Hacker News new | past | comments | ask | show | jobs | submit login
Octopart, The Little Startup That Hung In There (techcrunch.com)
184 points by jagira on May 19, 2011 | hide | past | web | favorite | 53 comments

I really like this new theme on TechCrunch today, giving some attention to under-reported startups doing it "the old-fashioned way". I know it can't last, because it won't get as many page views as other articles, but it's nice.

I'm really looking forward to having an excuse to use Octopart. I've only occasionally had to find electronics bits and pieces in the past, and it's always been a pain in the ass. Thanks for building something people need, guys.

This is my favorite YC startup. I remember in college trying to design and build my computer. It was just awesome to have all the parts, data sheets, and price in one spot. Sure beets an excel sheet, keeping track of data sheets, and going everywhere to find every last part! Glad to hear they hung in there and are doing well!

Thanks so much for the comment. It's great to hear that hardware hackers find Octopart useful. We have a lot of ideas on how to improve Octopart and now we can finally implement them.

Mine too. Worked in the Oil industry as a buyer during college and Octopart was pretty useful at finding things I had in my RFQ's.

I've been a hardware hacker since I could walk. Finding and actually obtaining any parts at all in the 80's was excruciating. Even in the bad old days of the web, datasheets and even pricelists seemed to be the locus of some very unseemly rent-seeking activity.

I can't even begin to say how awesome octopart seemed when I first discovered it. Even today, when I go looking for a part I find myself stopping and not believing how awesome this is.

datasheets and even pricelists seemed to be the locus of some very unseemly rent-seeking activity

I once made the mistake of paying $10 to subscribe to a datasheet web site because they seemed to have the full tech specs (i.e. register layouts) of an IC I was reverse engineering (see http://myhd.sourceforge.net/). It turned out to be the same marketing materials I had downloaded from the chip vendor's web site.

Octopart is like the million-dollar part spec databases that electronics manufacturers buy just to have available in their design software, but for free.

seriously not trying to be a downer, but doesn't this smack of starting a company just to start a company? they all took $0 in salary in 2008, I wonder if they are making live-able wages now in 2011. They took on their first employee and maybe had to again cut their salaries to afford it? YC worries me that it convinces everyone to start a company rather than go to school, work for another company as an employee or follow a different dream. We're being sold someone else's dream of "starting a company is the best thing in the world to do". The talent pool is being severely diluted just like too many expansion teams in a sports league. The end result? Less success for everyone. I admire their persistence, but I wonder if Octopart is in a market where they can build a company that can hire and pay the employees or maybe even generate a larger lump of cash. I may not get the new model yet, but I still think that revenue generation is the goal of a company. However, it's just roughly 3 years into their lifespan, so maybe they will eventually build a nice business. I definitely wish them luck if this is truly their dream.

Or maybe it's just the experience of building a company that is the reward. I don't know.

Actually, this is exactly what it's like to bootstrap a company these days. You scrape along with no success and no money for a long time before it grows into something that can support you. Hitting the $1000/founder/month profit mark is a huge milestone since it means you can just barely squeak by without digging further into your personal savings. It means that if you can keep growing the business, you're going to win.

You'll notice that it's a completely different game than starting with big VC. And as you mentioned, it doesn't fit a lot of people's definition of a "business".

That's fine. Just remember that there is in fact another definition. Revenue Generation is still part of it, but it's not the end goal. Generating enough revenue to live comfortably on while minimizing the amount of time and effort you need to expend is the ultimate goal for a lot of us. In that context, a company that brings in $20k/month and runs on autopilot is the highest form of success. For a lot of people with different definitions, that would be classed as complete failure.

Anyway, enough talk. I'm off to Scotland for a week or so to get some fresh air while my business runs itself without me.

Thanks for the comments - this was my first HN post. I hope you didn't see it as being pessimistic - As someone who has started companies in the past and wants to do it again, I am trying to understand this new movement.

I love the thought of building a self-running business. Is that realistic? Is that realistic statistically? $12,000 profit per month after taxes is not livable in the US, and probably not Canada. Continuing to grow past $1000/month doesn't mean you are going to win; it could mean you win for a little bit and then lose all at once when something goes wrong, or someone else decides your idea is good and scoops it up. Maybe, maybe not. Again, I don't know. I do know that you can live frugally and without a cushion for only so long if your dream involves a family.

