Designing e-commerce mobile app & website for a boba shop

Dias Boba

UI / Website / Mobile App / E-Commerce
Dias Boba is a local business in Los Angeles that delivers bubble milk tea to doors. The goal of this project is to design an e-commerce mobile app and a website for customers to browse and purchase bubble milk tea for home delivery.

Project Description

Role

UI Designer

Industry

Food & Drink / Shopping

Tools

Adobe XD

Duration

Nov 2022
VIEW HI-FI PROTOTYPES

Logo Design

Marketing Website

web admin

Desktop App Ideation

Marketing Website Ideation

Graphic Design

My Role

I have been the one and only designer for ShopSwap since the project's inception in February 2023. As of September 2023, the platform has grown from a blank slate to over 200 brands on board.

// Ideation & Coordination

I work with the leadership team to discover insights and turn concepts into features. I collaborate with the dev and marketing team to align user goals with business goals.

// Design Execution & Validation

I create sitemaps, user flows, wireframes, prototypes, and conduct usability studies and user testings to validate the designs. I lead efforts to evolve the service and address user pain points related to the brand partnership experience.

// Build & Maintain Design System

I create a modular design system based on reusable components and their states. I maintain the design system regularly, ensuring that UI patterns stay up to date, embrace evolving design and development best practices, and continue to address user needs.

// Art Direction & Brand Design

I design brand guidelines, transactional emails, social posts, and specific campaign related image and video content.

context

Back when our founders ran a digital marketing agency, rising CPM costs were becoming a huge problem for many of the clients. Since then, the problem has only worsened following iOS 14.5 and Google’s phasing out of 3rd party cookies.

At the same time, the number of DTC brands has skyrocketed since the pandemic. This has pushed up demand for digital ad space and has driven up advertising costs significantly. With the fact that marketing budgets are shorten due to the extra expense brands are forced to incur post-covid, such as logistics.

Therefore, DTC brands are left reevaluating their acquisition strategy, with 40% reduction in paid media spend, and 77% of brands saying that partnerships are a high priority.

Problem

The problem is, one campaign usually costs over $10k and takes months to coordinate. From finding the right brands, reaching out to them, to ultimately building the campaign is an incredibly time-consuming process. It is a huge challenge for brands with limited staff, working tirelessly to fulfill orders.

All of these pressures have made digital ads untenable for all but the largest brands. This has led to a belief that the time is ripe to build a scalable technology solution which allows brands to easily implement collaborative advertising campaigns.

Goal

Our goal is to empower brands to cooperate with each other instead of compete. We want to make the process of setting up brand partnerships with value-aligned brands as simple and time-efficient as possible.

In the early stage, our product aims to help brands achieve 3 key goals:

  • Connect with other brands who are actively seeking co-marketing partnerships
  • Offer brand partners the tools they need to effectively plan and execute partnership campaigns
  • Give brands the ability to evaluate the success of campaigns, see what works, and what doesn’t.

Our vision is to build a DTC SaaS toolkit leveraging the data from our current product to build new solutions for the alternative advertising space.

design

// Partner Matching

One of our product's key features is to allow brands with matching target audiences and core values to connect in seconds. The "Partner Matching" page was initially designed with a regular card layout, but it was later iterated into a Netflix-style design that displays brand details when hovered over. This makes the partner matching experience more visual and straightforward, while also reducing the unnecessary repetition of UI elements, such as the "Request Partnership" buttons.

// Campaign Management

The "Campaign Manager" provides partnering brands with simple workflows to easily coordinate multiple campaign types, such as discount sharing campaigns, giveaway campaigns, and gift-with-purchase campaigns. With the "Campaign Manager", brands can plan and launch their campaigns without ever leaving the platform. As of October 2023, there have been four major iterations of the “Campaign Manager”, each with upgraded features and improved user experience.

// Success Measurement

Analytics allows brands to monitor important performance metrics after launching campaigns, such as impressions, opens, clicks, and return on investment.

VALIDATION & REFINEMENT

User testing is a critical part of the design process. By interviewing users and observing their behavior, we identify problems with the designs and iterate them continuously for better user experience. Here are a few examples showing the process of validation and refinement of the designs.

#1 Onboarding

// Problem

To provide a better matching experience for brands, we created ShopSwap Scores, which use our algorithm to show how well a brand matches yours. However, the algorithm requires brands to provide detailed information during the onboarding process, which can lengthen the onboarding time and discourage users from completing it.

// Solution

  • Visualized the onboarding questions by using Tinder-style cards.
  • Added skipping options and encouraged users to go back and finish the onboarding questions after signing up by blurring the ShopSwap Scores.

#2 Pre-Generated Task List

// Problem

There are common campaign types that are often used by brands during partnerships, such as discount sharing, giveaways, and gift with purchase. We want to reduce the workload for brands as much as possible by providing pre-generated tasks after they have chosen a campaign type. However, we discovered that the complexity and specificity of each task varies. For example, some tasks require brands to collaborate, while others can be completed by each brand individually. Some tasks can be completed by one brand but require the other brand's confirmation.

// Solution

  • Defined each task as one of four types: individual, collaborative, hybrid, or general. Designed progress pills with tooltips to educate users how different task types work and what they should do next.
  • After user testing, we discovered that one brand typically does most of the work, while the other brand mostly just confirms. So we also removed the restrictions and dependencies between tasks so that brands don't have to wait for the other brand to take action before moving on to the next task. This improves the continuity of the workflow and reduces the bounce rate.

#3 Landing Page Customization Tool

// Problem

Discount code or gift redemption landing pages are essential in co-marketing campaigns, but they can be a hassle for brands to create. They need to build a working website that collects emails and displays discount codes. Additionally, brands have the need to customize the landing pages to match their branding.

// Solution

  • Created a landing page landing page builder that is easy to use and customizable.
  • Designed pre-made templates that brands can customize to match their branding.

#4 Campaign Broadcasts

// Problem

We discovered that some brands prefer to publicly announce a partnership opportunity with a specific campaign type and topic in mind, rather than cold-calling other brands without knowing if they are interested in collaborating.

// Solution

  • Designed and developed a new "Campaign Broadcasts" feature that allows brands to post and apply for partnership opportunities, as well as manage their broadcasts and applications.

Impact

As of August 2023, we have 200+ brands onboarded with 50+ partnerships launched.

Reflection

// Adapt to the company's work pace

In reality, the design process can be very different from what we have learned in classes or books. Different companies can have different paces and expectations for their employees. When I first joined ShopSwap, I was not completely accustomed to the startup work pace. I would spend a lot of time researching, ideating, and sketching before delivering my designs, trying to find the best practices and solutions to solve user pain points. However, after communicating with the leadership team, I began to adjust my pace. I omitted steps in the design process that seemed less necessary in the startup environment. For example, I sometimes skip wireframing and go straight to prototyping if most of the components are reusable in the design system. This is because the priority in the early stages is to make the product work as quickly as possible. As the saying goes, "we can always iterate later."

// Always consider responsive design and different states of design components

It is important to think about the compatibility of the design with different screen sizes before you find out that some elements will be difficult to present on smaller screens and have to modify the design. It is also important to include different states, such as empty states or error states of components. For example, most of the time the first state that users see is the empty state, so it is crucial to think about how to make the empty states attractive and effective as well.

// Direct communication with the dev team

When I first joined ShopSwap, I would have my designs reviewed by the leadership team, get their feedback, and then iterate on the designs. However, as designs and modifications became more complex, it became difficult to keep track of the design process. So I started to manage the entire design process myself and hand off the designs directly to developers after they had been reviewed by the leadership team. This not only lessened the workload for the leadership team, but also made it easier to sync up design details, discuss potential technical constraints, and modify the design more efficiently.

// Avoid showing redundant or default elements, making the design as clean and effective as possible

I would like to express my gratitude to our CTO Amit, who always encourages me to think outside the box. For example, hovering effects can be used not only for educational information, but also for repetitive elements such as the "request partnership" buttons on the partner matching page; progress pills can be hidden when they are in the default to-do state; and a separate card can be popped up to draw the attention of users when there is a necessary action.

discovery

The questions "what should I eat today?" and "which restaurant should we go to?" are common dilemmas that people face every day. Eating is a basic human need, but there is currently no easy way to get quick inspiration for food and restaurants. When you scroll through short videos on social media platforms like TikTok, Instagram Reels, and YouTube Shorts, have you ever seen a video of someone eating at a restaurant and thought, "that looks so good! I want to eat that tonight!" or "I want to go to that restaurant too!" People love taking photos and short videos of their food before they eat. While some of this content is shared on Instagram stories, the rest is going to stay in the phone forever. Foodies is a platform where people can share food content (especially photos and videos), get inspiration for food, and browse and book restaurants that are linked to the content.

USER RESEARCH

Intrigued by the need for a more visual solution for food and restaurant exploration, I began exploring these questions through competitive analysis and user interviews. Was this really a problem worth solving?

// User Interview

I created an Interview Protocol to ask potential users. The high-level goals of this study are:

  • To understand user needs when it comes to exploring food and restaurants.
  • To identify desirable and lacking features of current competitors.
  • To determine if users need a more visualized application to explore food and restaurants. If so, how do they currently address this need? What tools and processes do they use?

// Affinity Diagramming

According to the interviews with four potential users, I created four sets of sticky notes, and organized them into groups based on their relationships.

