Share
Embed
How to Incorporate UX Research into the Product Development Process
Kaho Tajima (Yahoo! JAPAN / Flea Market Div, YAHUOKU! Services Group, Commerce Group / Designer)
https://tech-verse.me/ja/sessions/208
https://tech-verse.me/en/sessions/208
https://tech-verse.me/ko/sessions/208
Tech-Verse2022
November 18, 2022
Transcript
96 slides
-
Kaho Tajimaɹ Yahoo! Japan Corporation •UI/UX designer of "PayPay Flea
Market" •Design Team Manager and Head of Design, Overall Service •Outside of the company, he belongs to Designship, a general incorporated association. -
Product timeline iOS App Android App Launch Web Launch PayPay
Mini App Launch Service launch 5% commission 10 Million DL 15 Million DL 2019/10 2022/10 2021/1 2021/10 2020/9 5 million DL ! " ! ! 2020/1 2020/4 2020/7 ZOZOTOWN alignment 2022/6 WEAR alignment -
How do we foster a culture of UX research and
connect it to product development? What’s going to be talked about today For people like • I want to incorporate UX research! I want to try it! • I want to know specifically what kind of UX research you do in product development. -
Table of Contents - Case Study 1: Loyalty Survey -
Case study 2: Persona development - PayPay Flea Market development flow and perspective on issues - UX Research Approach -
UX research, as a major premise, is a method of
UX design and by testing and repeating the hypotheses we aim to provide better services and experiences to users -
UX research, as a major premise, is a method of
UX design and by testing and repeating the hypotheses we aim to provide better services and experiences to users -
UX Research Methodology Interview Usability Test Survey (e.g. questionnaire) Qualitative
Research Quantitative Research etc… -
How well do you know the users of the product
you are developing? -
Frequently heard in UX research Isn't the interview just one
person's opinion? How effective is user research for your business? Doesn't research take a long time? -
In reality, survey and effects are intricately intertwined. Effect A
Survey C Effect B Effect D Survey A Survey B Effect C -
• Outputs are not prioritized ahead of the investigation. •
Do research only when necessary. Are you bound by expectations to do this? Survey A Effect A -
How well do you know the users of the product
you are developing? -
How do you try to get to know your friends
and family? -
Repeated UX research increases user resolution. Survey Effect A Survey
C Effect B Effect D Survey A Survey B Effect C -
Repeated UX research increases user resolution. Survey Effect A Survey
C Effect B Effect D Survey A Survey B Effect C Effect Creating a strong effect by increasing user resolution -
The culture of always having some kind of UX research
going on is valuable in itself. -
How do we foster a culture of UX research and
connect it to product development? What’s going to be talked about today For people like • I want to incorporate UX research! I want to try it! • I want to know specifically what kind of UX research you do in product development. -
Design PayPay Flea Development System Plan FE BE Release once
every two weeks PJ1 Assign a job category to each feature development project Product Management iOS Android Web BFF BE SRE PJ2 PJ3 PJ4 v1.x.x -
Design PayPay Flea Development System Plan FE BE Release once
every two weeks PJ1 Assign a job category to each feature development project Product Management iOS Android Web BFF BE SRE PJ2 PJ3 PJ4 v1.x.x -
PayPay Flea Development Flow The task Definition Feature Specifications Definition
Information Design UI Design -
PayPay Flea Development Flow The task Definition Feature Specifications Definition
Information Design Development UI Design -
PayPay Flea Development Flow The task Definition Feature Specifications Definition
Information Design Development QA UI Design -
PayPay Flea Development Flow The task Definition Feature Specifications Definition
Information Design Development QA Release UI Design -
What features are of value to users? Discuss this and
decide with PdM The task Definition Feature Specifications Definition Information Design UI Design Development QA Release PayPay Flea Development Flow -
What features are of value to users? Discuss this and
decide with PdM Design with Figma Communicate with engineers and have them develop accordingly The task Definition Feature Specifications Definition Information Design UI Design Development QA Release PayPay Flea Development Flow -
want to hear from users want to provide it to
users as soon as possible -
• Outputs are not prioritized ahead of the investigation. •
Do research only when necessary. Issuesᶄ Not many people could do UX research.It was a bad pattern situation -
UX research cycle Start by fostering a culture of regular
UX research Research was conducted at regular intervals without worrying about the timing of feature development and to be reflected upon later PJ1 PJ2 PJ3 PJ4 PJ5 PJ6 PJ7 PJ8 v1.x.x Release v1.x.x Release v1.x.x Release -
UX research cycle Start by fostering a culture of regular
UX research Research was conducted at regular intervals without worrying about the timing of feature development and to be reflected upon later PJ1 PJ2 PJ3 PJ4 PJ5 PJ6 PJ7 PJ8 v1.x.x Release v1.x.x Release v1.x.x Release reserch reserch -
Development of persona of the existing users Qualitative survey Loyalty
Survey Quantitative survey Today's case study Research from both quantitative and qualitative perspectives -
Development of persona of the existing users Qualitative survey Loyalty
Survey Quantitative survey Today's case study Research from both quantitative and qualitative perspectives -
Why was it conducted? Match everyone's user image. By developing
the persona of the existing users we align the common understanding -
PayPay Flea Market existing users outside the company were interviewed
on actual usage, and a survey is conducted -
PayPay Flea Market existing users outside the company were interviewed
on actual usage, and a survey is conducted The qualitative data obtained from the interviews are analyzed, and personas were created -
Grouping of subjects 1 Axis A 2 4 3 1
2 4 3 1 2 4 3 ɾ ɾ ɾ Axis B Axis C -
Grouping of subjects 1 Axis A 2 4 3 1
2 4 3 1 2 4 3 ɾ ɾ ɾ Axis B Axis C -
