Hacker News new | comments | ask | show | jobs | submit login
Why Do Startups Do This? (asymptomatic.net)
276 points by ringmaster on July 26, 2012 | hide | past | web | favorite | 74 comments

> The big logo-y thing at the top of your blog page? Yes, the one that currently links to your blog? Right, that one. It shouldn't link to your damned blog! Link it to your product's home page instead.

Every time I have to manually cut the /blog/ out of the location bar I wonder how many users were lost by requiring that little bit of extra effort.

This. I got downvoted the other day for saying that about Kicksend: http://news.ycombinator.org/item?id=4159367

Maybe because it was not on topic but it does get on my nerves.

On a very similar note, if you want people to upvote you on another HN thread, you should really probably link to news.ycombinator.com instead of .org so people like me won't have to sign in again. :)

I'd argue that .org is better so people like me won't have to sign in again.

Since you are on the same site already, use a relative url.

Curious, had no idea there were (at least) two tlds.

Interestingly, it seems that they did change it to go to the homepage.

I just upvoted your other post. This practice is quite annoying. I guess it's understandable on some level but obviously it should be easy to get to a company's home page from its blog.

I make a rough estimate that 50 per cent of the time I do not bother making that extra step.

I think it's like 99% of the time for me.

I tend to think that that is just the default behavior by whatever drop-in blog package they're using. You see this in nearly 110% of phpBB boards, where the corner link brings you to the top page of the webboard.

It's a missed detail, though whether by laziness or incompetence I'm not willing to speculate.

Sometimes it's blog.domain.com. I sometimes don't even bother to cut the subdomain to see the product itself.

Same goes for newsletters / emails that I joined via a splash page or holding page. At the very top of the email tell me:

    1. You're receiving this email because you joined XXXX's beta waiting list on xx June 2012

    2. XXXX is a product that helps you do ....
Basically remind me what you do and how you got my details.

I can't tell you how many times I've signed up for a beta release, totally forgotten about it, and then get a mysterious email with absolutely no information about what the product is, does, etc.

It's so frustrating and at the same time, right before I hit the delete button I think, "Well, if they're not taking much time to clearly explain this at the top of the email or newsletter, what else are they taking shortcuts on?"

Better yet, don't launch until you've actually launched. On the list of startup behaviors that drive me crazy, right after the two in my post, comes "Sign up to be notified of when (hah, IF!) we actually do something" pages. To paraphrase Yoda: Launch or launch not, there is no sign up to be notified.

I couldn't disagree more. Pre-launch signups give you a way to measure interest, a group to communicate with about features, betas and the like, and a crowd to talk to when you do launch. otherwise, you are developing in a black hole.

Then it is important that the emails that are collected are then used for doing that. I have yet to receive anything more than a "verify your address" email based on submitting mine to any of these pre-launch email collection sites. Nobody has communicated with me about features, or tried to use my interest to help develop their brand. Perhaps if that were the case, I'd look upon these pages more favorably. But my thought in general still stands: If you don't have anything to share, you possibly should reconsider collecting emails until you do.

I agree, pre-launch sign-ups are important but there needs to be engagement with the emails collected until the product is ready. Fab did a great job with this by creating a mood board that allowed folks to post stuff they like before their actual service launched.

And what does it do for the user? "I am going to build a space ship for less than $1000, give me your email address and I will let you know when I am ready." What did that just do for you that it would not when it is actually done?

Lots of information about who and how many might buy your spaceship, helpful in decisions thought the design and manufacturing process, a group to announce product and beta launch to, leveraging months or years of long tail attention rather than just launch day, ...

Pre-release sign ups are critical for anyone competing in an app store. You need to get a lot of downloads in the first few days to break into the Top 10 list, so you gather email addresses pre-release and notify them all at once on launch day.

Lots of smart people disagree with this. You should start marketing when you start coding. Find out who is interested, what they want, etc. When you launch, you don't want to do it to crickets.

If you have something people will search for, you can start your SEO and start adwords. 6 months down the road you are actually getting some search traffic, and your adwords have over time built up a mailing list.

