Thoughts on a failed launch
blog.styleguidance.com
blog.styleguidance.com
The sad fact of the matter is that us tech bloggers get way too much Email. No matter how many Gmail filters we set up, it's a constant struggle to find those 'diamonds in the rough' — the sites that nobody has ever heard of and have no funding. If you get ignored by everyone for multiple launches then maybe something's wrong, but don't get too dejected because of one bad press push.
Shoot me an Email next time you guys launch a cool new feature (it's in my profile). Tell me why it's cool. Tell me why you're more than a Digg for style questions. No promises about a post, but I'll check it out. Also, just from a five second glance at the site, I think you might want to look into changing up the layout/colors a bit. I'm no designer though.
I'm not really down, we are doing pretty well, all things considered. And I'll make sure to shoot you an email the next time something interesting happens(new feature or a milestone)
I know what you mean about the layout, to tell the truth we had a very nice web 2.0 look when we entered private beta, and the early users absolutely hated it. So we kept throwing redesigns at them, until they settled on the current one.
We'll probably try switching to a web 2.0 look, 5-6 months from now, to make it look a little bit more modern. And by then we should be able to get at least some decent numbers coming in.
I seems like StackOverflow for fashion newbies more than Digg. If it's largely occupied by the hapless, you might want to elevate a few users (beyond the concept of points, etc) to some sort of style-leader designation. And maybe add Gravatars - it really lacked personalities and imagery.
After reading the piece, however, my interest was in seeing the pitch that went out to blogs. How easy was it for them to give you coverage? Did you provide stats, incentives, quotes, ready-to-go screen caps, etc?
Gravatars already exist, you just have to enter the actual question to see them. No gravatars on the homepage, so that it loads faster.
And I'll make another post about the pitch email and the press page tomorrow.
First off thanks for taking the time to share it with us. I read the whole post. Good to see your incremental progress. Which brings me to the header of my comment. I don't see in any way how your launch was a "failure". You have "fail" and "sucked" peppered throughout your post. Why!
I wasn’t too surprised or worried by the results, since that was more or less what I expected.
2 problems here: Why expect failure? And secondly, can't call it a waste of time then since you knew beforehand what you were getting into - your choice.
Not trying to shoot you down more - quite the opposite! I don't see how your launch was a failure at all. You have more data than 100% of people with "ideas" and 100% of people in "stealth mode". And how is you gaining traffic an epic fail? Even if its one user! I see a positive slope in that graph damnit!
Personally I have erased the concept of "failing" from existence. Simply because if all things are trials, then you benefit from data received back. It doesn't matter whether that data is "good" or "bad", data is data. And one does trials so one can get better at doing those trials! So you fail when you die, till then don't recognize its existence.
Be happy man.
My graphs have a scale of 1-10 (yes im serious). so when i start to see the 1-100 (like yours) I will be ass-happy.
=) Congratulations!
But yeah overall it's been a very good month.
Thanks :)
I agree on the failure part, I think the key concept to take away, is that this is a marathon, not a sprint. Don't expect instant gratification
Pegging your product on one massive launch is a flawed idea - it seems to operate under the assumption that massive, immediate reach is what you need to attract users as opposed to leveraging/exploiting other mechanisms. Having product "out" at all stages in order to gain feedback and iterate helps.
In fact, it is far better that you had a "failed" launch now so that users can give you feedback as opposed to trying to remain stealth to build up to a bigger launch.
I think that Skribit's "successful" launch can be more attributed to how Paul handled the pro upgrade screw up. But the TC article isn't what will make or break the success of Skribit.
To tell the truth, we really haven't gotten that much feedback after we came out of private beta. In private beta we got hundreds of tips, feature requests etc, and now in public beta, nada. Not a single feature request last week.
Yeah an article on TC won't make or break a startup, but it sure does help. a) better response to your other coverage requests b) a small bump to your page rank c) a flood of potential new users to help jump start the site.
If I had to give it a numerical value, I'd say major launch coverage, can help you save 1-2 years of building your traffic the old fashioned way...1 person at a time
I want you to succeed and reach your potential, and I don't like saying things like this, but .. you count your uniques in 1 digits?
Dude. I have users who do nothing but post every few weeks about how much they hate their teacher and they are kicking your ass by orders of magnitude. I had kind of come to believe 100 or so hits a day, uniques that is, was the new baseline for any kind of non-copypasta content. Even 10k hits a day is sort of ... meh? Great proof of concept if the trend is heading upwards but useless in and of itself?
What I'm trying to say here my friend is .. you can do a lot, LOT better than <10 uniques. Seriously, any blog at all can do better than that. You absolutely need to start a company blog ASAP and just write about ANYTHING. Mention anything you learnt in any aspect of development, you will have 100 uniques/day in a week and maybe more.
I guess what I'm trying to say is .. I like your attitude but geeze, you need to recalibrate your scale a bit. Sub 10 uniques a day is pitifully low and almost below just random; I'd almost advise you to check your DNS etc. Any kind of useful company blog whatsoever will put you orders of magnitude about that. Yes, erase the concept of "failing" from existence. But don't erase "improvement" as well....
The problem is very simply that I am a programmer. I spend time programming; not marketing.
I don't have an (active) twitter, facebook, myspace, linkedin, blog, etc. Yes yes I know how important it is to have them! - but the point is I don't.
So I always just build out things because I'm a semi-noob programmer. I've learned from experience, that its all about customer development, so that's what I will do now. But it is indeed factual that I have about 8 visits a day =).
Lastly, I am a 37 signals guy. I measure (company) success in profit. So it is nice to measure the importance of how many uniques you get, only so much as it contributes to the bottom line. I.e. A blog might need 50trillion uniques to make $x. A paid app, eh probably not so much.
