SlideShare a Scribd company logo
Studying the Dialogue Between Users and
Developers of Free Apps in the Google Play
Store
Journal First Presentation | Empirical Software Engineering
Ahmed E.
Hassan
Safwat
Hassan
Cor-Paul
Bezemer
Chakkrit (Kla)
Tantithamthavor
n
We study user-developer interactions through the
distribution and rating mechanisms of the Google Play
Store
2
How developers leverage the distribution
mechanism to publish emergency updates
We study user-developer interactions through the
distribution and rating mechanisms of the Google Play
Store
3
How developers leverage the distribution
mechanism to publish emergency updates
The dialogue between users and developers
Mobile app stores offer rich data
4
The Google Play Store has around
3.6 million apps
March 2, 2018
Need to fix this update, it keeps on freezing.
User
For many years, developers were not able to
respond to user reviews
5
October 14, 2015
After listening to one podcast, sometimes I want to go back
and listen to it again. Where can I find the old shows that I
have already listened to?
User
Since recently, developers can engage in a
dialogue with the reviewer
6
October 14, 2015
After listening to one podcast, sometimes I want to go back
and listen to it again. Where can I find the old shows that I
have already listened to?
User
The app doesn't delete episode articles. If you don't see
them it means that you pressed on the toolbar 'eye' button
to hide read content. Press once more to display them.
Dev
Since recently, developers can engage in a
dialogue with the reviewer
7
October 14, 2015
After listening to one podcast, sometimes I want to go back
and listen to it again. Where can I find the old shows that I
have already listened to?
User
October 16, 2015
After talking to support, my questions were answered. It's a
good app
User
The app doesn't delete episode articles. If you don't see
them it means that you pressed on the toolbar 'eye' button
to hide read content. Press once more to display them.
Dev
Analyzing the dialogue between users and
developers
8
2) What are the common patterns of developer
responses?
1) What is the impact of responding to user
reviews on the review rating?
3) What drives a developer to respond?
A summary of the studied dataset
9
> 2K
Apps
> 355K
Changes
in reviews
> 128K
Developer
responses
> 4M
Reviews
There is a great benefit in
responding to reviews
10
Six times more likely to lead to
a rating increase
There is a great benefit in
responding to reviews
11
In 84% of the cases of rating
increase, users increase their rating
to four stars or five stars
Six times more likely to lead to
a rating increase
We identified three main reasons for rating
increase
12
Developer guides the user to
solve the reported issue without
having to deploy an app update
34
%
We identified three main reasons for rating
increase
13
Developer guides the user to
solve the reported issue without
having to deploy an app update
34
%
24
%
Developer deploys an update to
address the reported issue
We identified three main reasons for rating
increase
14
13
%
Developer guides the user to
solve the reported issue without
having to deploy an app update
34
%
24
%
Developer deploys an update to
address the reported issue
Details of the solution are
communicated outside the store
Analyzing the dialogue between users and
developers
15
2) What are the common patterns of developer
responses?
1) What is the impact of responding to user
reviews on the review rating?
3) What drives a developer to respond?
Our approach for identifying the common
patterns of developer responses
16
Step 1:
Collect
metrics
We collected 6 metrics:
Review title length, Review text length, Days since last
release, Review rating, Positive sentiment, Negative
sentiment.
Our approach for identifying the common
patterns of developer responses
17
Step 2:
Build
model for
every app
Step 1:
Collect
metrics
We built 415 models.
Our approach for identifying the common
patterns of developer responses
18
Step 2:
Build
model for
every app
Step 3:
Extract the
key
features
for each
model
Step 1:
Collect
metrics
We extracted 12 key features for each model.
Our approach for identifying the common
patterns of developer responses
19
Step 2:
Build
model for
every app
Step 3:
Extract the
key
features
for each
model
Step 4:
Cluster the
models
Step 1:
Collect
metrics
We identified three patterns of developer
responses
20
Only negative reviews
We identified three patterns of developer
responses
21
Negative or longer reviews
Only negative reviews
We identified three patterns of developer
responses
22
Negative or longer reviews
Only negative reviews
Reviews which are posted shortly after the
latest release
Analyzing the dialogue between users and
developers
23
2) What are the common patterns of developer
responses?
1) What is the impact of responding to user
reviews on the review rating?
3) What drives a developer to respond?
Our approach for studying what drives a
developer to respond
24
• We manually read a statistically
representative random sample of user-
developer interaction episodes.
• We study the contents of responses to
understand better what drives developers
to respond to reviews
We identified four main drivers for
responding
25
Thank the user63%
We identified four main drivers for
responding
26
Thank the user63%
Ask for more details45%
We identified four main drivers for
responding
27
Provide guidance25%
Thank the user63%
Ask for more details45%
We identified four main drivers for
responding
28
Provide guidance25%
Thank the user63%
Ask for more details45%
Ask for endorsement24%
We identified four main drivers for
responding
29
Provide guidance25%
Thank the user63%
Ask for more details45%
Ask for endorsement24%
For provide guidance, we observed that
similar responses can be used to provide
FAQs
30
AppLock
The user asks how to use the
“AppLock” app to lock other apps
“Please open phone settings security
apps with usage access enable AppLock”
User
Dev
520
For provide guidance, we observed that
similar responses can be used to provide
FAQs
31
PicsArt
The user complains that the app is
very slow
“Hi Vincent this issue can sometimes be
solved by clearing the cache. To do so
go to your device’s Settings - Apps -
PicsArt and tap ‘Clear data’ and ‘Clear
cache’
User
Dev
271
AppLock
The user asks how to use the
“AppLock” app to lock other apps
“Please open phone settings security
apps with usage access enable AppLock”
User
Dev
520
32
33
34
35
36
Safwat Hassan
shassan@cs.queensu.ca

