Heads up, I clicked on the "claim this page" link for BedquiltDB (my sideproject) at http://stackshare.io/bedquiltdb, and got a generic server error page.
FWIW, I went to stackshare today and was interested in the stackups feature. For the first time ever, I decided to click that I used something (bootstrap), then I ended up registering through github and submitting my site.
However, once I was logged in, I could not find a way to get back to stackups. The top/left home icon takes me to trending tools and stacks. Posts seems to be stacks of major companies. Tools and services doesn't have stackups.
I searched for searched for "Stackups" in your search bar, and the top result was "Stackups". I clicked more. It lists 50 results, but then no actual results were listed.
I was only able to get back to the stackups by logging out, and returning to the home page. Maybe I am missing something obvious, but just FYI that was my experience.
I'm not sure I would characterize this as boring stuff so much as only like a third of the picture.
When the article was titled "tech stacks" I was hoping to learn about the actual tech stacks, not just the front end stuff. Like what's actually answering the HTTPS requests? nginx? Apache? Node? Passenger? Tomcat?
What's the DB system?
What OS?
Heck even from front end we're missing who's using Angular, who's using Ember, who's using Bootstrap, etc.
I wouldn't have titled this "Tech Stacks". Maybe "marketing and analytics software."
FWIW their Chrome extension (free to try) does seem to detect Bootstrap consistently, but front end frameworks were hit or miss depending on the site for me.
Still the amount it does uncover quickly is impressive.
It seems like their tool gathers data from commonly exposed endpoints, etc. Doable for some backend web frameworks, web servers, etc, but much harder for the tech stack on the backend. To me that would be more interesting to have too.
StackShare [0] is working on that problem. It's all self-reported for now. It's pretty cool to be able to see the full stack of HyperDev [1] (exceptionally good post) or GitHub [2] for instance (average post).
I looked up the Datanyze report for my own startup and most of the information is outdated by 6–12 months and/or incomplete... I'm curious where they sourced it from. CrunchBase looks like one source.
"So besides the boring stuff listed in this article"
Not boring if you want to target YC batch companies. In fact it is quite the opposite. The niches here are hacker-rich and as such could be used as a guide to build new products. That is the reason I posted it.
"No self-respecting developer would ever include the items enumerated in this list as part of their "tech stack.""
There is a dis-joint here between the reality what ^is^ and what ^should be^. Here's your chance. Write a better list and submit it. Show us how it is done. Or find a newer, better article and submit it.
>2. Of the YC companies that use hosting providers, 55% use AWS, 13% use Cloudflare and 6% use Rackspace.
Comparing Cloudflare and AWS? Where's GCP or Azure?
>3. Of the YC companies that use a CDN, 68% use Amazon CloudFront, 7% use Cloudflare and 5% use Fastly.
Okay, so there's a separate section for CDN...
>8. Only 22% of YC sites use a third party CMS. Of those, 83% use WordPress.
"3rd party" CMS? As opposed to...a custom made web app?
>11. In terms of installs, the most common Analytics providers are Google Analytics (451), Google Universal Analytics (314), Mixpanel (183) and Optimizely (129).
I thought Google Universal Analytics was an updated version of Google Analytics.
That said, this kind of market research is incredibly valuable.
Hey, so i'll attempt to clear the air a bit. Didn't think this would trend like it did..
Cloudflare - chalk this up as a mistake. I have a marketing background, so not too well versed in the infrastructure space. I'll ask Macro to make an edit for me :)
"As opposed to...a custom made web app?" - correct.
Google Analytics vs Universal Analytics - Correct, Universal is an update, but many still use GA standard. The GA # (451) is all encompassing, the GUA # (314) is the subset of GA users that use the update.
Granted, there is much room for improvement in both the information given and the semantics behind it, but I'm glad you enjoyed it. Thought it would be interesting for people to see since a lot of this stuff usually operates behind the black curtain.
Yeah CloudFlare is not a hosting provider, so that piece seems erroneous. They are on the CDN line below, I just don't understand why they ended up on this line. Maybe because people's DNS points at CloudFlare?
That's correct, but it's very difficult to determine the hosting environment if you're using a CDN. I have a write up on my blog at https://www.gra.pe/?p=36. But the idea is that if you know the hosting IP, you now where the IP sits (ASN). If you determine the ASN, you know which provider is being used.
To add to this - what they probably did was run an nslookup or dig against the domains themselves. And whatever IP resolved for those domains, they checked who owned that IP space. With a CDN, it won't show AWS, it'll show the CDN.
83% WordPress usage among CMS users seems high, but 22% using a CMS at all seems low. Presumably most using a CMS just want a blog and a few pages? Any kind of complicated content structure goes straight to custom development? Does this account for static generators like Jekyll or CMSes that don't provide information about themselves?
I would think that many of them don't expose plainly the fact that they're powered by WordPress or only use it as a backend/authoring solution with a custom frontend.
Might be because often the standard content pages are a small part of many startup sites so they just tack something custom onto their site build rather than use a full CMS.
The chart at the bottom (now that it's fixed; thanks!), if you remove the Google outliers, shows that there is little correlation between the market share of a given SASS within its category and the number of YC startups that use it.
As a result, the conclusion is that YC companies just use whatever the hell they want, not what is "best."
Remember new YC cos get beaucoup Azure credit now. (I'm not sure how much they get on AWS.) It would be interesting to see if the offering has had an effect.
> We are happy to announce the Microsoft will be giving $500,000 of free Azure hosting credit to YC startups in our Winter 2015 batch and future batches.
Is that surprising? Early on, I wouldn't worry much about server cost. Salaries are usually the big budget item, and I'd lean toward boring/default choices for anything that isn't absolutely vital for demonstrating traction.
Then if a startup hits the growth phase, I'd expect server costs are still relatively low priority.
We tried to use Azure for a new system ~8 months ago (it's hard to pass up that much money). Spinning up a vm with the image we needed and a static ip was not possible with the GUI. The specific image we needed was only accessible on the "new" portal, but a static IP could only be assigned from the "old" portal. It's theoretically possible to set it up with PowerShell, but after spending over a week of poring over outdated and incomplete documentation and opening support tickets, we switched to AWS and had it running in a few hours.
Even if we did get it to work, I'm not sure I would feel comfortable investing in a platform where something as simple as "create a vm with this image and assign a static ip" isn't straightforward. It could be better now, but for us (and most other young startups) the time of our engineers is much more scarce than hosting costs.
I ran a similar study a while back - I was curious to see if the YC incentives for startups (Azure credits, Amazon credits, Digital Ocean credits, etc.) were significantly swaying founders to use a technology stack.
Also, I'm assuming Datanyze used the publicly facing domain of each of these websites (very similar to what I've done). What you'll notice is a lot of these startups have APIs or apps that have very different domains than the ones being used for their landing pages. I spoke to the folks at Bizspark about this and Azure adoption is pretty high (I believe these are private numbers).
Is the most popular domain registrar among YC cos really GoDaddy?
This seems pretty unusual given industry trends over the past, 5+ years. Do you separate GoDaddy resellers from GoDaddy direct?
Update: They are not separated.
> I see Namecheap come up in a lot of hackernews posts for recommended registrars. I’m assuming since Namecheap and many other popular startup-friendly registrars are both resellers of GoDaddy and eNom and are ICAAN accredited registrars, the WHOIS lookup may not reflect the proper registrar (i.e. they may show up as eNom instead of Namecheap).
I'm also surprised to see AWS Route 53 omitted as a registrar, and possibly as a name service, unless that's what "Amazon" under DNS refers to.
Still this data is useful and nice to have as a point of reference.
Yeah PG should attempt to validate his claim that a better language is actually an advantage. They should have enough data to draw some sort of conclusions. Or be able to ask ex founders if they didn't collect it beforehand. Of course all this compared against various success metrics.
Also good to know when people ended up switching stacks and for what reasons.
Reddit tried Lisp and moved to Python, quite famously. Viaweb it turns out is not extrapolatable to, well, any other website or company. The real lesson is, use whatever language you already know.
Good point. I don't know what PG considers a "better" language (aside from Lisp), but I'd make a random guess and say that most YC companies are using JavaScript, Ruby, Python, or Go for their backend.
All of those are basically Lisps if you squint hard enough. In particular, they are much closer to Lisp than the two languages people were using for Viaweb competitors at the time, namely C++ and Perl (according to the "Beating the Averages" essay), and specifically the mid-'90s version of C++ and Perl. Perl 5, released 7 months before Viaweb launched, added anonymous functions (lambdas) and the ability to have variables that were references to variables. Perl 5.004, released almost two years later, added funcrefs, which made anonymous functions useful. C++ technically let you define objects with operator() all along, but there was no useful support until the STL, first released less than a year before Viaweb launched. boost::function only showed up in 2001, and native lambdas in 2011.