If you haven't read this article by Rob Walling I highly recommend it. http://www.softwarebyrob.com/2010/10/14/startup-marketing-pa...

I understand that as an user/early adopter you dislike it, but the launch page with a signup form is an invaluable tool to gather information about interest for your app for cheap, even before you invest time in the development.

Yes! I'm not eating and breathing your startup like you are. You actually do have to remind me why I cared about you in the first place.

I hate it when people decide to slap a blog section on their website and it is completely disjointed from the rest of the site.

Usually this seems to happen because they have used some customised system to build their website and then just slapped wordpress or something on to use as the blog.

So as mentioned in the article you hit the blog page and "home" now takes you back to the home of the blog even though the site looks like the rest of the site.

Even worse when it's a tumblr or something and you have now ended up in a completely seperate Silo.

I can't imagine this is good for SEO purposes either.

The number of time this sin is committed by companies selling design services boggles the mind.

When starting O2O Trade to sell HDMI & USB cables, we included a blog because it was "one of the things successful startups do". None of us knew what to put into the blog, and most entries were a measly paragraph. Even so, our blog entries showed up in Google searches more often than our static product pages. Searching for our company leads people to these little stubs of text instead of our products.

Adding a blog seems like a "good idea at the time", but unless you have a good blogger on your team, it'll just be an afterthought-task that gets shuffled around because nobody wants to do it. Don't let a blog waste your time and steal focus from the rest of your company.

sounds like you just don't know what to do with the traffic to your blog posts

This all boils down to a rule I have for myself whenever addressing a large audience: start by stating the obvious.

This does two things. First, it gives people an easy mental on-ramp to follow the thread of what you are saying. And second, it forces you to back way up and cover the ground that is so central to your world you would probably forget to say it, even though it is completely unknown to most of your audience.

Yep, people tend to assume that everyone knows what they know by default. See http://lesswrong.com/lw/kg/expecting_short_inferential_dista...

The "Tell me what your product does on every blog page" is one that Atwood is great at. The SO byline is very unobtrusive, but I would also assume informative to anyone who doesn't know what SO is.

It was so simple, and it has probably brought SO a ton of first-time viewers.

I think some people want to blog without it seeming like one big advertising. Sometimes I read articles on startup blogs and think "why are they even writing about this" until I get 2/3rds of the way through and realize I'm reading an infomercial.

Seriously, just put your product link at the top-left (like OP says) and throw the tagline right next to it. I'm not going to be offended that I'm reading a branded blog.

Wholehearted agreement with Owen's post.

On point #2: describe your product in clear, what-it-does language.

Mistakes I see are emphasizing: how it does it (C, Java, OO, Rails, REST, ...), where it does it (PC, mobile, Mac, Cloud, ...), "ecosystems" it integrates with (Social, FB, Oracle, ...), who your investors or team are (VC, founders, investors...) etc. All of which may or may not be particularly relevant, but ... they're not key to me understanding what you do. Tell me these things, but focus on the what first.

Use direct, actionable language, not vague or nebulous terms. It's a "NFS file security permissions auditor", not "Cloud information assets security tool".

Describe a workflow or workflows from the perspective of your users. Not developers. Not architects. Not

This doesn't just apply to startups. I use a lot of Free Software, and many of these projects also fail to describe themselves clearly (though most, especially over time, eventually get it right, if only because other people can come in and rewrite idiotic descriptions). Reading through a list of package descriptions from Debian or Ubuntu, where a pithy, one-line description is your shingle to the world, should give a sense of good and bad descriptions.

Even long-established technologies such as Java suffer from this.