Pulling out of a phd program may be the right choice, but you are giving up on a tangible asset before you cross the finish line. Education isn't the be-all end all, but it still means something to a lot of people and that's a nice back up if your company doesn't work out.

I noticed you live in Thailand now which was apparently your dream. That lowers your costs significantly. Congratulations on that and the success you've had!

For me, living in a first world country changes the calculus. I'm not promoting the "big vc" route and I don't think YC invests in companies so the founders can make $20k/month and take vacations when they want. Your final point I'll assume wasn't trying to rub my face into the fact that you have this freedom that perhaps I don't have.

Again, congratulations on your success, and I think I'm going to try out your web meeting product.


$12,000/month after taxes is $144,000/year for a single founder company. U.S. median family income was $45,018/year before taxes in 2003. Even for programmers working in Silicon Valley, $144,000/year after taxes is pretty decent.

For comparison purposes, if you sold your company for $3.6 million, invested the money at 4% above inflation, and lived off the real interest, you'd earn $144,000/year.

Now, that's clearly not as nice as an $8.5 billion acquisition by Microsoft, but it's still pretty sweet. If nothing else, you can comfortably feed your family while working on a new startup. :-)

yes - typo. $1,000/month or 12k/year is not livable in the US.

Thanks for pointing this out. I actually did reread the post, but pre-coffee.

$1,000/month is great beer money. How much work goes into day to day operations?

One of the things to keep in mind that if something 'only' makes $1,000/month but only takes two or three days out of your time a month its "effectively" paying you $15K - $20K/month because you have those other days to do something else. People who set up linkbait pages (I'm looking at you content farms) which bring in 4x their operating cost just replicate the crap out of them.

1k/month before taxes :( If you are already doing a full time job in software dev, you are probably paying 30-40% tax already.

Just for the record, I was not one of the people who down-voted your comment. I personally think that your perspective is interesting.

"$12,000 profit per month after taxes is not livable in the US, and probably not Canada."

It seems to me that $12,000 per month is an eminently livable wage. Did you mean to write $12,000 per year there instead of $12,000 per month?

(Just a quick HN related tip. Something that helps me is to review a post carefully after I've posted it. It has helped me to catch myself saying some strange things a few times.)

In reference to your original post, Octopart is currently getting half a million unique visitors per month - and still growing. Given that they have low overhead, it is reasonable to believe that they are probably profitable at this point (including hefty owner's salaries).

I am currently in the process of bootstrapping a company in a similar way to the process that they took. I am not doing it because I was "sold someone else's dream of \"starting a company is the best thing in the world to do\"".

I saw a potential market opportunity (and something that I would enjoy working on) and I took action to meet what I saw as a need. Some people are just pre-programmed to want to own their own business. I know that it is something that I have been trying to do for a little over a decade.

I've had a few rough starts and a few no starts, but then that is how the learning process often goes. I realize that I would probably have been further ahead at this point financially if I had just taken a "normal job", but how much would that financial "security" have been worth with that constant feeling nagging me in the back of my head that I should have started my own business.

Did you mean to write $12,000 per year there instead of $12,000 per month

Looks obvious that it was just a typo - the parent mentioned $1000 per month per founder, $12000 would obviously be per year.

Guys, don't downvote self professed newbies without explaining why you are doing so.

I have to say, the kind of business you mention (20k/month & autopilot) would likely be my dream 'job'.

He said runs on auto pilot.

20k/month revenue or profit?

There's nowhere I say that "starting a company is the best thing in the world to do." I describe it as an efficient but intrinsically painful way to solve the money problem.

"Imagine the stress of working for the Post Office for fifty years. In a startup you compress all this stress into three or four years. You do tend to get a certain bulk discount if you buy the economy-size pain, but you can't evade the fundamental conservation law."

And far from trying to draw people out of school, we actually have a bias in the other direction. We're very reluctant to fund students unless they seem like they already want to quit.

"Imagine the stress of working for the Post Office for fifty years."

Maybe it's a matter of how pain is perceived. I think I'd rather do 10 5-year stints at startups than endure that sort of pain. Nothing against post office personnel, but I've been in government-like operations before and much prefer startups or earlier stage companies I've worked for.

