This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.
Front page of Kongregate! W00t!
Although most of our players never notice, Defender's Quest is a Flash game. We compiled it with Adobe AIR, which let us employ some tricks to get the standard "desktop quality" game features, like fullscreen resolution switching, to work. We also used Flixel, which gave us great performance (for a Flash game) by circumventing Flash's normal, unoptimized, display logic. Using Flash made it easy to develop a browser demo, which we could distribute through flash portals.
As I mentioned in my previous article, rather than paying to host the demo on our own servers, we uploaded it to Mochi Media, turned off advertisements, and embedded it on our site. Then, we enabled distribution, which means that Mochi would send the demo to lots of flash portals across the internet.
After seeing a truly astonishing number of visitors to our site from www.minijuegos.com of all places, we decided to go after the big boys -- Newgrounds, Kongregate, and Armor Games. We launched on Newgrounds and Kongregate, but Armor Games is invitation-only and we weren't able to reach an agreement with them.
Our plan was to create a special version of the browser demo for each major portal, implement each one's achievement APIs, and create a unique demo landing page for each offering a $2-off coupon. Since I had gotten complaints that our store's coupon field was hard to find, I made the "buy" button on each landing page pre-apply the coupon to the order. These coupons were worth more than our usual $1 off, but because we were afraid of "demo backlash" we figured the magic $5 impulse-buy price couldn't hurt.
We started with Kongregate and immediately had our worst fears confirmed. The game's page was quickly flooded with flames and one star reviews. Our score was around 3.4 and dropping steadily, which means our work would soon be consigned to video game oblivion. Many players complained they had been tricked, and that demos were unwelcome on free game sites.
We were also accused of being "greedy," "evil," and -- my favorite -- "illegal." Against my better judgment, I logged into my personal account and argued back, but my comments were swiftly down-rated by angry players, and my grumpy words (mercifully) disappeared in minutes.
Anthony acted much more rationally than I did. He jumped onto our developer account and instantly started damage control. He changed the name of the entry from "Defender's Quest Prelude" to "Defender's Quest (lengthy!) DEMO". He updated the description fields to say "THIS IS A DEMO!" right at the top, rather than tucked away after the game's feature list.
If people aren't expecting this, it makes them very angry!
He started responded courteously to each negative comment and within a day our review score had started turning around. Fans started buying the game, and a lot of them went to bat for us in the comments. After a few days we cracked the four star barrier, and were featured on the front page.
We're still getting the occasional one star hate comments (who doesn't?) lambasting us for being a demo, but the score climbed steadily. All of the metrics suggest Kongregate is overwhelmingly responsible for the majority of our total sales.
After things settled down, we got many comments like, "Glad you changed the description. I don't mind demos, I just wanna know what I'm getting up front." as well as, "I never buy based on a demo, but I bought this." So although our demo will never get a top rating, changing the context to manage expectations seems to have helped the perception.
A vocal minority will likely continue to down-rate us, but that comes with the territory.
We later launched on Newgrounds, using the experience from Kong to guide us. We made it absolutely clear in the title and description that the game was a demo, and included links to the RPS and Destructoid reviews of the game. Newgrounds received a similar $2-off coupon, just like Kongregate.
I also created some medals (achievements) for the game. Kongregate badges are all hand-crafted by Kong employees (mostly Greg McClanahan), but on Newgrounds you can create them yourself and submit them to Newgrounds chief Tom Fulp for approval. I created some for the usual stuff like beating levels and gaining party members, as well as a set of easy ones just for visiting the various option menus.
My hope was that players would spend the first five minutes of the game exploring the options menus to rack up some quick medals and discover all the options and customization in the game. Hopefully players likely to complain about "too much reading" would quickly see they can turn tutorials and dialogue off, and those demanding "more hotkeys" would discover that not only does the game have hotkeys, but they are all customizable. The entire game can be played with nothing but the keyboard should they so desire.
Newgrounds' feedback system is less noisy than Kong's, and oriented towards reviews rather than comments, so the reviews are dominated by players who have bought the full game. Furthermore, Kongregate's system only allows a user to vote once, but Newgrounds lets users vote every day, giving higher voting power to established users.
The upshot is that on Kong, the score depends more on the first impressions of a large audience, trends downwards over time, and is more vulnerable to one star hate-reviews. On Newgrounds a devoted fanbase of site regulars can gradually raise the score and easily offset drive-by down-ratings. Shortly after launching, our score on both sites was around 4.11. It's plateaued on Kong, but it's up to an unbelievable 4.40 on Newgrounds!
Shortly after I submitted my medals to Tom Fulp for approval, he responded to my message with great news. Not only had he approved the medals, he'd also placed the game on the front page! Unfortunately, Newgrounds went down for their long-delayed redesign on February 6, and when it came back up, we were just below the threshold for front page status.
Even so, coupons suggest we've made a lot of sales from Newgrounds -- even more than from Rock, Paper, Shotgun, making them our second-largest source of sales. Together, Newgrounds and Kongregate probably represent about 80 to 90 percent of our total revenue. This is a completely unexpected outcome; we thought a demo up-sell model was a long shot for Flash portals, and anticipated only modest sales.
People playing games on free Flash portals don't like spending money, and they also don't like going to another website to do so. Given we were asking them to do both, our success took me by surprise. We estimate that our demo-to-sales conversion rate is somewhere between 1 to 5 percent. Moving to kreds on Kongregate seems to have helped -- on-site currencies let players buy the game without going to another website or pulling out their credit cards.
Since we knew that many smaller flash portals "steal" game files from Kongregate and Newgrounds, we wanted the Kongregate version to work even on other sites. This is how Flash games spread virally. To be clear: we want portals to do this so that our demo gets more plays, and hopefully, more sales. The Kongregate version had special code that detected when it was running on a site other than kongregate.com, which would make it revert to the normal format, directing players to visit our website to buy the game.
So, that's everything we've done, and our sales numbers so far. I'm not prepared to draw broad conclusions as to what worked and what didn't in terms of driving sales. It's still a bit too early in the process for that. But I will say that I think there are some pretty clear spikes correlated with reviews and portal promotions. Our steady daily sales seem like they're coming from Kongregate, Newgrounds, and small flash portals.
Our last and most important strategy has been to focus on customer service. Anthony, James, and I all try to keep the backlog from building up. We do our best to stay abreast of glitches and compatibility problems, and to surmount the limitations of our less-than-perfect storefront. So far customer feedback has been positive.
We hope to move on to other stores like Desura and Impulse, etc, soon, but first we want to iron out the major bugs, solve compatibility glitches, and finish the bonus version so we can put our best foot forward on those platforms.
I hope all this information helps someone.