pexels-photo-76931.jpeg

Prezent

PREZENT

 
Prezent is an app to make the experience of selecting a gift for a person fun and less stressful.

Overview

Timeline: August 2017 - December 2017 (CS 6755 Introduction to HCI Foundations, Georgia Tech)
Team members: Meghan Galanif | Ethan Graves | Nikhila Nyapathy | Maria Wong
Tools: Sketch | Axure
My role:
UX Research - Recruited participants, conducted 5 semi-structured interviews, contributed to survey design and deployed the same, analyzed data through an affinity diagram, contributed to brainstorming system functionality and design criteria
UX Design - Participated in the brainstorming and ideation process, sketched two iterative versions of the selected systems, created sketches and wireframes of the social media scanner, collected feedback for the sketches, mocked up certain screens, contributed to usability criteria specifications
Evaluation - Performed roles of notetaker, error counter for 3 usability benchmarking tests and conducted 1 cognitive walkthrough

 
 

Problem context

The prompt given to us was to pick a user group in the retail space, and zero in on a specific problem area. We chose the space of gift-giving, through our research findings, we focused our goal on assisting individuals in choosing a well-informed gift for someone whom they have limited information regarding.

Research

Disclaimer: All the research was conducted with a handful of participants and is not representative of the entire user group.
We initially started with the idea of gift-giving for kids, especially by individuals who are unfamiliar with kids. We conducted research to collect broad information on gift shopping patterns of people, and to truly understand the trends and difficulties people encountered in the process, for both kids and adults.

                                       Questions we asked over the course of our initial research

                                       Questions we asked over the course of our initial research

Surveys
We deployed a survey for which we got over 200 responses. It branched based on whom they most recently shopped for (kid or adult). 

Interviews
We conducted interviews with 14 individuals, and asked them a set of questions in 2 different scenarios, for when they last shopped for a kid and for an adult. We used an affinity diagram for analysis. The key takeaways were the commonalities amongst the desired gift attributes (eg: wanting the gift to be unique and sentimental) and the goals different people wished to achieve with their gift (eg: wanting their gift to be useful). Some were specific to shopping for children (eg: wanting the gift to be educational). Further insights included how information about the receiver known to the person influenced their choices, and how they went about finding new information.

 

Click for high resolution.

 
In_store.png

In-store visits
Based on the insights we received from the interviews and survey, we decided to experience and learn firsthand about the in-store context of shopping for children. We conducted two in-store walkthroughs - Target and Toys R Us. We found that while there was a lack of information in-store for buyers to rely on, the ability to visually browse a variety of items seemed to be an appealing process. This was backed up by anecdotal data from the interviews.

Online shopping methods
From the interviews and survey, Amazon.com was the preferred choice of platform due to the variety, reviews and fast shipping. Analyzing the website showed various options such as themed categories (eg: Kids' birthdays), age-wise filtering as well as 'Best Toys of the Month'. Even after filtering there were an overwhelming number of options. A drawback was that it didn't give the users options of touching and feeling before picking out a gift.

Legacy and existing systems
Research on legacy systems showed that there were attempts made to leverage Facebook 'likes' to find out gift ideas for a person

Task analysis
Having abstracted the fundamental tasks involved in the gift-giving process as well as decisions the user has to make before coming to a conclusion, we created a user flow as shown below.

 
 

Key Insights
From our research we determined the following:-

  • People did not find gift-giving for kids as stressful as they did for adults. This led to us pivoting to include a larger user group, but this time, focused on gift-giving in general and not specifically for kids. This was surprising to us as we had gone in thinking shopping for kids would cause more stress - and this is where conducting interviews with people regarding shopping for both kids and adults helped.

  • One of the biggest pain points was gifting someone for whom they had limited information regarding - it was surprising to us that people cared a lot when in such a situation; despite not being close to the person, they were still sensitive about giving the right type of gift which was appropriate. They wished to know more about the person, but they did not want to directly ask the person - either in the spirit of surprise or basically uneasiness in asking directly.

  • Shopping styles were varied according to the convenience level it presented for the user. However, in a lot of the 'last minute' gift shopping experiences, we found that they leaned towards in-store shopping due to the lack of time. They did also like the 'touch and feel' aspect to in-store shopping. The choices presented in stores and online sites alike were described as overwhelming by some users.