I haven't read anywhere that you have said that, but would you agree that starting a company was the best thing in the world for you to do?

My network of company founders is much smaller than yours but I think if I were to ask them if starting their company was one of the best things they ever did, and I know it to be true for myself, they most certainly wouldn't say no.

I'm guessing that's also probably true of the majority, if not all, of the Y-Combinator backed founders. Whether their company is still alive or not.

To answer the parent, starting and running a company through the fight in the trenches is massively rewarding but also massively frustrating in many ways. I know I can't give it up and do anything else and am sure I'm not alone. I don't think anyone would try to sell you on the dream of diving into the fight. It's something you have to want, it can't be sold.

Regarding the talent pool: HN is just the squeaky wheel and doesn't represent the entire market so I would doubt that hackers starting companies is negatively affecting the available resources.

You mean as more than a way of making money? It depends on the type of person you are, and what your ambitions are. If you're (a) the right sort of person to start your own company, and (b) your ambitions in life would be best realized in the form of a company, then yes. So e.g. Steve Jobs clearly has (a) and as far as I can tell has (b), so for him starting a company was in itself a good thing to do.

Yes, I did mean as more than a way of making money. You've stated that you did in fact start Viaweb to get rich. I know a few other founders that also started companies as an attempt to get rich (I am among them) and continue to chase more. But if I were to ask those same people, whether their companies earned them wealth or not if starting their company was one of the best things they ever did I doubt that the money (or lack thereof) would be much of a factor. Realizing other ambitions (b) is a much, much greater factor.

I'm curious now though. The YC application tries to screen for (a) but doesn't test for (b). Or at least, I didn't recognize that to be the case. Why is that? Is it not just as important? I would think so as I think it's realizing those ambitions that seem to keep people going and provide them with the relentless focus on winning. Or am I wrong and you've designed the application to screen for that as well?

Why a founder will continue to push seems as important as the ability to continue to push.

Because (a) is enough. We're cool funding people who are doing it for the money; the startup doesn't have to be their life's work.

That's very interesting, maybe my perspective isn't quite accurate then as your strategy has clearly already been successful. My thought is that to build a large and successful company you need to develop an interest that goes well beyond the desire for money - but I haven't built a large company yet so perhaps I'm naive. Thanks for the reply, pg.

sorry - I knew I'd read an article about enticing people to leave school and associated it to YC. It was actually Peter Thiel: http://www.dawnmontgomerypresents.com/?p=2104 Now, having said that, there does seem to be a lot of pressure to go be an entrepreneur when the chances for success are still extremely small (depending on how you define success of course). success != happiness however and I agree that there are a lot of people out there that will grow immensely if they go take these chances.

I would much rather people go take a chance starting a company than go into finance - that's my personal bias.

Thank you for responding to my post and clarifying.

There are only like 4 companies that sell these products. Why would those companies bother giving Octopart a referral fee?

It's not like Octopart is going to remove their listing if they don't pay up, that would hurt them more than the companies.

What am I missing?

We're listing about 65 distributors right now. Early on we did remove listings from distributors who didn't want to pay. Usually they would call back the next day to make a deal. Sometimes it took longer. From what they tell me now, we're providing a good return on investment.

Ah cool. Good to hear. Hoping you guys continue to do well.

It's not really a business about the nuts and bolts, it's about the big orders that their customers can bring in if they decide to produce the product they're hacking on on a larger scale.

As Sam says, one of the best parts of Octopart is that it has so many distributors, especially the ones you've never heard of. So when Digikey and Mouser are sold out of what you need, you can find it somewhere else.

Octopart was the first company that came to mind when I thought about writing Startups Open Sourced. Having not heard about them for a while, I wanted to know what they were up to and writing a book about it was a good excuse to talk to the founders firsthand.

Here's an interesting part about their decision to drop out of the PhD program at UC Berkeley to do Octopart:

Q: When did you decide to drop out?

We wrote the first line of Octopart code during my third year in grad school. We launched a prototype as soon as we had something ready. The prototype was awful. The logo was an octopus drawn in MS Paint and the prototype took 20 or 30 seconds to return results; we launched as soon as we had something ready, and by launch, I mean we just emailed friends and family and told them, "Hey we're working on this electronic parts search engine, here is a link." We basically did that to motivate ourselves. We knew that as soon as we got it out there and told everybody "We're working on this company," we would actually have to have something to show for ourselves. We figured, with the prototype, the more embarrassing the better—it was an incentive to make the product better.

