IntroductionBuilding and running this business has been a fun experience. However, I don’t have the time or the bitcoins to keep it growing to the extent that it could, and have been running it on autopilot since August. Despite its good health (59.109 BTC was deposited and 40.056 BTC cashed out in the last 30 days), the site would be of much greater value to someone with the time to take it to the next level, so I’m putting it up for auction.
With a little effort, you could seize opportunities for more revenue that have been out of reach for me and avoid some of the mistakes I’ve made.
For instance, several people have inquired about the affiliate program, a major source of new users, and I haven’t responded. I also stopped all forum activity many months ago, which had been a decent source of business. Pursuing those avenues would undoubtedly increase revenue.
Because of my lack of time to devote to the site, I made some mistakes that have significantly impacted the bottom line. I failed to notice that, for about a month starting in 2/2014, new users weren’t able to make deposits (a 3rd party service provider changed their API, and new users couldn’t log in). Also, the limits were set way too high for a time, which nearly bankrupted the site. Both mistakes were due to my lack of time and inattention to the site; disregarding those errors, the site’s profit would have been at least several times higher.
Other opportunities are out of reach for me. For example, raising the limits and re-branding as a high-limit site would likely bring in a lot more revenue. I unfortunately don’t have a big enough balance to support high limit gaming (the blackjack limit is only 0.1 BTC).
The business is in good health and can continue on autopilot. However, you can take it to the next level by recruiting affiliates, promoting on the forum, and increasing the limits. You’ll get back what you put in; the sky is the limit. Revenue and profit[Figures last updated Tue, 20 May 2014 04:30:50 +0200]
[To view all deposits and cashouts, navigate to /transactions. You can enter start/end dates and see transactions within that period.]
Since the site’s inception last May, 316.478 BTC has been deposited by players, and 287.288 BTC cashed out by players and affiliates, for a total of
29.189 BTC earned for the house.
Our one affiliate has made more than the house: 35.738 BTC total. The figure above of 287.288 BTC cashed out includes the affiliate’s cashouts as well as the players’ (so the actual gaming profit is 64.927 BTC, shared between the house and the one affiliate).
The low profit is an aberration. Early on, I set the limits too high without simultaneously attracting enough high-rollers, and had a single very lucky player playing at the increased stakes.
If you disregard that one player’s activity, the house’s net gain would be 93.239 BTC.Up until mid-November, 180 had been deposited and 106 cashed out. Things seemed to be going well so I increased the blackjack limits to 1 BTC (without advertising that increase). In November, one user made by far the largest single deposit (20 BTC) and got very lucky, winning 50 BTC in a short amount of time. He was the only player to truly max-out the limits by playing every hand at 1 BTC. After that loss, I had to shut down the site for a week until I could reduce the limits to 0.25 BTC (I was traveling), because it became clear that I had insufficient funds to support gaming at the 1 BTC limit. No other max-limit players appeared, because I failed to advertise the limit increase.
Unfortunately, 0.25 BTC proved still too high of a limit for my balance. That same lucky player came back, and again was the only player to make large deposits and max the limits. He got lucky, winning 15 BTC more. I then lowered the limits to 0.10 BTC where they stand today.
That lucky player was an outlier: he played 719 of the 742 blackjack hands played at the 1 BTC limit. As stated above, if you ignore that player’s activity, the house would have made a net a net gain of 93.239 BTC (276.478 BTC deposited and 183.238 BTC cashed out (including all affiliate payments/cashouts)).
Another mistake I made was distributing promotional giveaways with low play requirements. One user won 3-4 BTC from a promotional deposit, and several others won significant but lesser amounts. I’ve since set up a better way to distribute promotions (it lets you see a lot more stats and a graph for each user), and have raised the default bet requirement.
Managed correctly, you can expect a very favorable deposits:cashout ratio. Balance sheetNotes about the figures below:
-Organic users are users who signed up during that month without clicking on an affiliate link. Referred users signed up via an affiliate link.
-“Affiliate” is the sum paid out to the affiliate that month. At the end of the month, we credit the affiliate’s account if their players have lost money. If the affiliate’s balance is negative one month, that negative balance carries over to the next month. That explains why the affiliate wasn’t paid for 2/2014 or 3/2014. Note that the affiliate actually earned about 0.500 BTC for 1/2014 but lost it at the tables.
-
Keep in mind that the transactions since August have occurred with almost zero effort on my part. I hired someone to help me with forum posting for an hour per day for a couple weeks (which brought in only a few players), but other than that (and occasional promotional giveaways to existing players) there has been zero marketing. Business has been self-sustaining.
5/2013
Organic users: 6
Referred users: 0
Deposits (9): 1.568 BTC
Cashouts (0): 0.000 BTC
Affiliate: 0.000 BTC
Net: 1.568 BTC
6/2013
Organic users: 32
Referred users: 18
Deposits (41): 8.037 BTC
Cashouts (19): 4.334 BTC
Affiliate: 0.500 BTC
Net: 3.203 BTC
7/2013
Organic users: 176
Referred users: 26
Deposits (84): 23.176 BTC
Cashouts (83): 4.456 BTC
Affiliate: 2.021 BTC
Net: 16.699 BTC
8/2013
Organic users: 135
Referred users: 32
Deposits (142): 40.957 BTC
Cashouts (83): 17.892 BTC
Affiliate: 10.461 BTC
Net: 12.604 BTC
9/2013
Organic users: 42
Referred users: 24
Deposits (122): 55.507 BTC
Cashouts (29): 18.638 BTC
Affiliate: 12.463 BTC
Net: 24.407 BTC
10/2013
Organic users: 89
Referred users: 22
Deposits (132): 45.911 BTC
Cashouts (53): 33.452 BTC
Affiliate: 1.986 BTC
Net: 10.473 BTC
11/2013
Organic users: 93
Referred users: 20
Deposits (116): 28.062 BTC
Cashouts (49): 71.319 BTC
Affiliate: 3.126 BTC
Net: -46.383 BTC
12/2013
Organic users: 42
Referred users: 24
Deposits (59): 3.147 BTC
Cashouts (26): 4.967 BTC
Affiliate: 0.590 BTC
Net: -2.410 BTC
1/2014
Organic users: 35
Referred users: 76
Deposits (237): 39.983 BTC
Cashouts (46): 48.025 BTC
Affiliate: 0.000 BTC
Net: -8.042 BTC
2/2014
Organic users: 9
Referred users: 13
Deposits (51): 9.903 BTC
Cashouts (12): 10.013 BTC
Affiliate: 0.000 BTC
Net: -0.110 BTC
3/2014
Organic users: 21
Referred users: 9
Deposits (18): 1.290 BTC
Cashouts (7): 4.847 BTC
Affiliate: 0.000 BTC
Net: -3.557 BTC
4/2014
Organic users: 14
Referred users: 14
Deposits (98): 37.117 BTC
Cashouts (27): 19.485 BTC
Affiliate: 4.592 BTC
Net: 13.040 BTC
5/2014 [as of Tue, 20 May 2014 04:41:38 +0200]
Organic users: 5
Referred users: 14
Deposits (44): 21.819 BTC
Cashouts (16): 14.123 BTC
Affiliate: 0.000 BTC [0.342 BTC owed as of 5/20]
Net: 7.696 BTC
[Last updated Tue, 20 May 2014 04:41:38 +0200]
From 2/2014 to 3/2014, there was a problem in which new users couldn’t deposit. It’s possible that new users couldn’t even sign up successfully. That partially explains the low revenue and signups during that period. Also, we lost our position on the front page of betwithbtc.com (our affiliate) during that period due to poor performance resulting from the unable-to-deposit bug.
11/2013 was a bad month (-46.383 BTC), but 50 BTC of that loss (and 20 BTC deposited) was from a single user. Keep in mind that that user was the only one to be playing at the limit. Of all the 742 blackjack hands played at the 1 BTC limit, 719 of those were from that player.
2/2014 was also a bad month (-8.042 BTC). That was also the result of the same player. The limits had been lowered to 0.250 BTC, but again, only that one whale was maxing out every hand at the limit.
PlayersThere are 990 registered, un-banned users, of which 685 have deposited. Navigate to /anon_users to see information about each user.
-470 have visited in the last 6 months
-190 have visited in the last 3 months
-104 have visited in the last month
-31 have visited in the last week
-130 have deposited more than once
-87 have deposited more than twice
-45 have deposited more than five times
-19 have deposited more than ten times
-181 have deposited more than 0.01 BTC
-91 have deposited more than 0.1 BTC
-41 have deposited more than 0.5 BTC
-30 have deposited more than 1 BTC
-19 have deposited more than 3 BTC
-7 have deposited more than 10 BTC
Operating costsI’m currently paying $225.50 per month for hosting and related services, but you could easily operate the site on $40/month. Here are the current monthly expenses:
Managed hosting platform: $190.50
I have been paying way too much for hosting. I recently launched a copy of the application on a $20 VPS and it works fine. Were I to keep the site, I’d probably run it on a $40 Linode VPS. However, the current setup is very easy to run and I suggest you keep the site running at the current host at first and set it up on a VPS of your choice later. That way, you can be assured that the hand-off goes seamlessly.
Airbrake.io: $20
This is an error reporting service. I actually don’t pay this anymore so it defaults to the free tier. It would be a worthwhile expense, but you could probably find a free or cheaper alternative.
Google apps account: $5
I had a Google apps account associated with the domain, but that account will not be transferred in the sale (the site has recently switched to a free Gmail address). I’ll cancel the Apps account and you can re-register it at a later point. But for now, you won’t be paying this cost. I switched to a Gmail account because the Google Apps account can’t be transferred.
Amazon S3: <$5
This is only used for hosting a few of the site’s images. You can host those assets from wherever you want though. For the handoff, I’ll have the assets hosted from the application itself.
Twilio: <$5
The application sends SMSs to an admin when things happen (when players sign up, deposit, or cashout, or the site shuts itself off, or the house balance dips below a certain level). You can turn this feature off.
Pusher.com: $0
I’m using Pusher.com’s free tier, which is fine for current traffic levels and then some. Traffic / message volume has never exceeded Pusher’s free tier limits. You might have to upgrade this to the $20/month package when traffic goes up.
Total: $225.50 USD per month
However, if you were to host on a VPS without the unnecessary services, you can expect to pay about $40 per month at current traffic levels. The sale and handoffThe sale is for the following items:
1) Full source code and images (on a remote code repository)
2) Account at the registrar where the domain is registered
3) Gmail account
4) A VPS hosting account running a parallel / identical version of the application with a copy of the production database (so that it’s easy for the escrow provider to quickly verify that the code being delivered is indeed the code that powers the site).
Even though the sale is technically only for the code, domain, email address, and VPS account,
I’ll transfer the complete, running application to you at its current host. I can’t make this seamless transfer of the running application part of the official deal, because I want all deliverables to be completely and easily verifiable by the escrow agent. When the escrow agent has verified that everything is in order, he will send the credentials of the accounts above on to you and release payment to me. At that point I’ll transfer the running application to you at its current host. Note that with only the accounts above, you’ll have complete control of the site and can deploy it to any host you want.
How the transfer of the running application will work: I’ll transfer the application to you through the managed hosting service I’m currently using, which is one of the leading platform-as-a-service providers. Note that this will require you to set up an account at the hosting service and enter a credit card number. You’ll probably want to deploy the site to your own VPS afterwards to save money, but transferring the site to you this way ensures a more seamless transition. You’ll be assured that the infrastructure is sound and that the box hasn’t been rooted, and will have the confidence of knowing the application has been sitting on that infrastructure, mostly untouched, for months.
Note that the release of funds will be at the discretion of the escrow provider, who will verify that my end of the deal has been completed.
The escrow agent will do the following before releasing the funds to me:
-Log into the following accounts to verify credentials: domain registrar account (and will verify that it indeed hosts the domain), code repository, Gmail account, VPS account.
-Visit the version of the application served by the VPS and test the games a bit in free mode to be reasonably certain that the VPS is running the code that powers the production version of the application. Also, verify the /transactions page and /users_with_balance page to be reasonably certain that the database is really a copy of the production database.
-Send those credentials to you, the buyer.
Note that when you send the payment to the escrow agent, you’ll have to provide me with the wallet address where you’d like to receive payments as well as an admin email address and telephone number (if you wish to use Twilio).
When the payment is released to me, I’ll also send you a list of instructions for completing some basic tasks which you should do immediately, which will give you an idea of how to perform various management tasks. I’ll send instructions on how to perform the following tasks:
-Log into each account associated with the site (hosting, registrar, Gmail, Pusher)
-Log into the web admin interface; visit various admin pages; and perform various user management tasks, issue a promotional bonus to a user, create a new promotion, log in as a certain user, manage affiliates, etc...
-Log into the programmatic interface console and perform various queries and tasks, including putting a balance hold on a user and then removing it, checking the cumulative user balance, checking the house balance, etc...
-Stop and start the site.
Turning it overYou will inherit the site’s debt.A few users maintain a balance. Also, the affiliate is due for a payout towards the end of the month. You’ll inherit that debt.
The cumulative user balance is minimal (few users keep a balance on the site) and does not fluctuate much. Navigate to /users_with_balance to see how much is owed to whom. You’ll notice that there are a few people with balances over 0.500BTC. Note that almost all balances are due to promotional giveaways, not deposits, and that those users can’t cash out until they wager a significant amount (60BTC or so- the amount is listed on that page). In other words, users who have significant-seeming balances are almost all recipients of promotional giveaways that are quite unlikely to ever be realized. Also, note that many of those users also haven’t visited the site in over six months and are unlikely to ever return.
To calculate the balance owed to users who do not have to place any more bets before they cash out, you can just add the balances of users that have 0.000 in the “must wager before cashout” column.
That figure is only 0.197 BTC as of Tue, 20 May 2014 20:04:35:45 +0200. That is a much better estimate of the debt.
Also, note that our one affiliate is due for a cashout on May 25. They are currently owed 0.342 BTC as of Tue, 20 May 2014 03:28:26 +0200.
My future involvementI’ll happily provide support for a month after the sale. During that time, you can ask me any questions you want about the code or how to use the site, and I’ll help get you up to speed on the mechanics of running and promoting it. I can only make changes to the site before the hand-off, however, and only superficial ones, such as changing text and game parameters (When I say before the hand-off, I mean before I transfer the running application to you at its current host. I will transfer the VPS version of the site to you immediately after you send money to the escrow agent.) I want you to be successful with the site, so I’ll do what I can towards that end, but I don’t want to over-promise.
The site is being sold as-is. Once the site it sold, it is no longer my responsibility to fix any bugs that may pop up. However, I’ll help you fix anything that occurs within the support period, and even afterwards I’ll be very inclined to help you out.
Rest assured that the site has been operating for many months with minimal intervention.
Who you areA non-technical person can perform most of the tasks required to run the site via its web admin interface, but it would be better if you have programming experience and are familiar with Ruby web development. If you are non-technical, you should at least have access to someone who is comfortable with programming, ideally with Ruby on Rails and Javascript experience.
There shouldn’t be any need to update the code regularly other than to change parameters (game limits and so forth). I can help you out with tasks like that after the sale. There’s no guarantee beyond the support period, but I’m very inclined to oblige reasonable requests.
AuctionThe bidding starts at 2 BTC and the increment is 2 BTC. The auction ends on Monday, May 26th at 6pm server time, There is a hidden reserve and I reserve the right to refuse sale.
EscrowIf you wish to use escrow, we can go through the following bitcointalk users:
Dabs:
https://bitcointalk.org/index.php?action=profile;u=54791DannyHamilton:
https://bitcointalk.org/index.php?action=profile;u=60820I’m still in communication with those users and others. This is a tentative list and I may add another later. I will pay all fees.
See the “The sale and handoff” section for more about how the transfer will proceed after sale.
Concerns about the transferA prospective escrow agent has expressed concern that the transfer poses a risk for the buyer in that I might be able to retain some control of the site.
Regarding the domain: once I hand over the credentials for the hosting account, you’ll be able to change the password and/or transfer the domain to any other registrar.
Regarding the application: you can host it anywhere you want and make any changes you want to the source code. There are no backdoors or malicious code, as you will soon discover; the code is structured as a normal Ruby on Rails app, so that if you’re familiar with the conventions you can scan through it quickly. Note that the infrastructure of the platform-as-a-service hosting provider I use can’t be rooted or otherwise tampered with, so you’ll be assured that the running version I transfer to you at the current host will be “clean”. Also note that after the transfer you can host the application from any host you choose.
This next point may be a bit confusing, but someone brought it up so I thought I’d address it. One concern is that the code running on the VPS account (a part of the “official” deal) is not exactly the code that powers the real production version of the site (which I will transfer after the payment is released to me, but is not part of the “official” sale). The reason I’m offering the app running separately on a VPS to you is so that the escrow agent can quickly verify that you’re receiving the code that powers the production version of the site (the code run by the VPS obviously resides on the VPS); the escrow agent will quickly and easily verify this by playing around with the running application until they’re reasonably certain that it works identically to the production version of the site. So with just the code on the VPS and control of the domain, you’d be able to host the site anywhere you want. I have no incentive to present a different version of the code to you on the VPS.
Another reason I’m offering the VPS is because I’d initially planned to host the production version of the site on the VPS (rather than the PAAS) and just transfer that to you, but I later realized that that presents several worse concerns for the would-be buyer, such as the possibility of the box being rooted (and the necessity to soon re-launch the site elsewhere which would be a pain not only for you, but also for me to coach any non-professional web developer through), and the need for more technical expertise to manage the site. So I scrapped that idea, kept the production version running on the PAAS service, and decided to offer the version running on the VPS as an added assurance to the would-be buyer and as an easy way for the escrow agent to verify that the code has been transferred.
A related concern is that the database on the VPS might not be the same one as on the production version. First of all, it won’t be an exact copy, because it takes a few hours to copy and download the production DB, and I’ll be deleting a couple of user accounts that I own (like my admin account) for my own privacy (none of the stats above include activity from my user account). Beyond that, I can’t think of a reason I’d withhold database records from the buyer other than plain maliciousness. Anyway, I’ve made it easy for the escrow agent to quickly verify that all the user records and basic stats on the VPS machine match those on the production version.
MiscellaneousProof of ownership: /proof
Bitcointalk thread:
https://bitcointalk.org/index.php?topic=214877 Appendix: About the siteGamesThere are three games: Roulette, Blackjack, and Slots. Roulette was the first game, launched about a year ago; Blackjack was launched about 10 months ago; and slots was recently launched (it was 99% completed since August but I just recently pushed it out). I built all of these games from scratch. There is a pretty ample test suite written for each game. All games are provably fair. See the site for more details of each game.
Note that my low balance means that I can’t support the “correct” slots payouts; the highest win could pay as high as x500 (the RTP would be 99% in that case), but currently the payout for that combo is set to x100.
Social playSocial play is one of the defining features of the site. In any game, you can see the other users who are logged in, as well as their balance relative to other players (reflected in the font size of their name). Also, in Roulette and Blackjack, players’ names flash green / red if they win/lose (in slots, players’ names only flash on win).
In Roulette, players play at the same table, and play against the same “spins” (which occurs every 30 seconds). If users have the setting enabled, their bets are made public. That means that if that user places a bet, their name is highlighted purple; if other users hover over that purple name, they see the bet that was just placed. If they click that player’s highlighted name, they can place the same bet as that user.
Note that most of the players you see on the site are actually site-controlled bots. These bots make it seem like many people are playing, to entice new users (the bots aren’t actually gambling; it just looks like they are). The bot system is fairly elaborate: each one is assigned a time zone and has a preferred time to play, and all bot actions are probabilistic. The average number of bots can be changed (you can also turn this feature off).
AffiliatesThe affiliate system has brought in a lot of business. When you sign up affiliates, they receive a referral link. They earn x% of the profit from the users they refer, for life. Cashouts are issued every month, but they aren’t automatic- you have to click a button to cash the affiliates out. Signing up affiliates is easily done via an admin page on the site and you can enter what % they earn.
There is only one affiliate: betwithbtc.com. We are usually on the front page of Betwithbtc.com, which is the third Google result for “bitcoin gaming”.
Several others have inquired about the affiliate program, and all inquiries have gone unanswered. I can forward you the emails of the interested parties.
PromotionsOne challenge is getting users to deposit again. To achieve this, admins can issue promotional deposits to users’ player accounts to get them back to the site. Applying a promotion alerts the user via email.
When applying promotions, you can either give them straight BTC which they can cash out immediately (not recommended), or apply a “bet multiple requirement” for that promotion. The bet multiple requirement is how much the user must put into play before they can actually cash out. For example, if you give out a 0.100 BTC promotion to some user, and the bet multiple requirement is 50, that user will have to wager 5.000 BTC (over as many rounds as they want) before they can cash out their winnings (losing 0.100 BTC will also annul that requirement). So you can make it very unlikely that promotions will actually be cashed out.
Admin interfaceYou receive an email (and optionally a text message) when important things happen: when users sign up, deposit, cashout, or when the house balance dips below a certain threshold.
There is an extensive web-management console. Here are screenshots of various admin screens:
View currently-logged-in players, and links for other admin pages: /assets/:admin.jpg
The admin’s view of a user’s page: /:users:show.jpg
Distributing promotions to players: /:promotions.jpg
The users table: /admin_users.jpg
CodeThe casino is a Ruby on Rails 3 app built from scratch. The code includes a large test suite.
Owner’s manualI’ll provide a short owner’s manual to accompany the site, with instructions on completing basic tasks, lessons learned, and tips on promotion.
HistoryRead through the main Bitcointalk thread (
https://bitcointalk.org/index.php?topic=214877).
Early on we ran raffles where users had to place banners in their signatures for a few weeks, which was somewhat successful.
There was a professionally-designed banner made (which you’ll get), and is visible at betwithbtc.com.
Ask all questions before bidding!
Show Posts