More Related Content

PDF
Studying the Dialogue Between Users and Developers of Free Apps in the Google...
PDF
Studying the Integration Practices and the Evolution of Ad Libraries in the G...
PDF
Studying online distribution platforms for games through the mining of data f...
PDF
Seene iOS app - social media buzz analysis
PDF
SURF: Summarizer of User Reviews Feedback
PDF
Recommending and localizing change requests for mobile apps based on user rev...
PPTX
Social Media Workshop - Internet Monitoring - Jakarta 2014
PDF
Reactions to Snapchat update - social media analysis
Studying the Dialogue Between Users and Developers of Free Apps in the Google...
Studying the Integration Practices and the Evolution of Ad Libraries in the G...
Studying online distribution platforms for games through the mining of data f...
Seene iOS app - social media buzz analysis
SURF: Summarizer of User Reviews Feedback
Recommending and localizing change requests for mobile apps based on user rev...
Social Media Workshop - Internet Monitoring - Jakarta 2014
Reactions to Snapchat update - social media analysis

What's hot (7)

PPTX
Customer support for mobile apps
PDF
Sonatype's 2013 OSS Software Survey
PPTX
Live 2014 Survey Results: Open Source Development and Application Security Su...
PPTX
Slingshot App by Facebook - Reactions across Social Media
PPTX
User Sentiment to Determine App Quality
PPTX
Hyperlapse - the new app by Instagram for time-lapse videos shooting - social...
PDF
When a machine learning researcher and a software engineer walk into a bar
Customer support for mobile apps
Sonatype's 2013 OSS Software Survey
Live 2014 Survey Results: Open Source Development and Application Security Su...
Slingshot App by Facebook - Reactions across Social Media
User Sentiment to Determine App Quality
Hyperlapse - the new app by Instagram for time-lapse videos shooting - social...
When a machine learning researcher and a software engineer walk into a bar
Ad

Similar to Studying the Dialogue Between Users and Developers of Free Apps in the Google Play Store (20)

PDF
Studying User-Developer Interactions Through the Distribution and Reviewing M...
PDF
ADVANCEMENT IN THE MOBILE APP REVIEW SYSTEM TO ENHANCE QUALITY OF MOBILE APPL...
PDF
How To Leverage App Store Reviews API For Unmatched App Performance.pdf
PDF
How Can I Improve My App? Classifying User Reviews for Software Maintenance a...
PPTX
Large-Scale Empirical Studies of Mobile Apps
PDF
What is AppFollow
PPTX
Exploratory Analysis On Play Store Apps.pptx
PDF
From Android App to Killer App - How to Reach the Million-Downloads Milestone
PDF
The Journey from User Acquisition to Advocacy - CleverTap & AppFollow
PDF
Tug of Perspectives: Mobile App Users vs Developers
PPTX
Growth Hacking- Organic and Paid App Installs
PDF
Ratings and reviews in the App Store and Google Play
PDF
Investigating country differences in mobile app user behavior and challenges ...
PPTX
GH_Final1.1
PDF
Acquire Customers at the Lowest Cost
PDF
How to get 30k+ App Store reviews every month
PDF
2015 app reviewer survey: the secrets of how to get your app reviewed
PPTX
App Ratings & Reviews: How to improve them and why it matters
PPTX
Apptentive at IndieDevLab - Grow your App Business
PDF
Empirical analysis on iOS app Popularity
Studying User-Developer Interactions Through the Distribution and Reviewing M...
ADVANCEMENT IN THE MOBILE APP REVIEW SYSTEM TO ENHANCE QUALITY OF MOBILE APPL...
How To Leverage App Store Reviews API For Unmatched App Performance.pdf
How Can I Improve My App? Classifying User Reviews for Software Maintenance a...
Large-Scale Empirical Studies of Mobile Apps
What is AppFollow
Exploratory Analysis On Play Store Apps.pptx
From Android App to Killer App - How to Reach the Million-Downloads Milestone
The Journey from User Acquisition to Advocacy - CleverTap & AppFollow
Tug of Perspectives: Mobile App Users vs Developers
Growth Hacking- Organic and Paid App Installs
Ratings and reviews in the App Store and Google Play
Investigating country differences in mobile app user behavior and challenges ...
GH_Final1.1
Acquire Customers at the Lowest Cost
How to get 30k+ App Store reviews every month
2015 app reviewer survey: the secrets of how to get your app reviewed
App Ratings & Reviews: How to improve them and why it matters
Apptentive at IndieDevLab - Grow your App Business
Empirical analysis on iOS app Popularity
Ad

