Soon after that I sent Andres the link, and we started throwing around ideas for startup companies. In the spring of 2006 I got a phone call from Andres. He was having trouble finding a low temperature capacitor for his experiment and suggested that we build a database of electronic parts and make it easily searchable on the web.
I love seeing successful YC startups that are outside of the fairly insular "startups serving startups and other stuff startup people are interested in" space.
There's so many verticals out there that have passionate users that are being under-served by today's technology. It might be less sexy from a TechCrunch perspective, but I have the sense that's where a good amount of the lucrative business opportunities are going to be for this next era.
We don't need "a better Twitter", "the next Google", or to be "as big as Facebook". But we do need to bring the power of these products to more and more people and markets.
The problem is finding them - that is the opportunities to serve relatively niche markets. Unless you are talking to and involved in such an ecosystem you will have no idea it exists or how to appropriately address the niche market's burning problems.
But I agree - from my experience these are absolutely the best avenue for a return on investment. On the flip side, 9 times out of 10, you won't need any outside investment to get the product off the ground, all you have to do is post on a few specific forums and you will have a business overnight.
Agreed, the question is how to find such niches. Though often it's deep, expert knowledge that provides insight, I highly value the idea of the "beginner's mind". An fresh perspective on a new field (or re-obtaining freshness on a familiar one) seems to often suggest new opportunities.
From a lifestyle perspective, having a wide range of interests and making friends in different fields might help...
Lately, the internet has been calling that 'selling shovels' in reference to the gold rush days. Arguably, the guys selling tools to the miners were much better off than the miners themselves. Levi Strauss did pretty good business selling blue jeans.
Is this mostly a social issue , i.e. sexiness of techcrunch like startups ?
Or is it an information problem, i.e. because it's hard for software engineers to come across worthy problems from other domains , that they have the ability to solve ?
A bit of both. When everything hyped around you is the next "social thing" or Google, or Web x.0 app, it can be hard to look beyond that. Now add in that most young developers don't have much exposure to anything beyond pure software, and it leads to 80% chasing the same 5% of application domains.
This is arguably one of the advantages of experience: you're more likely to have been exposed to a wider set of problems, many of which are not being addressed (well) by software.
It's not at all a difficult box to break out of, but the easy part is finding an application domain that you can build stuff for. The harder part -- without having experience in that field -- is figuring out how to reach your prospects. Many of these areas have little to no web presence. It's easy to say "just look at a few forums" but that doesn't help if your customer base isn't online at all.
As a Computer Engineering student, Octopart is defiantly one of nicest sites for buying parts or just looking for data sheets. I actually wasn't aware it was a YC company, but it makes sense!
This biz idea reminds me of Hipmunk. Even though I know Hipmunk is more or less just an affiliate referral service for Orbitz, the UX is enough "value add" to win my business.
I love Octopart and wish it would become even more wildly successful. However...
The problem I have with Octopart and other "better search" engines (like Hipmunk, Padmapper, et al) is that the coverage is not 100%--not every item in inventory is shown on the search engine site.
The lack of complete coverage kills me when I'm looking for a very specific edge case, which are exactly the sorts of cases that would most benefit from a better search experience. When I really want that one 10uF/6.3V X7R ceramic capacitor made by TDK, and I can't find it on Octopart, it renders the whole of Octopart useless.
Until these sorts of frontend sites get better adoption from the people who actually supply the inventory (e.g. Octopart -> Digikey, Hipmunk -> Expedia, Padmapper -> Craigslist), they cease to be useful in all the cases where they would provide major benefit.
You raise a valid point. While we don't have every part, we're constantly increasing the number of distributors and manufacturers who send us data. We have about 60 distributors on the site now which is about double what we had this time last year. We're always going to be getting better.
As for the part you mentioned, I didn't find a capacitor with those specs from TDK, but there are 12 alternatives here: http://bit.ly/hWnqhP
If you know of a supplier of the TDK part, let me know and I'll contact them.
That was just an example I pulled from nowhere, not something specific. I don't think TDK makes a 10uF/6.3V X7R cap (although I did notice the Octopart interface doesn't show temp spec/dielectric material as one of the specs that I can filter on... that's disappointing).
My point is just that in the course of normal use I want each one of these searches, be it for a capacitor or an apartment or a flight, to be very thorough. If I can't guarantee to myself that the search will cover 100% of the available inventory. I can guarantee that if I skip the nice frontend and go directly to the supplier.
FWIW, I picked the wrong one to pick on--Octopart has way better coverage than Hipmunk, Padmapper, or a multitude of other sites.
Hey, Eric from PadMapper here. Thanks for the callout :-) I agree that if you're looking for something very specific, it sucks to not be able to rely on a site showing you everything that's available.
Mind sending email to eric@padmapper.com about any obvious omissions?
Edit: you can also uncheck the "accurately located" filter and it should show more of the missing Craigslist posts.
The biggest issue I've noticed is when listings use a local reference for the location instead of a street address or intersection: for example, a Cambridge, MA listing might refer to "Kendall", a Portland, OR listing might refer to "alphabet district".
I do think Padmapper is great, though. It's still a huge improvement over the Craigslist interface.
Ahh, gotcha. Yeah, that kind of thing is hard to do reliably, though those should show up when one removes the accuracy filter (not necessarily in the right place).
And thanks! I really do like hearing criticisms, though, since it helps me figure out where to focus next.
I disagree. I use Octopart regularly. If something's hard to find, at least Octopart tells me where it's NOT at. One query, and I know 1/2 a dozen of the usual suspects that I don't need to bother searching. That's a big win for me. Now that they have Digikey, that's also a big win.
I have no idea where you're finding parts at, that they don't have in their database, besides the sleezy feeling parts brokers that tell you they have everything, even if they don't, so you'll contact them and they can try to locate the part for you at $5.00 per IC for a $0.35 part. I have no interest in these people. If I have to resort to them, than I won't be able to find production quantities anyway.
I was wondering what happened to these guys - I was watching a YouTube video this weekend of PG speaking at a conference circa 2008 and he spend a good deal of time talking about Octopart and their business model...
Octopart if you want to start making money, build in some BOM management features. It's not a sexy area but BOM management is a pretty good problem to solve with designers, suppliers, CMs, etc. Don't forget the EOCs and MCOs too.
Its funny how often successful products and services are just obvious modifications of pre-existing ones. Correct me if Im wrong but this sounds like the Google of electronic parts.
Thats not to downplay the genius or ingenuity of these companies by any means--in fact, if anything, its a compliment to their observational skills. Still, funny how it tends to be the obvious ideas which prove the most successful. Bravo
Its described as a consolidated search engine so i dont think its fair to say that its "nothing" like Google. Different problems though, you're absolutely right. I suppose I was generalizing but I think you see my point.
In terms of the product itself, "Kayak of electronic parts" might be a better description.
In terms of technology, the first thing we did was actually take a Google-like approach to the part search problem. For example, before we started working on Octopart, you had to use separate search boxes to search part numbers and descriptions. In order to implement search across multiple fields simultaneously we abandoned mysql in favor of a full-text search engine. So, there's probably more overlap with Google search than you would expect looking at our search results.
I love seeing successful YC startups that are outside of the fairly insular "startups serving startups and other stuff startup people are interested in" space.
There's so many verticals out there that have passionate users that are being under-served by today's technology. It might be less sexy from a TechCrunch perspective, but I have the sense that's where a good amount of the lucrative business opportunities are going to be for this next era.
We don't need "a better Twitter", "the next Google", or to be "as big as Facebook". But we do need to bring the power of these products to more and more people and markets.