Double Buffered

I just got out of an interesting talk by Fernando Paiz of Turbine entitled The Future of MMO Monetization: How Turbine Supercharged DDO by Adopting a Hybrid Business Model. The talk was basically a postmortem on the conversion of Dungeons and Dragons Online into a hybrid free-to-play model, and Fernando actually gave out some nice details. Here’s my notes, and I’ll add a link to slides once their up:

DDO Unlimited

• What went right with DDO: Handcraft RPG quests, fast combat, faithful adaptation of D&D rules and rich character customization
• What needed improvement: Hard to solo, was only 10-16 levels of content, no open world, no pvp, fairly hardcore. Basically it didn’t match with what players expected a subscription-based MMO to be. Felt like it’s own thing.
• DDO Unlimited: A Premium Free to play MMO. The idea behind free to play is that you can increase total revenue by allowing players to spend as much or little as they want. This provides a compelling marketing message that resonated, especially because of how rare it was in the US market. Also, it increases ease of referral and allows players to play at their own pace. Player choice.
• But, they kept the subscriptions. First, it maintains the flat fee option for the western players who really enjoy that paymsent system. We like all you can eat. Also, by providing a subscription option it allowed them to price content higher as players who objected to high a la carte would use subscription instead. Also, players could switch back and forth easily, so no existing player was offended by the switch.
• There were lots of naysayers “hybrid doesn’t work” “subscribers will quit” “they can’t co-exist”. It turns out they were wrong. DDO subscribers and free to play player coexisted very happily. Free players provided value for the subscribers and vice versa.

Implementing Free

• The game had to be really free. If you make it “half free” the players will figure it out and be irritated. It’s safer to launch with more toll gates in place, because removing them makes players happy while adding them is basically impossible. They eventually ended up removing “level tokens” because it made it not free enough. They didn’t sell end-game loot because they wanted to preserve the value of their in-game economy.
• Dual currency design is key. It limits the monetization of the in-game economy. They had success in offering larger bundled points at a better rate, the min-maxer in the audience means they bought more points total and ended up increasing revenue. Also, you can run sales on points and individual items.
• Balancing subscribers and free players is key. Each subscriber had full access to content, and was also given a small stipend of points to spend on consumables/services. They targeted the value of the subscription plan to equal $200 of a la carte content, which was about the lifetime value of a subscription plan. Players felt good with that pricing. Players could downgrade from subscription to free, they would keep progress but lose access. Characters with inacessible classes would be greyed out. Got to pick which characters to keep. There was no revolt  by the players over subscription vs. free value.

What to Sell

• You want to sell things that improve game play and make it more fun, but are not required to be fun. Free players got hirelings, but for pay you can get MORE hirelings. You need to be careful about what is per account and what is per character. In general, as long as you leave in old subscription model, you can sell anything you want to free players without revolt.
• Content was sold per account, as it was a “permanent” purchase. You can buy guest passes to allow teammates to do content. Classes, races, and early quest packs sold very well. This started out being a small % of revenue (because free players were still working through free content) but is now #1 source of revenue
• Convenience is good to sell per character. For instance, sell increased bag space per character, because they are willing to buy for individual characters. Shared bank to allow transfer between characters on account was very successful, and they flagged many bind on equip items to actually be bind on account. Also, they already had a “favor” reward system that allowed cashing in of in-game currency for access to cool things like Drow and 32-point character build. At launch they refrained from selling this, but players wanted it. So, they sold these favor rewards at a high price ($20 even), and players who already had it didn’t revolt, they just felt more valuable. Could still get it normally. For anything sold in the store, it has to be bind on character to avoid credit card fraud.
• Consumables are another key source of revenue. But, they never sold any named or rare items. Low level equipment was a popular because it helped early game levelling. Stat tomes of +1 were sold, but these were also found in game and the higher (+3,+4) tomes were NOT sold. These higher tomes would override the +1s. Temporary XP/loot boosts, packs of healing potions (purely as a convenience as they are purchaseable in game for not much), and resurrection items were popular.
• They admitted to not having done enough for Cosmetic items and Services. They did have success in selling reincarnation, which is a super respec. Cosmetic items were very limited and not high sellers.

Launch Results

• The press response at launch was amazing. Was called “accessible” despite being a hardcore rpg, purely because of the monetization model. Automatically became casual friendly.
• 2 Million new players since launch. 20% reacquisition of cancelled players. Acquisition costs drastically reduced, with 3x click through on ads. 5x Peak concurrent. 10x Montly uniques. 5x revenue. Even better is still growing, last month was best yet.
• Store was a big success. Was a web-based UI, and they put a lot of effort into it. Invested heavily in Customer Support to deal with shop issues. 70% of items in cart ended up going to checkout. > 20% of total players bought items. Even better, > 50% of subscribers ended up using store. Sales were effective. You have to promote and support your store throughout your entire system.
• By units, top sellers are resurrection cakes, experience boosts, potion packs, sigil of levelling before they cancelled it, loot bosts, and at the very end equipment
• By revenue, top sellers are the 32-point build (which allowed slightly better new characters), xp boost, +2 to all stats tome, Drow and favored soul races, character slots, monk
• Demographics shifted. Audience moved younger and more casual, both in play style and hours. But changes were relatively minor, stil a high hours per week game.

Conclusions

• If subscribers are a revenue baseline, the free to play customer averages to 70% the revenue/player of a subscription player. However, post launch the subscription + micro user averaged to 175% revenue/player under the old system.
• Prior to change the revenue/peak concurrency ratio (which corresponds to a revenue/operating cost ratio) was between $75 to $120. Post relaunch it lowered to between $65 and $90 revenue/peak concurrency. So, the revenue/peak went down a bit, but the peak concurrency went up 5x.
• Basically, the relaunch worked. Content monetization was a success, and the subscriber base was happy and ended up being more profitable. Doing a beta with real money was a good idea, they refunded the points post beta and let them spend it twice. But many spent a lot in beta.
• Some issues at launch. They eventually added some “retroactive points” for existing subscribers. They could have done a better job tuning for acquisition. They didn’t convert the gameplay as much as they would have liked to be casual.  The LFG system still isn’t great because it doesnt deal with teammates who lack content.
• The keys for others: make the joining process frictionless and easy. Reduce friction in shop. Focus on customer service in shop. Breadt of microtransactions is key.
• If they were to do it in the future, they would design their game specifically for the hybrid model.