Top Banner
Data scientists need not apply: How anyone can do game analysis Casual Connect, November 2014 @allisonbilas | [email protected]
22

Data Scientists Need not apply

Feb 20, 2017

Download

Data & Analytics

GameAnalytics
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Data Scientists Need not apply

Data scientists need not apply:How anyone can do game analysis

Casual Connect, November 2014

@allisonbilas | [email protected]

Page 2: Data Scientists Need not apply

Using data to understand your players and monetizers.

Page 3: Data Scientists Need not apply
Page 4: Data Scientists Need not apply

Analyst

Jr. Producer Producer Monetization Mgr.

Developer Senior PM

Page 5: Data Scientists Need not apply

Descriptive Exploratory Inferential Predictive Causal

Quantitatively describing data

Looking for previously unknown

relationships in data

Testing theories with a sample of

data

Analyzing current events to predict

future events

Measuring what happens to one

variable when you change another

Distribution; 5-number summary;

Before/after

Directionality; Visualizations

Regression models; Chi squared

Modeling, machine learning, data

miningA/B testing

OBSERVE EXPERIMENT

Page 6: Data Scientists Need not apply

Descriptive Exploratory Inferential Predictive Causal

Quantitatively describing data

Looking for previously unknown

relationships in data

Testing theories with a sample of

data

Analyzing current events to predict

future events

Measuring what happens to one

variable when you change another

Distribution; 5-number summary;

Before/after

Directionality; Visualizations

Regression models; Chi squared

Modeling, machine learning, data

miningA/B testing

“EASY”

Page 7: Data Scientists Need not apply

‣ Quantitatively describing data

‣ Distribution of data; 5-number summary; Outliers

Score Percentiles by Boost Used

0

28

56

84

112

140

10. 25. Median 75. 90. 99.

Valentine WinterPrism RubyBlaze BurstStone No Boost

Descriptive

Page 8: Data Scientists Need not apply

‣ Looking for previously unknown relationships in data

‣ Visualizations are key component

ExploratoryMonetizers by Amount Spent per Customer

0

1,000

2,000

3,000

4,000

$0 - $5 $5 -$10 $10 - $25 $25 - $60 $60 - $280 $280 - $2975

Total Bookings by Amount Spent per Customer

$0

$25,000

$50,000

$75,000

$100,000

$0 - $5 $5 -$10 $10 - $25 $25 - $60 $60 - $280 $280 - $2975

4% 6%

16%

22%

35%

17%

34%

18%24%

15%

9%

1%

Page 9: Data Scientists Need not apply

‣ Measuring uncertainty in analysis

‣ Linear regression; Chi squaredInferential

Page 10: Data Scientists Need not apply

Three Big Questions

INSTALLS How many people have installed my

game?

DAU How many people are

playing my game?

REVENUE How much money is my game generating?

Acquisition Engagement Monetization

Page 11: Data Scientists Need not apply

Really good analysis…

Page 12: Data Scientists Need not apply

Revenue declined by $2K (-5%) week-over-week after the re-engagement campaign ended.

This was caused by a 3% decline in DAU, which in turn reduced the number of monetizers by 3%.

Those that did spend continued to spend at the same ARPPU of $7.29.

Reporting Analysis

DAU = 1.25M

Revenue = $41.4K

ARPPU = $7.29

Conversion = 0.45%

Page 13: Data Scientists Need not apply

Revenue declined by $2K (-5%) week-over-week after the re-engagement campaign ended.

This was caused by a 3% decline in DAU, which in turn reduced the number of monetizers by 3%.

Those that did spend continued to spend at the same ARPPU of $7.29.

Reporting Analysis

DAU = 1.25M

Revenue = $41.4K

ARPPU = $7.29

Conversion = 0.45%

Page 14: Data Scientists Need not apply

Revenue declined by $2K (-5%) week-over-week after the re-engagement campaign ended.

This was caused by a 3% decline in DAU, which in turn reduced the number of monetizers by 3%.

Those that did spend continued to spend at the same ARPPU of $7.29.

Reporting Analysis

DAU = 1.25M

Revenue = $41.4K

ARPPU = $7.29

Conversion = 0.45%

Page 15: Data Scientists Need not apply

Revenue declined by $2K (-5%) week-over-week after the re-engagement campaign ended.

This was caused by a 3% decline in DAU, which in turn reduced the number of monetizers by 3%.

Those that did spend continued to spend at the same ARPPU of $7.29.

Reporting Analysis

DAU = 1.25M

Revenue = $41.4K

ARPPU = $7.29

Conversion = 0.45%

Page 16: Data Scientists Need not apply

DAU = 1.25M

Revenue = $41.4K

ARPPU = $7.29

Conversion = 0.45%

Revenue declined by $2K (-5%) week-over-week after the re-engagement campaign ended.

This was caused by a 3% decline in DAU, which in turn reduced the number of monetizers by 3%.

Those that did spend continued to spend at the same ARPPU of $7.29.

Reporting Analysis

Page 17: Data Scientists Need not apply

‣ Continue to use push notifications to re-engage users and support DAU

‣ Prioritize dev work for daily prizes to increase return rates

‣ Consider creating a high priced gem package to increase ARPPU

Recommendations

Page 18: Data Scientists Need not apply
Page 19: Data Scientists Need not apply

You do see, you just don’t observe.

Sherlock H.

Page 20: Data Scientists Need not apply

Allison Bilas, VP of Product

@allisonbilas

[email protected]

Additional Reading: https://bitly.com/bundles/sparklingallison/

We’re hiring!

Page 21: Data Scientists Need not apply

UNESCO Game Design Challenge

$100K Prize

Incubate a game on

peace, sustainable development

and global citizenship.

http://mgiep.unesco.org/

Page 22: Data Scientists Need not apply

BEST CHART EVER!