I shouldn’t have been so concerned about what the people on HackerNews would say about the design.
This needs to be starred about taking all advice, ever, but particularly taking advice from people who are a) like you and b) unlike your customers. I always ask my forum buddies for opinions on new things (never turn down another set of eyes), but you have to take them with a grain of salt: at the end of the day, they're not the ones who pay you money.
A lot of folks had very strong words for my last redesign -- including at least one person who thought keeping the designer's name on it was akin to professional slander. My customers liked it, as evidenced by metrics(+).
Guess whose opinion matters?
+ Metrics: accurate, appropriate, and timely numbers which inform decisionmaking. Pick any two. Grumble grumble... but I do love them.
Great article. I think you should consider an alternative subscription model.. one time access.
I am one of the demographic that initially used the site after either a HN or Cnet link (can't remember which one). I actually purchased a domain or two and made a mental note that I will use the site next time I am in the domain market, as I found it very useful.
For someone like me, I only look at domains once every few months at most. I just don't have an active appitite for purchasing - but when I do, it is a big priority. If you offered limited time access for a $5 a pop fee or something similar, I think you'd have a large audience (myself included).
this is one of the best posts I've read on Hacker News.
I love how you're open about your numbers. How you (debatably) made mistakes, and how you recovered -- and how I remember when you launched DomainPigeon :)
I think you could have avoided a pain point here by use of A/B testing: show premium users the same page you've always show them. Show some percentage of free users the new treatment ("premium, but for free") and the remainder the old default ("like premium, but less of everything you care about"). Then measure conversions. If your "premium, but for free" users don't buy statistically significantly more domains (enough to offset the premium revenue), then you can abandon the test and your premium users never have to know you had a crisis of confidence in your business model.
Its easy to say. Its hard to do, particularly with dynamic sites. Google Website Optimizer is a halfway decent solution if you can get it to work. I hope to have another halfway decent solution available for Rails users sometime in the not-too-distant future, as I'm sort of fed up with the GWO "all tests incur 1.5~2 seconds of extra load time as we figure out what URL to redirect to, because that totally won't murder your conversions" problem.
Most businesses fail due to not understanding their customer, not due to the technical side of the product. One strategy to combat this is to prove that you're making something people want.
"Design can’t make a bad site good and makeup can’t make an ugly woman beautiful. That is unless you’re drunk on optimism or alcohol."
Design can't slay the usability dragon alone. Well-said. I liked your article because you didn't even come close to saying 'X ways to succeed' but felt like you shared the highs and lows and mediums of startup.
Here's what I think: Short domains are nice, but in the new web, people search for things. So brands are necessary, and if your brand is searchable, the site is good, no matter how long it is. For example "I will teach you to be rich", is a pretty long domain name, but it's an excellent name to be found on a google search.
Add a new section to your site that scans books or other sources, and constructs sentences made up of 5 words or so, and offer them as domains to be registered. Maybe that will work better. Example: "Itwasadarkandstormynight.com"
Thanks for that. I am so prone to the endless tweaking of design minutiae rather than the much more effective sculpting of functionality and usability. I will try to follow your advice! The part about word choices and font-hue really hit home, ::sigh::.
The weird thing about this story is that it is so true… but if I would have read it one year ago, I wouldn’t have believed a word out of it. Especially for the “narrow your audience” tip. Couple of months ago, I’d say that our project monitoring app is for “anyone who works on a project”. I was wrong, we need to redefine our user description.
That is a tip that should be more written about. It currently exists in the “find your niche” form but it is wrong. People think “All I have to do is to find my niche then, everyone will use my product”. Instead of finding your niche, narrow down your audience.
It seems difficult to maintain an audience in a site like this without marketing. It's not a site people would be so eager to tell everyone about, as it might make it more difficult for them to register the good domains first. Perhaps the subscription should just be so expensive that it could be advertised? Conversion rate would have to be amazing though, considering how much competition there must be for terms like these.
I don't know. On one hand I've invested about seven months into this project and it feels like a waste to move on without having done much marketing or anything like that. On the other hand, I'm somewhat bored with it and really want to learn Python. Maybe I'll do both for a while. Not sure.
Well, be thankful you had it up, running, and actually making money BEFORE you got bored with it. I'm willing to bet a lot of side projects never reach those milestones before they were abandoned due to boredom.
This needs to be starred about taking all advice, ever, but particularly taking advice from people who are a) like you and b) unlike your customers. I always ask my forum buddies for opinions on new things (never turn down another set of eyes), but you have to take them with a grain of salt: at the end of the day, they're not the ones who pay you money.
A lot of folks had very strong words for my last redesign -- including at least one person who thought keeping the designer's name on it was akin to professional slander. My customers liked it, as evidenced by metrics(+).
Guess whose opinion matters?
+ Metrics: accurate, appropriate, and timely numbers which inform decisionmaking. Pick any two. Grumble grumble... but I do love them.