At www.java.com we have "What is Java?": "Java allows you to play online games, chat with people around the world, calculate your mortgage interest, and view images in 3D, just to name a few. It's also integral to the intranet applications and other e-business solutions that are the foundation of corporate computing." Um. OK. open http://www.java.com/en/download/whatis_java.jsp

At Oracle, we have a Java landing page with ... no description of the technology or its components (which aren't self-evident): http://www.oracle.com/technetwork/java/index.html

One of the best succinct summaries I've seen in recent memory is from jwz's "Java Sucks" page:

there are four completely different things that go by the name "Java": 1. A [programming] language, 2. An enormous class library, 3. A virtual machine, 4. A security model. http://www.jwz.org/doc/java.html

Now that is something I can wrap my head around (he also goes on to describe strengths and weaknesses of each component, good essay, read it, it's still disappointingly relevant).

Note though: the best product description comes from a critic. If you fail to clearly define yourself, your critics will.

You said to describe the product from the perspective of the user, not the developer, but then you preferred the developer description of Java (language, library, VM) over the user description (play games, chat with people, view images in 3D). It does seem like your average person landing on the Java page cares more about the second list. I mean, what the hell is a virtual machine anyway?

I'd say there should be two different pages: one for end-users, and one for developers. In his defense I'd argue the users of Java are developers, in which case they need to know specifically what it is. For end users who need it installed on their machines, the description should be "Java is a necessary plugin to make many of your favorite applications work correctly."

In the case of other tools, a lot of companies focus on useless buzzwords like "it's a cloud-enabled multi-tier architecture et cetera" instead of focusing on what the person using it cares about: "it's a content management system for blogs." OK, got it.

Seems to me that Java's users are developers. No non-developer will ever install Java by itself, they'll only install it because something else they're installing requires it.

Which Java are we talking about?

Again, as jwz noted, it's four things:

1. A programming language.

2. A class library.

3. A virtual machine.

4. A security model.

As a systems admin, I play mostly in 3, deploying, tuning, configuring, monitoring (to the extent that piss-poor Java tools allow any sort of monitoring -- want a dump of what's in memory? Sure ... let's just pause ALL activity on the VM for the next 25 minutes), troubleshooting, and patching/updating the VM.

I'm also concerned with the security model, within the parameters of my other system security concerns.

For the language and class libraries, it's largely ensuring that what my devs need and use is provisioned on our test, staging, and production hosts. There's also digging through some of the logging / crash / debug output to see if I can sort out what's wrong and fix it myself, or punt it over the wall to Engineering.

So no, it's not just developers.

Same could be said of this blog:

Asymptomatic There must be intelligent life down here

Doesn't tell me very much about who the author is, what he typically writes about.

Was going to say the same thing - presumably your goal is that I will like the blog post and by extension you, and if so I'll be curious what you do, what your area of expertise is, etc.

I can only hope the irony isn't lost on him - maybe Owen (whoever he is) is purposefully giving us an example of how annoying it is to not explain what you do on your blog.

Yes. Irony is ironic.

While I do agree that the product link should be obvious on the blog page, I don't think it should be overly pervasive. A logo on the sidebar is more then enough. The main purpose of the blog is to write engaging content that correlates with your product. Startups should stop trying to use the blogs just to increase SEO rankings, rather focus on creating conversations and build relationships with potential customers and users.

Thank you. It's irritating when the blog link is blog.companyname.com and I have to manually replace the "blog" part with "www."

Generally I think that's a side effect of outsourcing. They point a subdomain of their domain at wp.com or whatever blog host and leave it at that. Of course the blog host is going to make it a default that the corner link go back to themselves. Traceroute to the blog and I think you'll usually see that it's a separate host than their www.

Both very good points. I've even been guilty of it here and there... the trick is in making sure that the call to action is simple, present and not being deterred by anything else.

Having friends review is great, but sitting an intelligent stranger down and asking them to perform certain actions is what most startups need.

Many a time I've posted on HN comments something to the effect of, "that's great, but I have no idea what Your product is." it's extra lethal if your company or product name doesn't describe anything at all to those not familiar with "FooBarlr".

I've run into this so often; I'm glad I'm not the only one. I often have to revert to modifying the URL just to get to their main page (change blog.company.com to company.com) after finding no possible way to link to the home page.

"If you're using software like WordPress or PHPBB (What that hell were you thinking?)"

What exactly did WordPress have to do with his point?

Using WordPress for your entire product site makes a ton of sense for most companies. You can have your blog and your product info all on the same site. No subdomain blog. No separate SEO. It's simple to link the logo to the product, and have custom sidebars or after-content widget areas with a call to action. Your blog is part of your site, and feeds traffic to your product. Everything integrated. Easy to use. What's his problem again?

I assume he's referring to PHPBB there, which would be an odd choice for blogging software as it's forum software.

Owen wants more people to use Habari instead.

Habari was started by a group of ex-WordPress developers - http://asymptomatic.net/2007/01/09/29/whats-up

Please don't misunderstand. If you want to use WordPress that's great. My "What were you thinking" comment refers to PHPBB only, which I've seen used more than once for the primary blog for a startup. I'll use more commas next time or something.

That said, Habari is also a good choice for a blog. ;)

Cool. I completely agree that using PHPBB for a startup blog would be "strange".

Good points, I haven't really noticed this all that often.

I'm actually designing a blog for a new startup now and the first elements I created were at the top of the right sidebar with a quick "what we do" summary with a call to action button at the end. Also used the author summary box at the bottom of the article with similar content. I think just a quick "We're xyz company and we do [whatever benefit you offer]" and a "Learn More" button is a quick, clean and concise way of putting that out there.

Could not agree more.

Using something like Wordpress, it's not difficult to put a little box at the bottom of every post explaining what your product is and where I can learn more.

I agree with the advice, and I offer some in return about Habari: hire someone to make your demo video non-awful. The "muzak" and slow pacing alone were enough to send me running away from the whole project. A production that aesthetically displeasing is a sure sign of an unappealing product/environment.

When you post a link to Habari at the bottom of your post, it would be nice if the screenshot IMGs weren't broken on screenshots page. That is also annoying.


Yep, totally agree. I even wrote a very similar rant here several weeks ago: http://blog.jazzychad.net/2012/05/28/startups-fix-your-blog-...

Great advice! I just changed it for our blog, we had the big logo at the top linking to the main page already, but I realized we didn't have a short description of what we do. Blogger allowed this to be easily added to the sidebar.

I posted this not too long ago: http://technicalblogging.com/5-common-blogging-mistakes-made...

I cover the same two mistakes, plus a few more.

To the OP: You link to your product, but when I click on the link, the images don't work - http://habariproject.org/en/screenshots

Similarly, a huge % of bloggers overall make it impossible to contact them via their blog. I've never understood this unless you're intentionally being evasive.

I must have read about how Airbnb was "redefining the space"and taking all sorts of problems at least half a dozen times before I had any idea what it was.

I think there should be a #3 .... startups that devote their blogs to random hn fluff about being a startup rather than whatever they actually do.

Completely agree! The extra effort to go to the actual product's website makes me wonder if the startup even knows this is an issue.

Anyone have a mirror or different URL? It seems to be down for me...

Give it another try.. it was down for me too but seems to be back now. Or try http://webcache.googleusercontent.com/search?sugexp=chrome,m...

Glad it's not just me. I tried authenticating with Twitter to leave a comment, but then the site (not Twitter) b0rked on me.

Oops. Should be working a little better now.

same here.

Amen! I've been saying the same thing: https://twitter.com/dwradcliffe/status/209703542192222208

Good point. We don't currently do this very well on the Fogbeam[1] blog[2], and I'll be making it a point to address that later this evening. Thanks for posting this and bringing this point to the forefront!

[1]: http://www.fogbeam.com

[2]: http://fogbeam.blogspot.com

I recommend excessive use of the <blink> tag around any product mentions to really hammer things home :)

I always thought it classy when someone took the time to write something meaningful and share it without it just being a blatant advertisement for their product, I am alone on this one it seems, so be it...

They can write interesting architectural or business things without it being a "blatant advertisement". Often if I read that someone is using Redis in a clever way I'm curious what they are using it for and would appreciate an easy way to get to their main site.

Applications are open for YC Summer 2019

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