Around that time, Y Combinator was organizing a funding cycle so we decided to apply. Sam came out to Berkeley and we filled out the application. We didn't think Octopart was very good so we weren't planning on getting Y Combinator support. We didn't even think we would get invited for an interview. But we got invited and we went out to Boston. We met Paul, Jessica, Trevor and Robert and, that night, Paul offered to fund us.

Sam made the decision to drop out of grad school before we even applied to Y Combinator. His last day at CU Boulder was actually the day before the Y Combinator interview. I was closer to graduating and I really wanted to finish my degree so I was planning on doing both Octopart and grad school.

After we got YC funding, I took off for Antarctica to deploy the calibration device for my experiment. While I was there I worked on Octopart at night and when I got back I kept doing both Octopart and grad school. But by the end I was basically going into the lab and working on Octopart, almost exclusively. It took me a while to admit that I wasn't actually doing grad school anymore. I was just doing Octopart, except that I was working on it while sitting in my windowless grad student office.

Q: Was it your involvement with YC that finally convinced you to drop out of grad school?

At that time, Paul and Jessica would take all of the YC founders out to dinner. We met up with them at Jessica's favorite restaurant in Berkeley. We had a really nice conversation and towards the end of dinner, Paul asked me point blank if I was committed to Octopart and, I don't know, the question caught me off guard.

I told Paul that I was fully committed and he responded, "Well, as long as you're still in grad school, investors aren't going to invest in your company." I had been working on Octopart for six months already and mentally I had already gone all-in on the company. Before Paul's question I hadn't thought about how my decision to stay in grad school would cause somebody to think that I wasn't entirely committed.

The dinner with Paul and Jessica was on a Friday, I remember, and I said to myself, "Okay, this weekend, I'm going to figure out what I want to do; it's going to have to be one or the other." On Monday I went in to the lab and told my advisor that I had to drop out.

He worked on Octopart in Antarctica? That is pretty awesome.

In grad school I worked on IceCube which is a neutrino telescope at the South Pole. The detector is an array of photomultiplier tubes embedded in the ice underneath the Pole. At the time I was mostly working on electronic devices for calibration and measuring ash layers in the ice.

When we got accepted to YC I thought I was going to finish my PhD and work on the startup at the same time (ha!) so I kept working on the experiment. I went down to Pole for a month before YC to deploy the devices I was working on. While I was there I also worked on Octopart around the clock. Sam was actually in Argentina at the time so we would both log into our server in Berkeley to ychat when the satellite link was up. Somehow the timing worked out really well and I flew back from the Pole just in time for the first YC dinner.

Not the entire time. Andres was installing a sensor a mile beneath the water surface (if my memory is right; writing this from a mobile phone without the book on me). That was part of his research, I believe most of his time was spent in his grad student office.

Hah, I remember using all the way back to 2007 looking for special types of terminals from Keystone Electronics that were ridiculously hard to find because they were for some ancient 3/16" thick phenolic. Newark ended up having them, but Digikey and Mouser didn't. (I'm also a physics grad)

I'm kinda surprised this got YC funding.

I meant it's a good idea, but ultimately it's an affiliate marketing play right?

The main barrier to entry is getting good SEO (and I guess becoming known as a direct source) - there doesn't seem to be huge technical barriers.

Affiliate plays can be highly profitable, but I never thought that they were all that attractive to investors.

It's obvious in retrospect, but YC saw that Sam and Andres were animals[1]. (IIRC, Sam and Andres were the founders and brought on Harish just a bit after/during YC.) Octopart just would. not. die.

Personally, I think the current fad of pivoting and pivoting is probably makes for happier/better founders (my God, the trough of sorrow - I've been there and I wish that upon no one), but the Octoparts had a vision and just stubbornly stuck to it. A thousand kudos to them; they're stronger than I.

[1] http://www.paulgraham.com/start.html

Their search system is unequaled in any of the electronics parts sites I've used (Mouser, Digikey). That probably qualifies as a technical barrier.

I'm not an electronics expert, but I know a fair bit about search implementations.

I'd be guessing the partname->vendor price matching thing is non-trivial (because I bet the vendors all use different names and product codes). But it's not that hard.

The search itself is something that could be replicated fairly easily by someone who knew what they are doing.

The tech doesn't seem that complex, but successful businesses aren't about ideas, or technology. They are about applying technology to ideas and getting people to use it. The hard part to replicate of what octopart did is the 'putting all the pieces together' part.

Well yeah.

But "we put the pieces together" isn't a defensible advantage. Once one company has done it others can easily copy it (with less skill) unless they have something that makes it hard to copy.

Here's some common defensible advantages:

1) Technical. Google developed a technically better search algorithm than their competition and stayed better long enough to become dominant in the market. (That market share in turn allowed them to build a massively profitable business).

As discussed, I don't think Octopart have a technical advantage.

2) Network effects. Facebook had better network effects (and more reliable technology) than their competition.

Octopart doesn't seem to use network effects anywhere.

3) Business relationships. Microsoft leveraged its existing business relationships with computer manufactures in operating systems to build market share in web browsers by stopping them installing Netscape

Octopart doesn't have enough market power to do this.

4) Capital. Netflix offered movie studios enough money to make streaming attractive, which other companies had never been able to deliver.

Octopart doesn't have a capital advantage

There are other defensible advantages, too, but I'd be interested in knowing which applied here. What's to stop me hiring a bunch of cheap coders, copying Octopart and then beginning a race-to-the-bottom in terms of profits?

PG has replied to my initial comment and said that they have broader long-term plans. I think that's good, because I stand by my initial comment: I'm surprised this was attractive to investors.

This is key. "Putting all the pieces together" is usually less glamourous than coding something new and hoping it gets picked up on TechCrunch, but it's huge.

That's what it is initially, but they're not done yet.

It's amazing that a growing company from 2007 is considered "vintage". Just goes to show you that the timescales when it comes to web startups is highly compressed.

I remember spending hours and hours on DigiKey and Mouser looking for the right components for the prototype we were building. It's a huge pain. Especially the low value resistors and capacitors. We later found out that we should have fist checked if the parts were available before we did the design (the design itself was from reference design provided by manufacturer's, outdated). I wished Google would get into Electronics Part searching. But then it's too niche an area for Google.

Congratulations to the team. It is always worth celebrating when perseverance triumphs.

I was somewhat critical of Octopart when TechCrunch first wrote about the company years ago. My opinion was that there were already a number of similar sites that boasted the same features but (at the time) had MORE distributor listings. This continues to be the case.

In my opinion, if Octopart is interested in massively increasing usage then they should find a way to market to the enterprise side of things and not just hobbyists. Most all of the buyers I speak to on a daily basis (and the buyers employeed at our semiconductor distribution company) use OEMSTrade or Find Chips dozens (if not hundreds) of times every day. Furthermore, there are still a number of distributors that are not published by these two sites (Samtec, for example) that are increasing their footprint in the CM community. One place for differentiation - and a chance to "win" the sector - can come from adding some of these lesser known yet very important distributors to the network.

Regardless, congratulations on the hard-earned success. Best of luck in the future.

Here's one of their pages:


Interesting that their look and feel is pretty spot-on Google Apps. Probably a wise move, I'm sure by now customers grant a bit more trust to sites that 'feel' like Google, whether they realize it or not.

Glad to see a little startup like this making it though, good for them for sticking to their vision.

Hanging in there is key. I have found that sometimes you have to dig deep and hold on to the good idea that with good work everything will work given enough time. Something I have reflected on after three months at http://infostripe.com

I've tried Octopart but always go back to http://www.findchips.com

In any case, congrats to the Octopart guys.

I'd love to get feedback as to why you prefer findchips. You can email me at sam@octopart.com.

What's Octopart's backup strategy if Google somehow punishes their site in their next algorithm change (ala so many other innocent sites during the Panda Update)? I bet 80% of the traffic comes from search engines. If those 80% leave, it's game over in terms of revenue and partnerships.

I think Octopart should somehow develop a strategy for returning visitors that doesn't depend on search engines.

Registration is open for Startup School 2019. Classes start July 22nd.

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