This story begins in March ’08. Bored with the project I was working on and overly optimistic because of TechCrunch articles, I decided to create some sort of online business. I hesitate even now to use the word “business” as it implies some sort of organization or planning or something which my endeavor lacked in a big way.
Armed with a strong background in poker and Visual Basic, I decided to make and sell an advanced poker calculator.
For certain preflop situations you can calculate with mathematical precision which hands you should play and which hands you should fold. You make some estimations about your opponent’s hand, tell it about the pot and stack sizes, click “Calculate”, and it would chug through a series of somewhat complicated calculations show you exactly when to go all in and when to fold. I dubbed it – you guessed it – ALL IN Expert.
Some of the results can be counter intuitive and I reasoned that making this process easy would be valuable to a lot of people.
And so for about the next three months I worked ambitiously on the software and the website that would sell it. I was wildly optimistic about its future success. Its almost embarrassing to admit now, but before it launched I estimated it would sell 1000 copies at $40 each for a net of $40,000. I figured I’d be happy with that.
In the days leading up to the launch I was quite excited. I made the final adjustments and released it late one night in mid-June. I made a post to 2+2, a popular poker forum, and announced its launch. I went to bed, expecting great things the next morning.
Well, all did not go as planned. No one bought the software the first day or the second, or the third… A week went by. Zero sales. A few people had downloaded the demo version, but none of them made the leap to purchasing it.
Frustrated, I made some tweaks, resdesigned some things, and waited hopefully for people to come. I wrote several articles and posted them on poker forums, showing how ALL IN Expert could be used to help out in some situation. I added a feature where the results could be exported to a forum-friendly format which included a link back to the site, hoping that once a few people began using it others would see its utility and register their own copy. A few people found the site and eventually one person purchased it after I had lowered the price to $10. It became pretty clear that ALL IN Expert wasn’t going anywhere. I took down the site, ready to move on to other things and put the whole episode behind me.
My little entrepreneurial experiment failed pretty badly: I didn’t make much money and reality smacked me in the face. However, I did learn a lot in the process:
Don’t Rush Into a Project
There are a lot of romantic startup stories about founders who have this brilliant idea, start coding, and in no time at all there are investors knocking at the door touting multimillion dollar valuations. I’m sure this does happen, but it’s probably a lot rarer than we’re led to believe. The media tends to ignore the low-flying companies that never make onto their radar and rightfully so. The companies that do become successful get a lot of attention, creating the impression for the naïve that the numbers are better than they actually are. I learned later that this is known as the survirorship bias.
ALL IN Expert was not a startup, but if it was, it would have fallen in the never-makes-it-onto-the-radar category. Actually, ALL IN Expert barely made it off the ground before it crashed and burned. I think a lot of the problems could have been avoided if I had taken more time in the beginning to analyze the situation. Literally, I came up with the concept for ALL IN Expert one night, thought “Hey, that could work”, and began programming the next morning. I should have taken more time to figure out if this was something people wanted, to figure out how much time it would take, to look at alternative projects, etc.
Creating a successful business takes a lot of time and energy. Before embarking on an entrepreneurial journey, make sure you’ve done your homework so you don’t create another ALL IN Expert.
Research the Competition
Several free poker tools existed at the time I started on ALL IN Expert. PokerStove had been around for several years. Pokerazor was the new kid on the block and while it was quite advanced it wasn’t gaining much traction. Stoxpoker had also just released some similar software.
My analysis went something like this:
PokerStove – generally powerful and easy to use, but missing some important features
Pokerazor – bloated and prohibitively difficult to use
Stoxpoker – screenshots look like Pokerazor, must be the same:
My plan was to make a simpler, more practical tool that focused on usability. Sounds good, right?
The problem is that I barely used Pokerazor and I never used Stoxpoker. In my mind it didn’t matter what they could do–they were ugly and hard to use; ALL IN Expert would be pretty and easy to use.
I went with more of a “Ignore the competition and just make something great” approach, which sounds good until you realize that you’ve made a huge mistake by not learning from your competition.
When I finally launched this was the very first response:
So this is like an extremely simplified version of the Stoxpoker EV calculator for one very specific scenario?
Yes. No. I don’t know. I never used it. Can you just pay the $40 registration fee and not ask that question?
About a month into the development I did one thing well: my brother was visiting and I asked him to test out the software. This was the second most eye opening thing I got out of this project next to realizing after it launched how badly I had misjudged the entire thing.
He had no idea what was going on. “Click on the hands you think your opponent might have,” I explained to him…. No not there… click the colored box. No, not that colored box. See, that means you think your opponent can have that hand. OK, now click they gray box that says “Calculate”. OK, now click on the tab that says “Results.” No, its at the top of the window. At the very top center of the window, about an inch from the top, it says “Results”…
It went on like this for about twenty minutes.
Could it be that ALL IN Expert was just as hard or harder to use than the other software?
In the weeks that followed I spent a lot of time redoing the design and in the end I think it turned out pretty well thanks to that one night of watching my brother use it.
His feedback helped me avert a disaster in terms of usability, but what I didn’t see at the time was that no amount of usability will make a crappy product useful.
My big mistake here was stopping with my brother. I did have a few people test it prior to launch, but no one in between the initial version that my brother saw and the final release version. Had more people given me feedback early on, I might have seen sooner that there wasn’t a market for the product. I also never had anyone test the website, which might have helped as well.
Goodbye Desktop Apps
Did I mention ALL IN Expert was desktop software? I was most proficient at desktop programming languages so that’s what I went with. Never again.
Big picture, I think the trend is clear: desktop apps are a dying breed; web apps are the future.
There’s also all sorts of issues with deployment, bugs, and upgrades.
When my semi-final product was ready to be launched I had a friend test it out. Little did I know that there is a known compatibility issue with some Microsoft DLL file which causes some computers not to restart after installation. His computer happened to be one that was affected by the problem. After he installed ALL IN Expert his computer wouldn’t boot. He wound up having to spend four hours on the phone with Microsoft tech support (I’m so sorry) fixing the problem.
What if his computer hadn’t crashed? What if I had released it and dozens of people’s computers crashed because they had downloaded it? That could have been the meteorite that killed the business, but luckily I discovered it beforehand. Not that anyone downloaded it anyway, but still…
Make Something People Want
I had read that advise at least a dozen times from Paul Graham and Hacker News denizens, but somehow I thought I’d be the exception. The product would be so great that it would cause people to want it.
In reality, the competition’s tools, while not perfect, were adequate. No one was looking for a new poker calculator. Sure, they might take something that was better, but it wasn’t a burning need.
There were also some practical issues.
ALL IN Expert produced some fascinating theoretical results. The software crunched through the numbers and in many situations suggested that I had been making costly mistakes.
The problem was that the software was not designed to be a real time adviser. It was designed to be a study tool to analyze specific situations after they had already occurred. This severely limited its usefulness. Had I made a feature that enabled it to attach itself to a poker window and extract the information in real time, people might have been able to utilize the results. As it was, it would take too long to input the necessary data to be of any good real-time.
Also, while the calculations were exact the user still had to estimate his opponent’s hands. In an actual poker situation you won’t ever know your opponent’s range of hands with absolute certainty and even if you did, you wouldn’t be able to chug through all the necessary calculations to figure out the profitability of each possible move. You wind up using your judgment and making some estimations. ALL IN Expert would help educate you, but as a practical issue, it didn’t matter that it could show you your ROI to three decimal places.
Niches are Good, Super-Niches are Bad
I thought my original audience was online poker players. Period. In reality, my audience was online poker players who wanted to calculate the results for all 169 starting poker hands at once vs one at a time. Turns out there is a big difference.
Focusing on a niche is a good strategy, but make sure you know exactly who those people are. If you’re thinking thousands of people need your product when in reality only a few hundred do, you’re in for some trouble.
I started the price at $40. After it was clear no one was buying it I dropped it to $30, then $10, and now I’m just giving it away. Someone smart at 37Signals wrote that the best way to figure out how much you should charge is to ask yourself how much you would pay for it.
In this case, $10 sounds about right. What I should have taken into consideration is that given my small audience and the small price, I was probably never going to make more than a few thousand dollars even if I was really successful. That alone should have deterred me. If you’re going to focus on a super-niche, make something incredibly useful then charge accordingly.
Additionally, it’s very, very hard to charge for a product when your competition is offering it for free, no matter how bad their usability is.
You’re told to be persistent and not to give up, but you suspect that you’re racing towards a dead end. What do you do?
Five or six weeks in this occurred to me. I started to think I had overestimated the market and that I might be headed for trouble. I had a choice: press forward and see what happens or abandon what I had done and move on to something new. I decided to give it a few more weeks and see it through to the end.
It’s not easy to admit that you’ve made a mistake, but it also doesn’t get any easier as time goes on. The closer I got to completing it the more I realized I had a problem, but I couldn’t stop because I was so close. It’s a bad spot to be in. Do yourself a favor and if you strongly suspect bad weather ahead, take a different route.
Difficulty Gaining Users
Robert Stephens, founder of Geek Squad, wrote in the Oct 08 issue of Inc magazine that “Advertising is a tax for being unremarkable.” My first thought was ALL IN Expert.
My advertising didn’t come in the form of outward appeals to customers. Instead, I found myself focusing on ridiculous design issues in the vain hope that people would suddenly flock to my product. If you’re spending too much time early on worrying about the font and the colors and not focusing on functionality, chances are you’re in trouble.
I made a lot of strategic errors with ALL IN Expert. In the end the software looked good, worked well, and performed the job it was designed to do. I learned a lot of new things in the process – Photoshop, Paypal integration, the Inno installer, the importance of usability, some critical business concepts, and most importantly that I really enjoyed doing it, which despite what it looks like makes it a success in my book.
Coming soon: Domain Pigeon, where we find out just how much I’ve learned.
Few screenshots from ALL IN Expert, which is now free to download.
Telling ALL IN Expert a bit about the situation:
Estimating your Opponent’s Hands
ALL IN Expert shows you which hands to go all in with: