1. whirled metrics
1.1. overview
1.1.1. http://bit.ly/Dx5S
1.1.2. Whirled
1.1.2.1. three currencies
1.1.2.2. aggressive repricing needed
1.1.2.2.1. user's generating content do this well at Whirled
1.1.2.3. money booking unclear
1.1.3. Puzzle Pirates
1.1.3.1. has item decay
1.1.3.2. as no money payout, book the money as it comes in
1.1.3.2.1. no liabilities
1.1.3.3. players value their time at 25 cents / hour
1.2. spending / earning
1.2.1. http://bit.ly/npcD
1.2.2. at first gave out too much money out
1.2.2.1. changed this
1.2.2.2. try to balance the in and out flows
1.3. creators / purchases
1.3.1. http://bit.ly/L7U4
1.4. item listings by type
1.4.1. http://bit.ly/LI0F
1.5. item purchases by type
1.5.1. http://bit.ly/180BJ
1.5.2. by money spent
1.5.3. avatars, furniture & toys
1.6. item purchase metrics
1.6.1. http://bit.ly/c0mw
1.6.2. total coins spent
1.6.2.1. shows low end and high end products all earn well
1.7. purchases vs sales
1.7.1. http://bit.ly/TT8H
1.7.2. puzzle pirates
1.7.3. coin shop purchase
1.7.3.1. privilege items
1.7.3.1.1. e.g. "captain's badge"
1.8. item sales distribution
1.8.1. http://bit.ly/wrEq
1.8.2. different view on the same data as in previous slide
1.8.3. 160-170 k$ a month in product sales
1.8.3.1. mostly vanity items
2. billing payment breakdown
2.1. http://bit.ly/85HP
2.2. some figures are obfuscated due to confidentiality agreemetns
2.3. strategic
2.3.1. SMS
2.3.2. Target card
2.4. use somebody else's infrastructure
2.4.1. TwoFish
2.4.2. FatFoogoo
2.4.3. PlaySpan
2.4.4. GlobalCollect
2.5. if you go for kids & want lots of small payments
2.5.1. must have prepaid card
2.6. older people with big onetime payments
2.6.1. credit cards
2.7. fraud
2.7.1. http://bit.ly/vVKw
2.7.2. always do a refund
2.7.2.1. otherwise problems with credit card companies
2.7.3. verify big purchases
2.7.3.1. e.g. fax in signatures
3. Q&A
3.1. sample model of virtual goods LTV
3.1.1. http://bit.ly/virtual-goods-ltv
3.2. search vs. display
3.2.1. people who are searching for you (in some way) will convert a lot better
3.2.1.1. use google adwords for search
3.3. analytics tools
3.3.1. typically own built tools
3.3.1.1. lots of data, hard engineering problem
3.3.2. facebook specific
3.3.2.1. SoMetrics
3.3.2.2. developer analytics
3.3.2.3. Cotigent??
4. Presenters
4.1. Daniel James
4.1.1. Three Rings
4.2. Andrew Chen
5. intro & terms
5.1. some key terms explained
5.1.1. e.g. LTV = life time value
5.2. introduction to Whirled
5.2.1. rooms, items, trophies etc
5.2.2. soft currency = earned with gameplay
5.2.3. hard currency = bought with real money
5.2.4. third currency = earned by content creators when content is sold for hard currency
5.3. Puzzle Pirates
5.3.1. roughly 4 M$ revenue
5.3.2. started in 2005
5.3.3. male/female 50/50
5.3.4. wide age range (7-70 year olds)
5.3.5. key metrics
5.3.5.1. see flickr photo of slide
5.3.5.1.1. http://bit.ly/3tqbUT
5.3.5.2. conversion rates vary a lot of depending of the source of the user
5.3.5.3. top 10% of paying customers generate 50% of revenue
6. what factors drive LTV
6.1. games are rare as they make money from end customers directly
6.1.1. so you can actually buy traffic
6.2. see flickr photo of slide
6.2.1. http://bit.ly/XOX2
6.3. cost per acquisition
6.3.1. the key figure to understand
6.3.2. factors that drive it, flickr photo
6.3.2.1. http://bit.ly/T0lp
6.3.2.2. ad networks
6.3.2.2.1. plenty of sources
6.3.2.3. user requirements are crucial
6.3.2.3.1. installs may drop conversion from 20% to low single digits
6.3.2.3.2. flash trumps java hands down
6.3.2.4. audience/theme: horizontal vs vertical
6.3.2.4.1. horizontal (e.g. a socnet) works fine
6.3.2.4.2. vertical (niche) challenges in finding in audience, buying ads, scaling
6.3.2.5. funnel design
6.3.2.5.1. must be testing all the time
6.3.2.5.2. all steps
6.3.2.6. viral marketing
6.3.2.6.1. gotta have it, especially for Facebook games
6.3.2.6.2. potential to really drive down your cost / acquisition
6.3.2.6.3. not just growth, focus on retention
6.3.2.7. A/B testing
6.3.2.7.1. pretty much the most important thing
6.3.2.7.2. test your hypotheses
6.4. funnel
6.4.1. understand this and optimize it
6.5. profit = LTV - cost/acquisition
7. customer retention metrics
7.1. sample dashboard
7.1.1. http://bit.ly/16W0K
7.2. if your LTV > cost/acquisition, just go and buy traffic
7.3. puzzle pirates
7.3.1. CPA campaigns in march
7.3.1.1. looks like the traffic wasn't really worth anything
7.3.1.2. e.g. $4 CPA led to $1 LTV
7.3.2. have weekly figures for all metrics
7.4. retention
7.4.1. http://bit.ly/11Haa
7.4.2. cohort analysis
7.4.2.1. how fast do customers drop off
7.4.2.2. http://bit.ly/12WpG
7.4.2.3. at Whirled haven't done a lot to prompt retention
7.4.2.3.1. e.g. missing notifications
7.4.3. revisit rates
7.4.3.1. track
7.4.3.2. day-by-day basis
7.4.3.3. how many times do they come back
7.4.3.4. factors that drive it
7.4.3.4.1. http://bit.ly/P9C9
7.4.3.4.2. product depth crucial
7.4.3.4.3. community
7.5. conversion / retention
7.5.1. http://bit.ly/aiE3