More from SAIL_QU (20)

PDF
Studying the Integration Practices and the Evolution of Ad Libraries in the G...
PPTX
Improving the testing efficiency of selenium-based load tests
PPTX
Understanding the Factors for Fast Answers in Technical Q&A Websites: An Empi...
PDF
Investigating the Challenges in Selenium Usage and Improving the Testing Effi...
PDF
Mining Development Knowledge to Understand and Support Software Logging Pract...
PPTX
Which Log Level Should Developers Choose For a New Logging Statement?
PPTX
Towards Just-in-Time Suggestions for Log Changes
PDF
The Impact of Task Granularity on Co-evolution Analyses
PPTX
A Framework for Evaluating the Results of the SZZ Approach for Identifying Bu...
PPTX
How are Discussions Associated with Bug Reworking? An Empirical Study on Open...
PPTX
A Study of the Relation of Mobile Device Attributes with the User-Perceived Q...
PDF
A Large-Scale Study of the Impact of Feature Selection Techniques on Defect C...
PDF
What Do Programmers Know about Software Energy Consumption?
PPTX
Threshold for Size and Complexity Metrics: A Case Study from the Perspective ...
PDF
Revisiting the Experimental Design Choices for Approaches for the Automated R...
PPTX
Measuring Program Comprehension: A Large-Scale Field Study with Professionals
PPTX
On the Unreliability of Bug Severity Data
PPT
On the Link Between Mobile App Quality and User Reviews
PDF
Mining Software Engineering Data
PPTX
The Impact of Using Regression Models to Build Defect Classifiers
Studying the Integration Practices and the Evolution of Ad Libraries in the G...
Improving the testing efficiency of selenium-based load tests
Understanding the Factors for Fast Answers in Technical Q&A Websites: An Empi...
Investigating the Challenges in Selenium Usage and Improving the Testing Effi...
Mining Development Knowledge to Understand and Support Software Logging Pract...
Which Log Level Should Developers Choose For a New Logging Statement?
Towards Just-in-Time Suggestions for Log Changes
The Impact of Task Granularity on Co-evolution Analyses
A Framework for Evaluating the Results of the SZZ Approach for Identifying Bu...
How are Discussions Associated with Bug Reworking? An Empirical Study on Open...
A Study of the Relation of Mobile Device Attributes with the User-Perceived Q...
A Large-Scale Study of the Impact of Feature Selection Techniques on Defect C...
What Do Programmers Know about Software Energy Consumption?
Threshold for Size and Complexity Metrics: A Case Study from the Perspective ...
Revisiting the Experimental Design Choices for Approaches for the Automated R...
Measuring Program Comprehension: A Large-Scale Field Study with Professionals
On the Unreliability of Bug Severity Data
On the Link Between Mobile App Quality and User Reviews
Mining Software Engineering Data
The Impact of Using Regression Models to Build Defect Classifiers

Recently uploaded (20)

PPTX
Oracle E-Business Suite: A Comprehensive Guide for Modern Enterprises
PPT
Introduction Database Management System for Course Database
PDF
System and Network Administration Chapter 2
PDF
Audit Checklist Design Aligning with ISO, IATF, and Industry Standards — Omne...
PDF
Navsoft: AI-Powered Business Solutions & Custom Software Development
PDF
Nekopoi APK 2025 free lastest update
PDF
Understanding Forklifts - TECH EHS Solution
PDF
2025 Textile ERP Trends: SAP, Odoo & Oracle
PDF
Flood Susceptibility Mapping Using Image-Based 2D-CNN Deep Learnin. Overview ...
PDF
Claude Code: Everyone is a 10x Developer - A Comprehensive AI-Powered CLI Tool
PPTX
ai tools demonstartion for schools and inter college
PDF
Internet Downloader Manager (IDM) Crack 6.42 Build 42 Updates Latest 2025
PPTX
CHAPTER 12 - CYBER SECURITY AND FUTURE SKILLS (1) (1).pptx
PPTX
Introduction to Artificial Intelligence
PPTX
Operating system designcfffgfgggggggvggggggggg
PDF
medical staffing services at VALiNTRY
PDF
Which alternative to Crystal Reports is best for small or large businesses.pdf
PPTX
CHAPTER 2 - PM Management and IT Context
PDF
T3DD25 TYPO3 Content Blocks - Deep Dive by André Kraus
PDF
PTS Company Brochure 2025 (1).pdf.......
Oracle E-Business Suite: A Comprehensive Guide for Modern Enterprises
Introduction Database Management System for Course Database
System and Network Administration Chapter 2
Audit Checklist Design Aligning with ISO, IATF, and Industry Standards — Omne...
Navsoft: AI-Powered Business Solutions & Custom Software Development
Nekopoi APK 2025 free lastest update
Understanding Forklifts - TECH EHS Solution
2025 Textile ERP Trends: SAP, Odoo & Oracle
Flood Susceptibility Mapping Using Image-Based 2D-CNN Deep Learnin. Overview ...
Claude Code: Everyone is a 10x Developer - A Comprehensive AI-Powered CLI Tool
ai tools demonstartion for schools and inter college
Internet Downloader Manager (IDM) Crack 6.42 Build 42 Updates Latest 2025
CHAPTER 12 - CYBER SECURITY AND FUTURE SKILLS (1) (1).pptx
Introduction to Artificial Intelligence
Operating system designcfffgfgggggggvggggggggg
medical staffing services at VALiNTRY
Which alternative to Crystal Reports is best for small or large businesses.pdf
CHAPTER 2 - PM Management and IT Context
T3DD25 TYPO3 Content Blocks - Deep Dive by André Kraus
PTS Company Brochure 2025 (1).pdf.......

Studying the Dialogue Between Users and Developers of Free Apps in the Google Play Store

