Data driven decision making: Avoiding pitfalls and building value08.04.2012, Quo Vadis, Dr. Stefan Behrens

1

Aeria Games and ProSiebenSat.1 Games have combined forces on April 1, 2014

Industry leading monetization

Unparalleled marketing power

Strong developer relationships

Best-in-class operations

Global communities & reach

Top 3 player in

Europe

&

2

Some facts on the combined business

Registered Users

Countries Serviced

Game Licenses Operated

Mobile Share of Revenues

35

77m

16%

39

&

3

Data driven decision-making effectively drives almost all aspects of our business at Aeria

Talent Product Acquisition Retention Monetization

Values

4

At Aeria, decision making starts with strong values

VALUES

5

Hiring the right talent is the basis for success

• Don’t hire (only) gamers

TALENT

• Put them through „boot camp“

• Let them crunch numbers

• Keep setting the bar higher & higher

6

Picking the right games is not an easy feat

PRODUCT

Target Group Fit

Committed Developer

Proven Success

Adequate Art Style

7

Lifetime Value

(365d ARPU)

Customer Acquisition Funnel

Buy & SpendAeria Points

(7d Conversion)

Return toPlay

(3d Retention)

GameLogin

(Reg-to-Play)

Landing PageRegistration

(Click-to-Reg)

Click onAd Banner

(CTR)

CTR: 0.23%

Clicks: 107,000

Click-to-Reg: 5.6%

Regs: 6,000

Reg-to-Play 33%

Players: 2,000

3d Retention: 45%

Retained: 900

7d Conversion: 8.5%

Spenders: 170 $14.09

Max CPC:

0.13 USD

Max CPL:

2.35 USD

Max CPP:

7.05 USD

Marketing: Maximum Customer Cost Calculation

Successful acquisition is about balancing a simple, but fundamental equation...

ACQUISITION

8

…but the equation can get mighty complex

$-

$5

$10

$15

$20

$25

Jan-

12

Feb-

12

Mar

-12

Apr

-12

May

-12

Jun-

12

Jul-1

2

Aug

-12

Sep

-12

Oct

-12

Nov

-12

Dec

-12

Jan-

13

Feb-

13

Mar

-13

Apr

-13

May

-13

30d LTV (a) 90d LTV (a)

1yr LTV (a) 1yr LTV (e)

Actual Predicted

Predictive LTV ModelingMulti-Channel Optimization

Search22%

Display/AdNetworks

38%

Branding1%

Facebook4%

PR5%

TV30%

ACQUISITION

9

Retention rates can tell you a lot…

0%

10%

20%

30%

40%

50%

60%

70%

29

.09

.0

2.1

0.

05

.10

.0

8.1

0.

11

.10

.1

4.1

0.

17

.10

.2

0.1

0.

23

.10

.2

6.1

0.

30

.10

.0

2.1

1.

05

.11

.0

8.1

1.

11

.11

.1

4.1

1.

17

.11

.2

0.1

1.

23

.11

.2

6.1

1.

29

.11

.0

2.1

2.

05

.12

.0

8.1

2.

11

.12

.1

4.1

2.

RR 3 days organic RR 3 days paid

Game Case: 3 Day Retention Rate (RR)

RETENTION

10

…if you know how to read them

0%

10%

20%

30%

40%

50%

60%

70%

29

.09

.0

2.1

0.

05

.10

.0

8.1

0.

11

.10

.1

4.1

0.

17

.10

.2

0.1

0.

23

.10

.2

6.1

0.

30

.10

.0

2.1

1.

05

.11

.0

8.1

1.

11

.11

.1

4.1

1.

17

.11

.2

0.1

1.

23

.11

.2

6.1

1.

29

.11

.0

2.1

2.

05

.12

.0

8.1

2.

11

.12

.1

4.1

2.

RR 3 days organic RR 3 days paid

Game Case: 3 Day Retention Rate (RR)

RETENTION

Technical issuewith a patch

Secondary accountsof existing users

11

However, sometimes, the answers areless than obvious

30%

35%

40%

45%

50%

55%

60%

6.1

.20

12

6.3

.20

12

6.5

.20

12

6.7

.201

2

6.9

.20

12

6.1

1.2

01

2

6.1

3.20

12

6.1

5.20

12

6.1

7.20

12

6.1

9.20

12

6.2

1.20

12

6.2

3.20

12

6.2

5.20

12

6.2

7.20

12

6.2

9.20

12

7.1

.20

12

7.3

.20

12

7.5

.20

12

7.7

.20

12

7.9

.20

12

7.1

1.20

12

7.1

3.20

12

7.1

5.20

12

Germany

UK

Game Case: 3 Day Retention Rate

RETENTION

12

Especially, when soccer is involved

30%

35%

40%

45%

50%

55%

60%

6.1

.20

12

6.3

.20

12

6.5

.20

12

6.7

.201

2

6.9

.20

12

6.1

1.2

01

2

6.1

3.20

12

6.1

5.20

12

6.1

7.20

12

6.1

9.20

12

6.2

1.20

12

6.2

3.20

12

6.2

5.20

12

6.2

7.20

12

6.2

9.20

12

7.1

.20

12

7.3

.20

12

7.5

.20

12

7.7

.20

12

7.9

.20

12

7.1

1.20

12

7.1

3.20

12

7.1

5.20

12

Germany

UKEuro Cup!

Game Case: 3 Day Retention Rate

RETENTION

13

Good sales strategy is fueled by analytics

Monthly Product Sales PlanContinuous KPI Analysis

1[0-5)

2[5-50)

3[50-100)4[100-200)5[200-400)6[400+)

0.00%

10.00%

20.00%

30.00%

40.00%

50.00%

60.00%

70.00%

0None

1[0-5)

2[5-50)

3[50-100)

4[100-200)

5[200-400)

6[400+)

1[0-5)

2[5-50)

3[50-100)

4[100-200)

5[200-400)

6[400+)

Item Saturation

TS Redemption

CCU

Spender Buckets

Retention

MONETIZATION

14

Season #1 Price: $12 per spin No. of Spins: 16 No. of Jackpots: 1,500

Season #2 Price: $25 per spin No. of Spins: 14 Number of Jackpots 1000 Optimized composition of weapons

Season #3 Price: $20 per spin No. of Spins: 20 Number of Jackpots 500+500 Optimized weapons and Jackpot

Shooter Game Case, Revenue 2013 Jun-Dec Event Optimization

EventSeason #1

EventSeason #2

EventSeason #3

Execute, analyze, optimize for best results

Jun Jul NovOctSep DecAug

MONETIZATION

15

Beware… It’s a trap! (at least sometimes)

Data consistency can be a b*tch

Correlation ≠ Causation

What get’s measured can be improved,but what about intangibles?

You can plan a pretty picnic,but you can’t predict the weather!

PITFALLS

Data driven decision making: Avoiding pitfalls and building value08.04.2014, Quo Vadis, Dr. Stefan Behrens