Doubt discussion platform
Doubt pe charcha
Now Every doubt will be resolved!
Project type
Feature Redesign
Company - Doubtnut
​
Project duration
2 months
(Jan 2022 - March 2022)
Role
My role was to
- understand the requirements from stakeholders and Product managers
- Analyse the existing problems ( User research and Data analysis) - Team effort
- Define the problems, ideate possible solutions and deliver designs and test.
.png)
What
Problem Statement
40% doubts asked by students daily remain unresolved on Doubtnut…
Doubtnut is a platform where students can ask their doubts by uploading question images from various sources and the OCR scans the questions and provide solutions from Doubtnut repository within seconds, but the match rate of the doubts asked by students on Doubtnut is 60-62% due to technical constraints, insufficient data in doubtnut's repository, questions in other languages that we do not cater so on and so forth.
Feature timeline
June 2021
Feature launch
October - Dec 2021
Metrics drop
Dec 2021
Feature revamp
In June 2021, Doubtnut launched a feature called " Doubt pe charcha" to solve the above stated problem. It is a platform where students can have an interpersonal communication with an agenda to resolve each others doubts that are not being resolved by Doubtnut's repository of solutions.
Initially the feature did really well in terms of its performance and improved retention as well. But slowly and steadily we saw a downfall in its performance. Room opening and helpers joining the P2P room have significantly dropped from September to December 2021
In December 2021, we planned to do a revamp based on our analysis and observations. We'll be looking at the revamp process further in this case study.
Analysing the problems in exiting version
We kickstarted with doing an analysis of the existing version of "doubt pe charcha" and finding out anomalies. Analysis done to find out insights and actionables are :
​
- Data Analysis
- Uxcam Analysis (User sessions, Heat maps, user flows etc)
- User Interviews
Current flow :
The process
Data Analysis ( June-Dec 2021)
​
How P2P feature is performing currently?
1. P2P room creation
- Room created daily has been reduced by 27%.
- Number of users who create rooms has decreased by 25% in the last 6 months.
2 P2P Room open
- P2P room open has been reduced by 50% (1.14 M to 527.5K)
- There are 5 entry points for P2P currently : SRP Page, P2P icon click, Notifications, Homepage widget, Menu sidebar (P2P icon). The major dip (70-80%) is seen in notifications and P2P icon click.
3. Helpers joining the room
- No of helpers joined has been reduced by 41%
- Percentage of rooms where helper joined has been reduced by 3%
4. Messages in the room
- Messages sent daily has decreased from 30k to 20k but message/room is almost constant.
5. Doubts solved
- We can’t track this from backend currently. Data is just based on feedback provided by user
- Total rooms where the user marked they got a solution has declined from 17% to 16%. Out of these 17% cases, 16% are where the user marked it on the same day.
Doubts asked daily - 13,892
Doubts solved daily - 2367 ( 17%)
Helpers joining daily - 6694 ( 48.18%)
Room where helper joined : 3000 ( 21.5%)
​
Helpers are joining rooms that are at the top of the list of doubts.
​
Solvers' profile
Who are one's who are solving doubts currently?
​
1. Class wise
10th and 12th class users are most active in solving doubts(19% and 23%). followed by 11th and IIT and NEET exam aspirants class (average 14% respectively) The percentage of users solving doubts in lower classes is less.
2. Locale wise
English (59%) and Hindi (37%) users are most active
3. Active hours
Helpers active hours : 7 to 9 pm
4. Top Helpers from boards :
UP Board (62.9%) > CBSE Board (16.9%) > Class 14 - IIT-JEE(16.5%)
& NEET(11.7%) > Other exams and Bihar board (10.69%)
5. Age wise
Old users (d120+) are solving most of the doubts
Uxcam Analysis
1. Timer issue
- When user asks a doubt, there is timer of 90 seconds and most of the users are waiting and not getting answers in 90 seconds which is frustrating.
- When askers are coming again on P2P page and checking if their doubt is resolved or not, they are again asked to wait for 90 seconds which is even more frustrating.
2. Rating popup revamp
- Users close P2P after being asked to rate multiple times in one session. They are being asked to rate even if they have not initiated any conversation in group.
3. FTUE flow
- Inconsistent flow : Only doubt community section is visible to users who have not asked doubts.
4. Doubt prioritisation
- Hi user is getting en doubts to solve and vice-versa.
5.Few groups (primarily new doubt groups) are being joined by multiple helpers while others are never joined by any user.
6. If doubts are not resolved by any helper, askers never get their solution
User Interviews
Study type : Remote Interviews
Participants : 15
Length : 30 minutes
Location - India, Remote
Themes and Insights
​
40% participants ask question on Doubt pe charcha when they don't receive an answer from Doubtnut
.png)
13/15 (86%) have asked question on P2P previously.