  • 1. Studying the Dialogue Between Users and Developers of Free Apps in the Google Play Store Journal First Presentation | Empirical Software Engineering Ahmed E. Hassan Safwat Hassan Cor-Paul Bezemer Chakkrit (Kla) Tantithamthavor n
  • 2. We study user-developer interactions through the distribution and rating mechanisms of the Google Play Store 2 How developers leverage the distribution mechanism to publish emergency updates
  • 3. We study user-developer interactions through the distribution and rating mechanisms of the Google Play Store 3 How developers leverage the distribution mechanism to publish emergency updates The dialogue between users and developers
  • 4. Mobile app stores offer rich data 4 The Google Play Store has around 3.6 million apps March 2, 2018 Need to fix this update, it keeps on freezing. User
  • 5. For many years, developers were not able to respond to user reviews 5 October 14, 2015 After listening to one podcast, sometimes I want to go back and listen to it again. Where can I find the old shows that I have already listened to? User
  • 6. Since recently, developers can engage in a dialogue with the reviewer 6 October 14, 2015 After listening to one podcast, sometimes I want to go back and listen to it again. Where can I find the old shows that I have already listened to? User The app doesn't delete episode articles. If you don't see them it means that you pressed on the toolbar 'eye' button to hide read content. Press once more to display them. Dev
  • 7. Since recently, developers can engage in a dialogue with the reviewer 7 October 14, 2015 After listening to one podcast, sometimes I want to go back and listen to it again. Where can I find the old shows that I have already listened to? User October 16, 2015 After talking to support, my questions were answered. It's a good app User The app doesn't delete episode articles. If you don't see them it means that you pressed on the toolbar 'eye' button to hide read content. Press once more to display them. Dev
  • 8. Analyzing the dialogue between users and developers 8 2) What are the common patterns of developer responses? 1) What is the impact of responding to user reviews on the review rating? 3) What drives a developer to respond?
  • 9. A summary of the studied dataset 9 > 2K Apps > 355K Changes in reviews > 128K Developer responses > 4M Reviews
  • 10. There is a great benefit in responding to reviews 10 Six times more likely to lead to a rating increase
  • 11. There is a great benefit in responding to reviews 11 In 84% of the cases of rating increase, users increase their rating to four stars or five stars Six times more likely to lead to a rating increase
  • 12. We identified three main reasons for rating increase 12 Developer guides the user to solve the reported issue without having to deploy an app update 34 %
  • 13. We identified three main reasons for rating increase 13 Developer guides the user to solve the reported issue without having to deploy an app update 34 % 24 % Developer deploys an update to address the reported issue
  • 14. We identified three main reasons for rating increase 14 13 % Developer guides the user to solve the reported issue without having to deploy an app update 34 % 24 % Developer deploys an update to address the reported issue Details of the solution are communicated outside the store
  • 15. Analyzing the dialogue between users and developers 15 2) What are the common patterns of developer responses? 1) What is the impact of responding to user reviews on the review rating? 3) What drives a developer to respond?
  • 16. Our approach for identifying the common patterns of developer responses 16 Step 1: Collect metrics We collected 6 metrics: Review title length, Review text length, Days since last release, Review rating, Positive sentiment, Negative sentiment.
  • 17. Our approach for identifying the common patterns of developer responses 17 Step 2: Build model for every app Step 1: Collect metrics We built 415 models.
  • 18. Our approach for identifying the common patterns of developer responses 18 Step 2: Build model for every app Step 3: Extract the key features for each model Step 1: Collect metrics We extracted 12 key features for each model.
  • 19. Our approach for identifying the common patterns of developer responses 19 Step 2: Build model for every app Step 3: Extract the key features for each model Step 4: Cluster the models Step 1: Collect metrics
  • 20. We identified three patterns of developer responses 20 Only negative reviews
  • 21. We identified three patterns of developer responses 21 Negative or longer reviews Only negative reviews
  • 22. We identified three patterns of developer responses 22 Negative or longer reviews Only negative reviews Reviews which are posted shortly after the latest release
  • 23. Analyzing the dialogue between users and developers 23 2) What are the common patterns of developer responses? 1) What is the impact of responding to user reviews on the review rating? 3) What drives a developer to respond?
  • 24. Our approach for studying what drives a developer to respond 24 • We manually read a statistically representative random sample of user- developer interaction episodes. • We study the contents of responses to understand better what drives developers to respond to reviews
  • 25. We identified four main drivers for responding 25 Thank the user63%
  • 26. We identified four main drivers for responding 26 Thank the user63% Ask for more details45%
  • 27. We identified four main drivers for responding 27 Provide guidance25% Thank the user63% Ask for more details45%
  • 28. We identified four main drivers for responding 28 Provide guidance25% Thank the user63% Ask for more details45% Ask for endorsement24%
  • 29. We identified four main drivers for responding 29 Provide guidance25% Thank the user63% Ask for more details45% Ask for endorsement24%
  • 30. For provide guidance, we observed that similar responses can be used to provide FAQs 30 AppLock The user asks how to use the “AppLock” app to lock other apps “Please open phone settings security apps with usage access enable AppLock” User Dev 520
  • 31. For provide guidance, we observed that similar responses can be used to provide FAQs 31 PicsArt The user complains that the app is very slow “Hi Vincent this issue can sometimes be solved by clearing the cache. To do so go to your device’s Settings - Apps - PicsArt and tap ‘Clear data’ and ‘Clear cache’ User Dev 271 AppLock The user asks how to use the “AppLock” app to lock other apps “Please open phone settings security apps with usage access enable AppLock” User Dev 520
  • 32. 32
  • 33. 33
  • 34. 34
  • 35. 35

Editor's Notes

  • #11: Decrease = 942 (942/11813) = 8% In 84% of the increase cases users increase the rating to 4 or 5 stars.
  • #12: Decrease = 942 (942/11813) = 8% In 84% of the increase cases users increase the rating to 4 or 5 stars.
  • #21: 1  56% 2)  23% 3)  12% 4)  11%
  • #22: 1  56% 2)  23% 3)  12% 4)  11%
  • #23: 1  56% 2)  23% 3)  12% 4)  11%
  • #25: dom sample of 347 responses from the selected 3,431 responses
  • #29: Only in 3 out of 83 developer responses, users change their review after a developer asks for a rating increase and in only two cases the rating actually increases.  2%
  • #31: Check the alignment
  • #32: Check the alignment
  • #33: Add your name and contact email
  • #34: Add your name and contact email
  • #35: Add your name and contact email
  • #36: Add your name and contact email
  • #37: Add your name and contact email