There were certain key channels that we identified as being part of a successful gifting experience:-

Key elements for a successful gift giving experience

Key elements for a successful gift giving experience

Drawing from the above findings, we aimed to provide the following functionality:

1. Provide the user information about the gift receiver’s preferences.
2. Allow the user to input the information they already know about the receiver.
3. Keep the user anonymous to the gift receiver.
4. Assist users in making an informed decision.
5. Aid the users in making faster decisions.

Design criteria

design_criteria

User group & Personas
Once we made the pivot, we defined our user group as customers who have limited knowledge about the people they shop for, but still genuinely care about purchasing a gift that matches their preferences. Some social contexts which matched this (but not limited to) were:

  • Users who had decreased contact with certain friends circles

  • Users new to the corporate world and were purchasing gifts for their coworkers

  • Users who had newly joined a family

  • Users who have moved to a new place

  • Users who are newly financially independent

We built our personas to cover such situations and also across different shopping styles and goals as informed by our research.

 
 
persona_cole
 
 

Ideation

Brainstorming
For our "informed brainstorming" session, we came up with a wide variety of ideas for systems to provide a way to collect information about the receiver of the gift and help the gift buyer come up with ideas for the gift. Broadly, the ideas were categorized on the basis of whether the goal was collecting data on the preferences, or providing gift recommendations, or a hybrid of both. Our top three ideas were:

1. Social media image recognition - Analyzing data from social media (images, hashtags, places) to build a profile of the preferences
2. Anonymous Messaging - Discreetly asking the gift receiver simple questions about their general preferences
3. Virtual Assistant - An in-store assistant to help the user figure out what a good gift for the occasion would be based on available information
In our first round of ideation, we went for a divergent approach, by sketching out individually each of these ideas. We then discussed strengths, weaknesses and inspiration for each of the sketches as a group.

iteration_1
We pinned up the drawings and used sticky notes to mark positives, negatives and new ideas.

We pinned up the drawings and used sticky notes to mark positives, negatives and new ideas.

 
 

Refined ideas
Social media scanner
This was developed to leverage the wide presence people have on social media. Key features included being able to connect different social media platforms, adding and viewing your friends from the various platforms, choosing what type of data (images, videos, hashtags) to scan which generate gift ideas backed by analysis results.

social_media_storyboard_2

Anonymous messaging
The idea behind this application was for a user to find out the gift receiver's preferences by anonymously messaging them. It would be a two-way application, similar to other social media. A user is prompted to frequently maintain their own profile by adding/removing 'tiles', or images of possible interests which are randomly generated.

Virtual Assistant
This developed out of the need for a more informative in-store experience, as users were turning to brick and mortar stores whenever they were running short on time. The intention was to have a kiosk which would suggest gift ideas combining whatever information the user had about the receiver along with the store's database of customers' buying trends and reviews. Another feature was to locate the item in the store for a more efficient experience.

We asked different people what their thoughts were on the 3 ideas, what they liked and disliked.

 
feedback
 
Icons made by Eucalyp from www.flaticon.com is licensed by CC 3.0 BY

Final designs
 

Final_sketches
evaluation_methods


Key findings

  • Users would trust the system more if information that validates the recommendations is provided

  • Users would prefer to have more detailed information about gift ideas over general descriptions

  • More incentive needs to be provided for users to add their interests to their profiles

  • Users want to be able to complete a purchase in order to have the full gift-giving experience

  • Experts suggested we change the color scheme to have better contrast, and making the icons and fonts more consistent with the real world

  • Experts also suggested we increase the information provided to improve learnability

The average weighted SUS score measured was 86.5, with a standard deviation of 7.52 and a median of 87.5.

In Conclusion: Takeaways

  • Pivoting may be necessary, depending on user research insights. We thought the main problem was shopping for kids, but quickly came to discover through interviews and surveys that people found that easy as kids are easily satisfied as opposed to shopping for adults with more specific interests.

  • There is a trade-off between privacy and the 'surprise' element in a gift. Our solutions which did not involve the receiver's participation were viewed as potentially creepy even though it used the social media information that they has voluntarily put up. At the same time, involving the receiver may lessen the surprise element, and it's important to strike a balance.

  • It is important for users to have a sense of agency. From our research, we found that some users wanted specific gift suggestions, while others wanted general gift ideas so that they can make the final decision, adding a layer of personalization to the gift.

  • We must make the system trustworthy. Any decisions taken by the system need to be backed up by data, which needs to be made transparent to the user to build their trust in the system.