The curious thing is that, even so, none of those languages are really Lisps at all. I think the lesson is not so much "Use Lisp" as "Use a programming language that actually has some basic features and which your tech folks are familiar at", and most programming languages in the '90s didn't have what we'd rightly consider basic features today.
IIRC PG's main argument was that macros were so powerful he could write stuff to write features vs writing features directly. Hence being super fast compared to competition.
Even if that was accurate, I would guess with modern web apps or UIs there's gonna be a large overhead that might wipe out any benefit. Using F# in a live app, for instance -- sure, the main parts are nicer, shorter. And yes, it makes me feel sane. But compared to the overall effort a company needs, I'm not sure it matters in most cases.
http://stackshare.io/stacks might have what you are looking for. Although not all YC companies are on there, some of the more well-known companies are.
Yeah, I'd say this is more of a service stack. If someone asks you what you do for email, it would be odd to say "my tech stack is Gmail". Is Gmail's tech stack Gmail?
Also, I take it to mean 21% use a client-side A/B testing framework.
You could be performing A/B testing server-side and it would not be detected using this approach (which ultimately boils down to greping the HTML for known strings).
I agree, that said, I think it's hard to say the 79% _don't_ A/B test. Could easily be homegrown, or round-robin based, or perhaps it was only included on a specific page.
93% of the YC companies that we found using an A/B tester were using Optimizely. This, however, does not account for server-side testing as mentioned in the thread.
I think it's because they mostly spend money on current customers not new ones. To attract new customers they probably use another ways (like SEO, cold call, and so on).
The value of AWS is all the other services that they offer. While you could assemble them yourself, building on AWS allows you to move really fast and scalable infrastructure.
Both are great for hobby projects, and offer pretty great bang for buck, but I'd never run consider running anything serious on either, let a lone a business.
Some examples:
== Not pay bills (yeah, yeah, I should check my email more often when a credit card expires) ===
* Digital Ocean: All is working, then one day they flip the switch and shut down your servers and delete the backups (that you paid for)
* AWS: Has a super reasonable policy, first by shutting down the servers so you know
* Linode: Not sure
== Get DDoSd ==
* Digital Ocean: 24 hour blackhole, zero support or human contact in this time
* AWS: Servers kept alive, got some engineering assistance
* Linode: 24 (?) hour blackhole, a bit of really bad support
== Physical Server Upgrades ==
* Digital Ocean: Not bad
* AWS: Amazing. Zero downtime, I have 4 year uptime on many servers
* Linode: Botched my servers, more times than I can count. Even had support reboot the WRONG servers of mine when I was talking with them
== Bandwidth ===
* Digital Ocean: shitty and cheap
* AWS: Amazing and stupidly expensive
* Linode: Decent and cheap
I could go on. If it was a hobby project, I'd probably just use Digital Ocean because of their cheap prices and sleek control panel. (Or actually I'd use vultr)
But for anything serious where I have a budget for servers, you'd be crazy to use anything about AWS or Google Cloud
Instead of scraping the sites and trying to guess what YC companies are using...
Perhaps just ask them?
Surely every YC company would answer, or at least enough to make the information more valuable than a set of guesses.
Additionally, it would be good to know when YC companies choose other YC company products based on incentives offered (it's free to other alumni), as this disclosure allows others to evaluate whether the choice to use a product was because it was the best product for that task or was potentially skewed by the incentive offered.
It would be interesting to have also information about the frontend. Are they using Bootstrap/Foundation/etc? What other JS libs are they using for the presentation layer?
Advertising technologies I'm almost sure it's wrong. How do you know from the javascript used if the company uses Adwords? My company uses adwords extensively and all the events and conversions are passed using Google Analytics and the Adwords API.
Amazon gives away a silly amount of free AWS credits to some accelerator programs. At 500 Startups I think they give you $50,000 or $100,000 in credits that expire after a couple years. I assume they do the same for YC. Obviously most companies can't even use more than a tiny fraction of that.
None of these numbers point to a monopoly on the bizdev/marketing side - is that due to the tools/players being new (I consolidated) or is it likely to stay this fragmented?
http://stackshare.io
Some well known startups are "verified" with a blue checkmark, which means they update their tech stack themselves so look out for that.