Congratulations. Our stories are a lot alike. I created Feedback Army in 2008 (I continue to operate it though). One thing I liked about reading your story was the lessons learned part. For a novice entrepreneur there's a lot of value in shipping something, marketing it, providing customer service, and dealing with the whole process. You'll find it's a good springboard to what you do next. Best of luck.
Dude great to hear from you! I'd really love to talk some time if you're interested. Feedback Army is awesome and I think we might have some cool stuff to share. Couldn't find any contact info in your profile but my email is dshipper [at] gmail.
$1000 to run something like that on EC2 seems like a lot. That type of site could probably be run off one or two micro instances, and if you bought reserved instances you can get your long term costs way down.
I was running small instances - probably should have been micro. And there were definitely ways to decrease costs and stay on EC2. Heroku is free though so for me (at my scale) it seems like the best option for now.
The most interesting part of this post is the guy who wanted to build a service that reviews you online dating profile. That seems like something that has real potential.
In my experience, one of the most frustrating things about being on an online dating site is that it's really hard to get feedback from women about how your profile is perceived by them.
You send messages to a bunch of women and 90%+ never respond, but you have no idea why. I would pay $$$ to find out why they don't respond, and I'm sure that a lot of other people would as well (especially men).
Could this be built into a site like OKCupid? Perhaps some kind of revenue sharing arrangement (between reviewers and the dating site) where users pay to get feedback, and other users get paid to review the profiles of men or women who message them(or maybe even those who don't)...and it's anonymous feedback...or semi-anonymous.
I really like the idea as well! From what I know the site didn't go anywhere because the minimum advertising commitment for dating websites is in the thousands. So I think he dropped the idea.
But definitely it could work for someone with some marketing skills and/or cash to burn. It's something that can be built in a weekend as well. Feel free to reach out to me and I can try to help point you in the right direction if you're interested in pursuing it.
Yeah, I think it would be very difficult to make this work as a standalone service, but I think it could be potentially be very successful as an internal feature of OKC or a similar site. It would be valuable to me to find out what the women that I am interested in think of my profile, especially the ones that don't respond to me. I would pay to get some decent feedback on myself. One of the problems is that I see my own profile in complete isolation. Actually, what I should do is go look at the profiles of other men to find out what the competition is like, but I'm not going to do that. But there are probably lots of women who would give me feedback on my profile if I paid them even $1.
Congrats on your first deal, Dan. Sounds like you learned some valuable lessons along the way. They will serve you will in the future. Cheers. PS: Will be interesting to see what new owner does with the site.
Very interesting story and I love these low maintenance ideas. Congratulations on that sale!
Can you elaborate a bit on how you drove customers to that site? I was looking at a similar idea just a few weeks ago but search engine competition on usability keywords is incredibly tough.
In short: I drove customers poorly and infrequently.
The only marketing I did was through my blog/Twitter. That spawned a fair amount of word of mouth and some other blog posts by people who had used the service. Beyond that I think it was just people talking to eachother because the product really does provide value if you're building a website.
I also have a hunch that the custom thank you videos played a part too but I can't prove it.
Dan, great article and congrats on the sale. Domainpolish is a great product.
For the metrics did you ever set up Google Analytics? I typically tell everyone to just install that first thing, it can give you ballpark estimates for almost any basic metrics question with about 5 lines of JS on each page.
For EC2, I agree with PaulHoule that it's possible to save money if tuned properly, but the learning curve is steep. I know I've served 5 or 10 minor projects of a single mini reserve instance ($6 / mo) with no issues, but getting a sense for it requires a bit of work.
Thanks for your comment! I did definitely set up Google Analytics. But I didn't have goals or Web Master Tools or funnels or anything like that set up. So I had a few metrics but nothing really in-depth.
Yea I definitely could have done the EC2 experiment much more cheaply. But Heroku is free if you're small scale like DomainPolish :) Thanks again for commenting!
Good stuff, keep it up!
When I was your age, I was making $200/mth through hard labor, and it was 10 times higher than an average student :-)
Since you seem inclined on entrepreneurship, I would highly recommend that you tackle negotiation skills, since those tend to define the financial outcome of idea+implementation. You can start with Jim Camp book "Start with NO...The Negotiating Tools that the Pros Don't Want You to Know" and reading about Richard Branson, and go from there.
Good luck.
Congrats, Dan! Fully agreed that given how most projects land the entrepreneur (and/or the investors) in the red, the real achievement here is executing an idea and seeing it through the full lifecycle of a business, in this case, through to a sale. The learning experience will be invaluable when you exit your next company.
This was a great story to follow through Hacker News, and your candor, detail, and introspection are admirable and indicative of the success ahead.
Thanks Jason I'm glad you enjoyed following the story. It's definitely not a huge exit but I'm happy to have turned a profit and there's more where that came from :)
When talking about "exits", it might also be useful to keep in mind the amount of money you could've made while working a regular in the same period of time.
Of course I'll be the first to admit that working on your own projects is much more fulfilling and interesting, but it does have opportunity costs just like everything else.
Definitely - although it's also worthwhile to take into account the fact that you own what you produce when you work for yourself. So while after working for someone else for 6 months I would only have cash to show for it, after working for myself (very part time) for 6 months I had the option to turn what I produced into immediate cash, or do anything else with it that I wanted.
But I definitely agree that it's something to consider.
> I set up 3 instances (two production and one staging) and thought I was a boss. [..] Lesson learned: don't use anything but Heroku until you have to.
Did he really need 3 instances? Or even a separate staging environment?
Thanks for the comment! Obviously 3 instances were too much. But the point remains that if you're running a low traffic Rails app it's a lot cheaper (free) to host on Heroku than it is on EC2. And if you're running at the scale that DomainPolish was that makes all the difference in terms of take home profit.
I was looking at Heroku as well for hosting some MVPs and I was wondering how good they are for hosting Python. Any ideas ? Naively, I'd hope they're as good as hosting Rails.
I wish I could help you but I've only ever used them for Rails. I have used GAE for Python and had a pretty good experience. I think that's also free at the beginning.
I was wondering about GAE as well. After the pricing change, it looked like it got horribly expensive. Any comments on how a side-project would do on the free plan ?
Web frontend-wise, there isn't much difference, so it depends on your framework. But I can't find a Python equivalent of HireFire or the likes to keep worker dynos count down. I haven't gotten around to write one yet.
So I ended up licensing the codebase to someone else so he could build a different Mechanical Turk app. I kept all the IP but he got the codebase to use. Thanks for asking I'll try and make that more clear in the post!
Not really. Yea it's actually just stupid looking back at it. But I hadn't done any pricing before and I wanted to experiment. So now I know at least I won't make a boneheaded mistake like that again :)
Thanks! I guess to me, more important than the selling price, was the process as a whole. It's a product that I took from an idea, built out and eventually sold. In there were a lot of lessons that I thought might be interesting to people.
$3,500 might not be a lot to you. And obviously it's not enough to go away on. But it does mean that I can work full-time on my startup this summer.
I thought the post was great. $3,500 is better than the $0 that most people get for their side projects.
Congratulations on the sale and it sounds like it was a great learning experience.
edit: If you don't mind sharing, how did you end up finding the people who were developing their own sites? The post mentions customer engagement but not how you found them in the first place.
Thanks it really was - although I regret not spending as much time on it as I could have.
I basically ended up doing no marketing for DomainPolish whatsoever besides a few posts on my blog which were popular on HN. A few other people tried out the service and wrote popular blog posts as well. I think doing the thank you videos in the beginning was key because it gave people a huge incentive to tell their friends about the site.
I wish I had more to say on marketing but that's something that I'm working on a lot now for my new startup. Hopefully in 6 months I'll have some more lessons to share :)
It really is difficult to name a price, I would have ignored any offers below 10-15k but on the other hand you can now concentrate on a new project and for 6 months from 0 to sale is remarkable.
according to some people a quite common exit is between 12 and 24 months[1] of the net profit. Considering his expenses (1700$) and sales (2500$), and initially ignoring the licensing fee, he had 800$ profit for six months, so let's round it to 150$/month, 3500$ is 24months of this profits.
If we want to consider the single licensing fee, let's say he licenses once a year, it's 1500+1800 = 3300$/year, and so 3500 is a 12month evaluation. Still acceptable.
10-15k$ I think it was not acceptable (by the buyer of course) with this numbers, even thought he could have reduced significantly his EC2 expenses (and so having a much higher monthly profit)
I always assumed the 12-24 month-rule only applied to "content"-sites like a blog or a forum, but Dan built a unique service that cannot be copied with the same effort as a techblog or a bulletin board
Start-ups are not judged by the same metrics as brick-and-mortar companies or established service companies.
When you're starting out everybody knows that the first couple of years will likely (but not always!) be a time when money is tight if you want to maximize your growth and hence profits will likely be depressed. You can reduce your growth to increase your profits but most parties would rather aquire a much larger company with an amazing growth track-record than something that earns a few thousand $ / month on a much more flat growth curve.
Thanks! That's one part of the experience that I'm definitely upset at myself for. But I'd rather get the mistakes out of the way while the stakes are low right?
Ummmm, congrats, but it sold for $3,500 - that's not really an exit imo. It's like selling a decent domain name.
What's the point of even being on an entrepreneurial site if you can't celebrate someone's first success?
For instance, Dan complains about screwing up the offer he accepted. If I was in his shoes, I'd be very happy to learn this lesson now rather than when looking at five or six-figure sums - let alone anything bigger! ;-)
If you're going to use that definition, then everything becomes an exit... Work 12 months to make $2500, less than you would make in 3 months working part time job at Wal-Mart, must be an exit. Flip a used car for a $300 profit, another exit! Put up a garage sale, exit again!
Seriously, it's a lesson learned, experience gained. Not an exit.
Edit: to the downvoters, when you use the words "hate", "love", "epic" to validate some relatively-minor (to actual meaning) event or emotion, you're taking meaning out of that word ... in some cases just to feed that meaning to your ego.
Good write up. It's always nice when people share that kind of information / data. If I learn one thing from the time spent reading it, it was well worth it.
So, review a web site; review a dating profile (buyer). Why not a 'review anything' service?
I think you could definitely white-label the DomainPolish idea and sell it to people hoping to build a Review X service. The problem is that if you just try to build a site yourself that lets you review anything it's very difficult to target your customers.
People search for "get feedback on my website" not "site to review anything". Picking a specific vertical makes marketing much easier. But it's a cool idea! I can put you in touch with the current owner of the DomainPolish source - he'd probably license it to you :)
If you were going to do the 'review anything' concept you'd need serious marketing cash. You'd also build verticals within the service to target, to benefit from search engines and try to narrow on the big core markets (but still stay open to literally any kind of review someone might want to do). Same way Oracle etc customizes their business software "solutions" for verticals.
I'm terrible about that (the licensing idea), I'd be far more likely to just code it from the ground up, even if that's the stupid monkey approach. I like having a personal relationship with my code (lol), the structure of the site, and knowing how that correlates to projected scaling, and so on.
Congrats. What was the justification for that price? A certain number of months of earnings?
For something that required little work to maintain and produced consistent revenue, that should be 18 months earnings at an absolute minimum. 36+ months for this probably wouldn't have been difficult at all if you had put it on Flippa or similarly marketed the sale. Not trying to make you feel bad, just giving you some info for the future.