Add persona to the specification format Change the flow to
revisit when deciding the specifications The task Definition Feature Specifications Definition Information Design UI Design Development QA Release -
• Interview external users about actual usage • Create a
persona based on the qualitative data obtained from the interviews • Make time to check the persona in the development flow to try to establish the persona How to Result -
Development of persona of the existing users Qualitative survey Loyalty
Survey Quantitative survey Today's case study Research from both quantitative and qualitative perspectives -
Why was it conducted? NPS: An index that measures customer
loyalty such as "trust" and "attachment" to services It's unclear what works for NPS and there's nothing I can do about it. -
Enable fixed-point observation c onduct at regular intervals To be
based on psychological loyalty methods Structure and quantify in 3 layers Ҿ༻ɿ͓٬༷ͷ৺Λ͔ͭΉ ৺ཧϩΠϠϧςΟϚʔέςΟϯάɹ෦ ߂ؽ ஶɹਡ๚ ྑ म -
Enable fixed-point observation c onduct at regular intervals To be
based on psychological loyalty methods Structure and quantify in 3 layers Measure at each touchpoint Make loyalty visible Ҿ༻ɿ͓٬༷ͷ৺Λ͔ͭΉ ৺ཧϩΠϠϧςΟϚʔέςΟϯάɹ෦ ߂ؽ ஶɹਡ๚ ྑ म -
Positive Experiences Negative Experiences Positive Experiences Negative Experiences Conduct a
survey on NPS across services Survey with a questionnaire asking about item-wise satisfaction level Itemize experiences that seem to have a strong impact on satisfaction -
Positive Experiences Negative Experiences Positive Experiences Negative Experiences Conduct a
survey on NPS across services Survey with a questionnaire asking about item-wise satisfaction level ɾɾɾ Touchpoint A Touchpoint B Touchpoint C Itemize experiences that seem to have a strong impact on satisfaction -
Positive Experiences Negative Experiences Positive Experiences Negative Experiences Conduct a
survey on NPS across services Satisfaction in the mind Satisfaction in the heart Survey with a questionnaire asking about item-wise satisfaction level ɾɾɾ Touchpoint A Touchpoint B Touchpoint C Itemize experiences that seem to have a strong impact on satisfaction -
Satisfaction level for each touchpoint No translation Experiences that influence
the satisfaction Overall NPS -
Satisfaction level for each touchpoint No translation Experiences that influence
the satisfaction Overall NPS -
Satisfaction level for each touchpoint No translation Experiences that influence
the satisfaction Overall NPS -
Satisfaction level for each touchpoint No translation Experiences that influence
the satisfaction Overall NPS -
Satisfaction level for each touchpoint No translation Experiences that influence
the satisfaction Overall NPS -
Satisfaction level for each touchpoint No translation Experiences that influence
the satisfaction Overall NPS -
Mapping Map the measurement results and prioritize improvements Satisfaction in
the heart is high Satisfaction in the mind is low Satisfaction in the heart is low Satisfaction in the mind is high High Satisfaction in the heart Satisfaction in the mind High Low -
Mapping Map the measurement results and prioritize improvements Satisfaction in
the heart is high Satisfaction in the mind is low Satisfaction in the heart is low Satisfaction in the mind is high High Satisfaction in the heart Satisfaction in the mind Reduce the negative experiences (At the same time, increase the positive experiences) High Low -
Mapping Map the measurement results and prioritize improvements Satisfaction in
the heart is high Satisfaction in the mind is low Satisfaction in the heart is low Satisfaction in the mind is high High Satisfaction in the heart Satisfaction in the mind Reduce the negative experiences (At the same time, increase the positive experiences) Reduce the positive experiences (Increase the negative experiences at the same time) High Low -
Improving experiences that seem to have a significant impact on
loyalty Aggregate results Sum up by dividing into two segments, sellers and buyers -
Improving experiences that seem to have a significant impact on
loyalty Aggregate results Sum up by dividing into two segments, sellers and buyers -
Improving experiences that seem to have a significant impact on
loyalty Aggregate results Sum up by dividing into two segments, sellers and buyers -
There were many positive experiences, such as, “auto-input was very
helpful”. Conversely, comments like, “there were many items to be entered in the listing form, and I gave up halfway through”, expressed the negative experiences. Until the listing is completed -
There were many positive experiences, such as, “auto-input was very
helpful”. Conversely, comments like, “there were many items to be entered in the listing form, and I gave up halfway through”, expressed the negative experiences. Until the listing is completed Enhance auto-input and reduce input cost -
Increase the priority of planned measures for improvement Payment towards
sale proceeds Many selected the positive experience item, “I was happy that the sales proceeds were reflected so quickly.” -
Increase the priority of planned measures for improvement Payment towards
sale proceeds Many selected the positive experience item, “I was happy that the sales proceeds were reflected so quickly.” To increase the speed of reflection Take steps to obtain valuable evaluation from the buyer at the time of payment towards sale proceeds -
• Structure and quantify in 3 layers, and measure at
each touchpoint • Visualize touchpoints where negative and positive experiences are likely to have a significant impact • Straightforward Improvement proposals • To review the priority of existing scheduled projects, use this as a tool How to Result -
Development of persona of the existing users Qualitative survey Loyalty
Survey Quantitative survey Today's case study Research from both quantitative and qualitative perspectives -
UX research cycle Start by fostering a culture of regular
UX research reserch Research was conducted at regular intervals without worrying about the timing of feature development and to be reflected upon later PJ1 PJ2 PJ3 PJ4 PJ5 PJ6 PJ7 PJ8 v1.x.x Release v1.x.x Release v1.x.x Release reserch -
The culture of always having some kind of UX research
going on is valuable in itself.