Discoverability issue : 9/15 (60%) participants dont solve because they dont know from where they can do it
100% (15/15) users say : Very few / no doubts are getting resolved via this feature
Why you don't ask on Doubt pe charcha?​
​​
-
a. 10/15 (73%) users say it takes a lot of time to receive a solution from Doubt pe charcha and sometimes, users never get a solution
-
b. 4/15 (24%) users say Waiting 90 seconds is making us frustrated.
-
c. 2/15 (13%) got a solution only
Are there any problems that you are facing while asking doubts on P2P? Please elaborate
-
a. 46% ( 7/15 ) waiting for 90 seconds is frustrating
-
b. 20% ( 3/15 )students don't know what to do after asking a question.
-
c. 13% (2/15) users say notifications are not locale specificNotifications are not locale specific
-
d. 6% (1/15) users find people chatting in groups
Are there any problems that you face while solving doubts?? Please elaborate
-
a. 66% (4/6) They ask to rate everytime while leaving a group even when there is no activity in groups
-
b. 50% ( 3/6) Notifications and questions are not locale specific.
-
c. 50% ( 3/6) Users find it difficult to search a desired question
-
d. 33% ( 2/6) There is no motivation for solvers
How satisfied are you with "doubt pe charcha" experience? Rate on a scale of 1-5, with 1 being not satisfied and 5 being completely satisfied.
-
20% (3/15) users rated 1
-
40% (6/15) rated 2 -
-
13.3% (2/15) rated 3
-
13% (2/15 )rated 4
-
13% (2/15) didn't rate
User Personas
Based on the user research, we found two types of personas
​
1. Asker : Who asks question on P2P
2. Solver : WHo solves question on P2P
User flow diagram
To understand and anticipate cognitive patterns of users, user flow was made to display the complete path a user takes when using the Doubt pe charcha feature
Feature prioritisation
We prioritised featured with the help of PRoduct managers and created versions and in order to split up in sprints and focused on the following problems in version 1
Rethinking Doubt pe charcha : AB HAR DOUBT HOGA SOLVE
This version targets to solve every doubt asked by students via a community of students
More and more helpers will be attracted via Gamification features like Earn Badges and rewards like DN Goodies on solving doubts. We’ll also educate them that they’ll be able to revise via helping others..
​
Helpers can filter doubts based on language, medium and class of their choice.
It also aims to solve the problem of not getting a reliable confirmation from asker’s end to know whether the doubt has been solved or not.
Then
Auto generated messages to tell the current status
Meanwhile doubt askers can explore the feature and help others solving their doubts.
Now
Track the status of your doubt
Then : Right after posting the doubt in student community, askers were asked to wait for 90 seconds till the helper joins. It used to be so frustrating for the asker to wait for 90 seconds and still not getting their solution as on an average in 80% cases helper was not joining in 90 seconds.
Now : Askers will get notified about the current status of their doubt and will be given an estimated time of 24 hrs ( calculated from data analysis) via auto generated messages by which they are most likely to get their solution.
Provide feedback
Then : Earlier every user who used to enter the chatroom has to give rating even after not sending a message or trying to solve a doubt. This journey wasn't right.
Now : We defined the set of users whom we'll ask to give rating. Only askers who got a solution from a helper will be given an option to provide rating on backpress. They'll rate the quality of an answer.
Solvers can redeem amazing rewards by solving doubs
Then : There was no way to track whether solver has solved a doubt or not. Users come on the chat room and share few messages and leave. There was
Now : We have introduced a cta to record the number of clicks users intend to solve a doubt and take confirmation from askers whether they got a solution
Gamification
Then : There was no motivation factor for helpers to come and solve doubts again and again.
Now : We introduced profile badges and rewards that they'll be getting on reaching certain levels. We gamified the experience to attract helpers.
Success metrics
1. P2P room creation
- Room created daily has been increased by 59%
2 P2P Room open
- P2P room open has been increased by 77%
3. Helpers joining the room
- No of helpers joined has been increased by 89%
4. Doubts solved
56% Doubts were being solved which was 17% before
​
5. Messages sent per room daily has been increased by 16%
​
Upcoming version : Adding internal and external teachers of Doubtnut - coming soon
After checking the success metrics we have planed to roll out this feature for doubtnut teachers also, so that more and more doubts can be solved and also to generate monitisation.
This version targets to solve every doubt asked by students via a community of students & Teachers (Internal + external)
Students will get quality solutions directly from teachers. They can chat with teachers or even get on a one to one call and close the session when they are satisfied.
​
Teachers can earn money by solving doubts
Doubtnut can save answers given by teachers in repository. It will also help in increasing business credibility and improve retention in longer run.
Learnings
We should always keep a check on numbers. What's working today for the audience can become obsolete tomorrow. ​
​
Satisfaction is the enemy of creativity. Try to brainstorm in all possible directions
​
Iterations are a part of design. So, one should never settle. Discussion, brainstorm and iterate