too late - you aren't hearing their pitch, in which they explain why they could be the next billion dollar startup.

Investors who want to make money take seriously the idea that one real success is the only thing that matters.

Also, in this case (and presumably others) rejection is exactly what founders needed. Judging by the write-up they were ill prepared and rejection is what forced them to take a huge step back and re-evaluate their approach. Would it happen if they were accepted or let go less harshly? Who knows.

PG and others have posted here several times saying they definitely do this. (And of course, what incubator or VC wouldn't.)

I would not expected anything less.

Maybe it's a tangent, but Overleaf is hands down one of my favorite services.

It follows UX designer, Bret Victor's suggestion that creators need to be able to see what they're creating. It also has a ton of convenient features like forking, auto-saves, and inviting people to edit through private links without requiring that your collaborators sign up for an account on Overleaf.

LaTex creates beautiful documents, but it was too difficult for me to use before Overleaf. Also, I only use the free version of Overleaf.

Thank you for creating Overleaf!

Thanks for this amazing comment! Overleaf definitely came from us solving a problem we had ourselves, and we've always continued to use it, which definitely helps you catch bugs and UX issues.

We've also hired a great team who have expanded on using Overleaf with clients -- for example, we use Overleaf for documentation when onboarding a new institution or publisher. It's really helped, and it's broadened out the type of feedback we receive too.

Thanks for using Overleaf! :)

Overleaf is an amazing service that needs more exposure. Right off the bat it takes away the steep learning curve of setting up a local LaTex creator. The real time generation is actually a lot faster now too.

Thanks, and yes, we've been doing a lot of work on scaling over the past few years to help with the compile times :)

We've just announced some exciting news too -- the ShareLaTeX team has recently joined Overleaf! More details on that here if you're interested: https://www.overleaf.com/blog/518-exciting-news-sharelatex-i...

Consider adding a link in that blog post to Overleaf, eg in the third paragraph where you say "(now Overleaf)." Clicked through to check it out after some of the positive comments in this thread, and found no direct link to it.

Good suggestion, thanks -- I've just added an update to the top of the post, and mentioned it there :)

Edit: just a note to say that I added this update to http://johnhammersley.com/?p=428 as that's where this HN story originally pointed to.

If they were accepted they might not have become the success they are. Programs like YC can put different types of pressures on a business.

Interesting read.

I've been using Overleaf almost every day for about a year (and absoultely love it!) and I am still not sure how it can be profitable :D

Glad to hear you've been finding Overleaf useful!

Re the business model: we don't rely on the user subscriptions for revenue -- we provide services for institutions (e.g. Overleaf Commons: https://www.overleaf.com/universities) and publishers (e.g. templates and direct submission links: https://www.overleaf.com/publishers#!publisherslist), as well as private cloud / local installs for companies: https://www.overleaf.com/enterprises

(and that allows us to provide a solid free plan that works well for most users, especially those just starting out with LaTeX)

But yes, the "how does it make money?" question was always one we found difficult to articulate well -- we certainly didn't do it in the YC interview! :)

> that allows us to provide a solid free plan that works well for most users

The fact that you offer a real (i.e. full-featured) free plan is really nice. Congrats on your success and finding a business model that works. :)

Browsing through their gallery of templates I found this lovely material CV/resume


Does it look very unprofessional or just slightly? I'm not actively looking but I like to have something up-to-date to send when people ask and this is tempting.

I think it is quite ugly. I have been using this plain modern cv template, which has gained me a few compliments :)


Personally, I think it looks unprofessional. It's often hard to not cross the border from Material Design to Fisher Price, and in this case I think they crossed it a bit.

Material design was so nice when they showed off images on the website advertising Google IO, and had shadows and subtle textures on things.

It turned out it was just using the MSPaint filled rectangle tool with bright colours. And then spacing everything so far apart that your 6.5" phone screen is pointless.

Unless I'm hiring for a position where design sensibility matters, I actively ignore things that are down to personal taste. But personally I find the colors way too loud, and I would question burning a lot of space on an image unless it was strongly relevant somehow.

I think the major function of a resume is to get you to the next stage. If this is for people who are asking you for something, then you already have their interest; getting to the next stage is about maintaining that interest. In that context, having something eye-catching and polarizing has little positive value and substantial risk. (If you were submitting paper resumes into a slush pile, getting attention can be more worth the risk, though.) So in your shoes I wouldn't do it.

Overleaf just rocks.

One of the things I wish existed earlier. The new default for collaborative research paper writing.

After getting into YC, you can see the large amount of different companies and markets and products they try to reach, but for sure the one thing the all share is billions market size.

Also, they are always clear that not being accepted into YC should not be the end of the company and you can try later even with the same idea.

Overleaf/ShareLateX are really great tools ...

However ... god I hate working with them if you have a lot of images that need to be compiled each time. It gets really really slow. That's my only gripe with Overleaf/ShareLaTex: the slow compiling.

Applications are open for YC Summer 2019

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact