[The world of Flash game sponsorships and hosting is a complicated one, and developers are often at a loss as to what sites and APIs provide the greatest tangible benefit. In this in-depth case study, SteamBirds developer Andy Moore walks through his experience monetizing his game, going so far as to provide rough dollar amounts.]
My free PC Flash-based action-strategy game SteamBirds
[YouTube trailer video
] launched on March 3rd, which means weíre coming up on the one month birthday soon. There hasnít been enough time to collect all of the game's secondary licenses, some of which are scheduled for mid-April and beyond, and there definitely isnít enough time to get a clear picture of the user traffic and how big the tail is.
But these figures provide a picture of the game's core revenue, and the general shape of how business works in the single-player Flash game space, and I hope to release further data as things progress in the future.
Please note that some sponsors have asked I donít reveal exact figures or associate their names with their price tags. Iíve fudged all of the numbers by a certain percentage and made some anonymizing edits so I donít step on any toes, although Iím sure some basic sleuthing can get you any answers you need.
The Development Split
I had a tough time deciding on a proper revenue split model. I wish there were 150 percentage points so I could pay everyone a bit more, but I think the ratio I have is fair to everyone. I wonít lie; I do wish I had more money for myself. Who doesnít? But even if I cut someoneís wages back Iíd feel like giving it to someone else, not myself. At least Iím happy with the final result.
Of the game's upcoming iPhone port, courtesy of SemiSecret Software
(makers of the excellent games Canabalt
and Gravity Hook HD
of all revenues go to Apple first;
of the remaining revenue goes to SemiSecret Software; and
then goes to the SteamBirds Team (see below).
Outside of the iPhone:
of all profit FlashGameLicense
inspires goes back to them. They get paid first, because without them Iíd only have made a fraction of the money. Their service is truly excellent for the Flash portal distribution model, and not using them is foolhardy. Not paying them is just mean.
Of the remaining profit for the SteamBirds Team:
goes to myself, for game design and programming. I dare say I put some kick-ass lines of code into the game, and I wish I could showcase them as easily as a work of art.
goes to Daniel Cook
, a true game design genius and excellent artist to boot. Before SteamBirds
, I donít think I ever would have worked with a ďgame designer,Ē thinking it was mostly elementary. I primarily hired Dan on to be an artist. Now I donít think I can make a game without his advice.
goes to Danny Baranowsky
for his musical genius (he also pays Jordan Fehr
for sound effects out of his cut). I get a lot of shocked looks at this one, and I agree; Iím used to paying only a meager sum for audio. But the services Danny offers (perpetual unlimited music!) and the quality of the end result is astounding. Absolutely astounding. I do believe in the gameplay as a standalone product, but the music takes it to a whole other level of awesome. I think itís worth it.
(Percentage splits have been classified upon request.)
The Bidding Process
Sponsorship bidding for SteamBirds
opened on FlashGameLicense
on January 30th, 2010. I purposefully chose an early-month entry so any larger companies bidding would have fresh monthly budgets at their disposal. I also consulted with FlashGameLicenseís sales statistics
to give me some insight; it seems January is the worst month of the year and February is a close second. I wanted to wait until later in summer, but I didnít want to wait any longer to launch the game!
Eventually, seven sponsors bid on the game, and a dozen more emailed me offers and discussed options on the side; the latter group tended to be complex-enough offers that posting them to FlashGameLicense wasnít appropriate. Those seven actively-bidding sponsors made 27 bids in a ten-day battle for control of the game. Bidding opened at $500 and quickly made it above $6,000. Hereís a quick rundown of the stand-out bids I received through all avenues, including email:
- 1/30: $500: Opening LOL bid
- 1/31: $1,500
- 2/01: $2,000
- 2/01: $6,000: A big jump right away. There were no bids between $2,000 and $6,000.
- 2/04: $8,000: A bidding war follows.
- 2/04: $10,000
- 2/05: $16,000
- 2/06: $20,000
- 2/07: $21,000
- 2/10: $22,000
- 2/10: $25,000: The deal I went with.
- 2/12: $30,000
- 2/12: $35,000
- 2/15: $45,000
I closed bidding around February 16th and notified Dan at Armor Games
that his bid had won.
"ButÖ but why?" I hear you incredulously shouting. "Why would you take such a low offer?!"
"I'll tell you," I respond with an eerily calm demeanor.
The high-value deals ($35,000+) had some very restrictive licensing terms. Restrictions like these were common:
- Rights to first refusal for a sequel
While I understand that "rights to first refusal" arenít all that binding, it still makes me feel uncomfortable. What if I have a bad experience with the sponsor? I donít want to feel locked into anything. If the sponsor behaves and is a nice partner in the venture, Iíd be more than willing to go with them again for a sequel. Why put a friendship in writing?
- No sequels, spinoffs, or use of any existing assets or similar assets for one year
I have some pretty big plans for the future, and SteamBirds
as an IP is ripe for the picking. Especially now that the game is a resounding success, I cannot ignore it. Yes, I understand it is important to space out sequels and launches; I donít want to over-saturate the market with my products. But I donít want to be forced out of my own work for a year.
- This deal is considered pre-payment for a spin-off version containing bunch-of-features XYZ, in addition to primary sponsorship
These offers were tempting. But in the end the extra amount offered wouldnít make up for the time and effort that goes into the new features.
- Perpetual exclusive license; no secondary sales or ad revenue
That term is fair enough -- as long as you can pay enough to outweigh that lost revenue stream.
How I Chose My Sponsor
To help me sort through the bids, I devised a ďvalue scale." I projected how much I thought the game would make on secondary license sales, advertising revenue (which I estimated to be near-zero, for what itís worth), and other such income. I included figures like bonuses for performance (if any), prize money (if allowable), and any other special-case funds. I then added these values to the base price of the offer.
In my mind, the Armor Gamesí offer was worth around $30,000 to $40,000, despite its $25,000 face value. By accepting Armor's deal Iím fully expecting to reach $35K (right in the middle of my predictions) in total revenue over the lifetime of the product.
Unfortunately for the larger offers, their restrictive licensing terms didnít result in a whole lot of added value. In fact, some had negative
added-value when dealing with longer timeframes and exclusionary periods.
In the end I was left with two very similar deals I was seriously considering. They had comparable terms, and Armor Games was actually a little behind on my value scale. But I went with Armor anyway, for one very important reason: personality.
Dan McNeely of Armor Games is amazing to work with. Early on in the bidding, we struck up a rapport and we discussed many things -- even those not game-related. Several times he suggested other deals might be better for me. He often tried to work out new agreements, suggested other sponsors I might want to approach, and is on good terms with many other sponsors. Best of all, he always kept his licensing terms wide open with amazingly few restrictions.
Most sponsors say, ďThis deal is the best deal for you, the developer.Ē It always sounds like something a used-car salesman might say. Dan earned my trust, and I legitimately think heís looking out for me and my well-being; he didnít have to say those words. The words were implied.
Sponsors are more than just money-machines. Theyíre partners with the future of your game-making career. Accepting Danís bid is likely to have a long and prosperous relationship between us, where I can trust him to help me find the best deals (hopefully with him again!). Though not tangible, thatís worth a few extra grand in my pocket any day of the week.
You simply wonít believe the number of sponsors who approached me with aggressive stances, insulting me or calling me names for not taking their offers. Some sponsors wonít even give me the time of day and havenít responded to emails in the last two months. Some seem to think they are superior to developers and put on an air of ďyou will bend to our will.Ē Seriously, you guys? Not very professional.
The Terms of the Deal
Dan at Armor Games was kind enough to allow me to post the details of the primary sponsorship (with aforementioned number-fudging, of course) -- not that thereís much to say. As I mentioned, Dan was very lenient with his terms. They are as follows:
- Up-front payment of $25,000 for the game
- Bonus of $2,000 for providing a set of bonus missions exclusive to Armor Games
- Requirement to implement the Armor Games AGI (high score interface)
- One-week exclusivity period on armorgames.com ad-free
- Following that, one week of viral-version distribution, with ads allowed
- Following that, secondary sales/sitelocks/etc. can be sold, with ads allowed
Dan didnít send me a big contract to sign; he was happy exchanging a few emails confirming the details. He has been very flexible and takes every suggestion on a case-by-case basis. Iíve gotten many exceptions out of him, and in return for his generosity I gave him a longer exclusivity period than he asked for. These relationships are two-way streets, and nothing is set in stone.
Revenues to Date
It's about time for a money summary.
ďOtherĒ is the combined forces of CPMStar (around $300) and Mochi ($10) advertising, as well as some award/performance money. As you can see, my near-zero ad estimates were fairly spot-on.
As a grand total, the game has earned approximately $34,000 so far, which is also fairly spot-on to my initial estimate. Minus FlashGameLicense and the SteamBirds Team, that puts $15,000 in my pocket for one month of full-time work. Even if you spread it out across my full schedule -- around five months to date -- thatís still $3,000 per month, a decent salary.
Keep in mind that the ad revenue will continue growing as time passes and might end up making a dent long-term. There are a few more secondary licensing deals to be made as well.
The game has done surprisingly well. I wonít go into the details on the reviews
and good comments
Iíve gotten, as Iíve discussed them previously, but here are some of the average scores around the internet. Flash developers seem obsessed with these numbers, but I donít see the allure.
- 87% @ ArmorGames.com (8.7/10)
- 86% @ Kongregate.com (4.29/5), before Badges were introduced
- 84% @ Kongregate.com (4.19/5), after Badges were introduced
- 88% @ Newgrounds.com (4.4/5), ratings
- 94% @ Newgrounds.com (9.4/10), reviews
- 91% @ FlashGameLicense.com (9.1/10), developer rating
- 80% @ FlashGameLicense.com (8/10), editor rating
- 75% @ FlashGameLicense.com (7.5/10), first impressions
The game has also won a few distinctions so far:
- Daily 4th Place: Newgrounds.com
- Weekly 6th Place: Newgrounds.com
- Front-Page Feature on Kongregate.com
- Weekly Prize Winner on Kongregate.com
- Monthly Prize winner on Kongregate.com
Game Performance Tracking
Huge shouts-out are due to Ben at SWFStats
. His very-easy-to-implement API is quite awesome and Iím very impressed with it. His data aligns perfectly with my data on other collection services, so I can vouch for its stability and authenticity. Right now itís tracking everything for me. His beta is available now and free to use.
First, an overview of the gameís traffic:
As of this writing, here are some stats for the game at a glance:
- 3,500,000 game loads (people that make it to the main menu of the game; doesnít count bounces)
- 24,000,000 level plays
- 26:10 average total play time
- 1,000 sites hosting the game
A statistically insignificant number of players play the game for less than two minutes. The official stats say there is a 100 percent player retention rate. That means if you see my title screen, you are playing two levels on average.
- 53 percent of users play more than five levels.
- Only 1.3 percent of players hit the mute button (without un-muting it later)
- Only 3 percent of players look at the credits screen
Some per-site traffic stats:
- ArmorGames.com has the majority of my traffic.
- Kongregate.com is a distant second place, with fewer than one third of the views
- A Chinese site pirated the game on day four, breaking my sitelock, and is third place so far, although by a wide margin
- All the rest of my 1,000 sites have fewer than 50,000 views each, with most having very low figures
[This article was first published on Andy Moore's personal blog.]