One other thing: For a site about style, the site is not very stylish itself. I think for your site, a designer or a few CSS books (SitePoint has some good ones) would be a good investment.
Again, thanks for the writeup. Looks like it has a lot of growth potential.
Edit: maybe it didn't render properly on my iPhone on the first visit. Upon checking it again, I can see the design. It was just white text on a plain black body the first time around. Looks better now.
I know what you mean about the design, a redesign is definitely in the plans. Just didn't want to spend that extra time redoing the design, instead of launching.
That said, I like the idea -- one of the better StackExchange verticals I've seen. But to beat other similar sites -- perhaps even sibling StackExchange sites -- your initial community is all-important. The only successful 'launch' is one that brings the right kind of askers/answerers... and I suspect most of those people aren't rushing to check out 'launched' products; they instead stumble onto new sites by a variety of paths generally unaffected by 'launch' publicity.
And to echo other commenters: the design/color-scheme doesn't say 'stylish' to me, and I suspect for your audience, that could be important. (The dark purple makes me think more of a sports site, and the manila-yellow makes everything look like a field-needing-input.)
Can you follow the design cues of some emblematic sites in the fashion/beauty category? Does StackExchange platform provide any way to A/B test different looks?
Good luck!
As far as actual competitor sites, most of them are generic, and we have much higher quality controls for posts. On Yahoo Answers, "What does the color red look like?" is a valid question, and on ours that gets deleted.
For design cues, I have a whole list of sites that I'll be using for inspiration.
No the SE platform doesn't let you A/B test anything. It's really very basic at this point. Pretty much all you can do is edit a CSS file, and add stuff to a box just above header, footer, 3 spots for ads.
So pretty much if you want to do anything, you need to add a ton of javascript and css.
I don't know whether it's a problem with the stack-exchange sites in general, but I have yet to see a single one (other than the mother ship) that does not look like utter crap.
Maybe it's the fact that we are going for a non-techie audience, but they seem to like the "crappy" design. But yeah I agree with you, it's pretty bad, which is why we'll be redoing it a few months down the road.
Pretty much, just wanted to get it over with and launch, since if you keep perfecting your site, you'll never think it's good enough to launch. I'd rather launch and have a 20% churn rate because of a bad design than spend an extra 3 months in development. Waiting to make it perfect.
You can always change your design later, but you'll never get those 3 months back.
P.S. You think SO looks good?
A massive-to-us 34% of new user signups that day were through Twitter, thanks to a riduculously low one-click barrier of entry.
As for contacting bloggers, we only got in touch with TC for launch day and have yet to reach out to my other blogger friends, et cetera. Gonna wait til after the holidays.
That being said - you are definitely getting some traction with the 700+ uniques/day. Grats!
I think the lack of registered users is mainly because you don't need to register to ask/answer questions. The site is 99% useful w/o registering. The only feature they don't have is voting.
So pretty much the only people who end up registering, are the super active users, who want to dive in, and participate.
We aren't really focusing on registering users, the goal is to make it easy to create new content, more content = more ways for Google to find our site.
Basically at this stage it's a SEO play, the goal is to become THE answer site for fashion/style questions.
There are plenty of big beauty/style blogs out there who might have covered you :)
I mean, great, but this is just another Stack Exchange instance .. how long before that software is considered as basically a standard tool like WordPress? If it isn't already? Do we talk about some lightly-customised Wordpress blog "failing to launch"? Hell at least that might have original content.
Making available an instance of server software you didn't write, with some minimal twist, != startup, IMO.
And hey, I had a phone interview with Venture Beat...when I launched my "non-startup"...which is more than some "startups" can say.
Techcrunch is just a wordpress blog and it gets 7.5 million uniques. WikiHow is just a mediawiki instance and it gets 20 million uniques. You can't focus too much on a platform. You think those guys couldn't build their own product from scratch? Sure they could, but then they'd have wasted 1-2 years in development time, when they didn't have to, and that's just stupid.
I'm sure there are plenty of other examples of turn key solutions being used to make a business. Meanwhile your cool new startup with 3 years of coding, 100,000 lines of code, gets 50 hits a day.
Sure I could have went out and coded up my own Q&A site, but why do that when I can just use a turn key solution?
Plus there are already >4,000 lines of customization code in the site, which is probably more than some people have in their "unique" web apps, who you'd label as a startup.
Being a startup is a state of mind, you can be a "startup" with a hello world application, as long as you feel you are. It doesn't matter how you get there, all that matters, is that at the end of the day you are working towards building something positive.
And we are getting plenty of traffic to justify that we are doing something people need. Hence the startup label.
That's precisely what they care about :)
BTW those examples you give built up content and traffic over a long period of time. They were innovative in their areas - but not with their technology, with their content and their focus.
(Im not commenting on you being a startup or not btw)
Same goes double for me. Sure I could have coded the thing up myself. Wasted a year, two, building the product, made 2-3 extra features. And for what? To get a "good job, but it's just another answers site" and a pat on the back? Screw that, I'd rather use those 2 years to build up my content and traffic.
Content is all that matters anyways. Content drives traffic, content gets users to stick around. Who wants to waste 2 years building a site, only to have to start out with 50 users a day traffic numbers?
What you have isn't something with cool features, your right it is content. That kind of site will take a year or more to build up. But it is the same process your pouring scorn on :-) there is still no guarantee.
But yeah it'll probably take 1-2 years to build up the content and traffic. But you gotta start somewhere, 10 people here, 10 people there, and before long, you are hitting millions of people a month.