// Results

As a result, I came up with the functional requirements for Foodies:

  • Photo Stream
  • Video Stream
  • Reviews & Ratings
  • Location & Distance
  • Restaurant info (address, phone number, menu, and opening hours)
  • Sort & Filter
  • Save & Collect
  • Reserve
  • Maps

COMPETITIVE ANALYSIS

I divided the competitive landscape into four types of apps:

  • Maps
  • Business Review
  • Restaurant Booking
  • Social Media

ideation

Combining with interview responses, I drafted three system goals that address three pain points in the current competitive landscape:

#1 Explore restaurants by browsing through the photo stream

// Pain Point

Restaurants are typically displayed as either a text list or a list with small thumbnail images.

// Solution

Provide a more visual and image-rich restaurant feed.

#2 Explore food by swiping through the full screen video stream

// Pain Point

Most restaurant reviews are text-only, but people today prefer to consume information in other formats, such as video or audio, especially when making quick decisions.

// Solution

Provide a short-form video stream to effectively inspire people of food / restaurant choices.

#3 View restaurant details and reserve a time

// Pain Point

Traditional restaurant apps make it time-consuming for users to choose a restaurant and explore its food. Users have to browse through a list of restaurants, choose one, view its food, go back to the list, and explore the next ones.

// Solution

Reverse the traditional path and provide a streamlined experience from exploration to decision. Allow users to explore various kinds of food through a quick and visualized method, and then directly go to the restaurant page to see detailed information or reserve a time.

PERSONAS & SCENARIOS

// Scenario 1

Tom, a 28-year-old marketing manager, lives in a rental apartment in downtown Los Angeles with his girlfriend. He recently discovered an app called Foodies, which he finds helpful in finding new restaurants. After a long day at work, he and his girlfriend were starving and decided to go out to eat. They were looking for a place that was close by and open late, so they opened the Foodies app to find a recommendation. Tom found a video of someone eating Japanese BBQ and decided to check out the restaurant. The restaurant was only three blocks away, so they headed over. While they were eating, Tom took a short video of the food and the restaurant environment and posted it to Foodies. The next day, he found that he had received 60 likes and collects on his post. He enjoyed using Foodies and decided to continue using it to find new restaurants.

// Scenario 2

Jenny is a 26-year-old art director who recently graduated from New York University. She loves to dance and has three best friends who she met in dance class two years ago. They usually go out to eat together after class. Two days ago, one of her friends recommended the Foodies app to her. After work today, Jenny decided to give it a try. She found the photos on the photo stream page to be very appealing. She wanted to eat a healthy salad for dinner, so she selected the "Vegan" filter. There were so many choices, so she searched for "salad" and found a restaurant that served her favorite salmon salad. She booked a table for 6:30 and called her friend Lisa to join her for dinner.

INFORMATION ARCHITECTURE

After completing my initial UX research and planning, I created a site map to outline the key pages for the MVP. The main pages are:

// Home page (photo stream)

Users can explore restaurants by browsing user-generated content, searching, viewing restaurants on maps, and applying sorting and filtering options.

// Explore page (video stream)

Users can quickly get food/restaurant ideas by swiping through the full-screen video stream, interacting with video content, and going to the restaurant detail page.

// Saved page

Users can view the posts or restaurants they saved.

// Explore page/video stream

Users can view their posts and liked posts, and edit profile information.

task flow

Then I outlined three task flows for the phase 1 of design to inform the wireframing and prototyping processes and identify gaps in the user experience.

design

I worked on the wireframes with the task flows to create wireflows and a testable prototype of the MVP.

USABILITY TESTING

With the prototypes, I conducted usability test to understand:

  • Can the app meets the needs of its target users when it comes to finding food and restaurants?
  • What problems do users encounter when trying to use the app to explore food and restaurants?

Each test was divided into four parts:

  • Pre-Test Interview
  • Tasks
  • Post-Questionnaires
  • Debriefing

Key Findings:

  • Participants found the app to be visually appealing and easy to navigate.
  • Participants appreciated the restaurant recommendations and suggested adding more filtering options for searching.
  • Participants found the current sharing task flow is not efficient enough and suggested adding in-app sharing and messaging features.

Recommendations:

  • Improve the sort and filter options to enhance the user experience.
  • Consider adding more options for sharing, such as adding a messaging feature that allows users to share posts and chat directly on the app.

NEXT STEPS

  • Implement the changes and improvements recommended in the usability testing.
  • After making changes, conduct another round of usability testing to validate the effectiveness of the improvements and ensure that the product is meeting the needs of users.
  • Communicate the results of the usability testing and any improvements made to the relevant stakeholders. For this specific project, I will communicate with the other co-founder and a developer, ensuring that the product is technically and financially viable and sustainable.
  • Continuously improve the product based on user feedback and ongoing user testing.
  • Current designs are focusing on content consumption task flows, such as browsing and interacting with content and viewing restaurant details. The design goal for the next stage will be focusing on content generation task flows, such as taking and uploading photos and videos, writing reviews, managing personal profile, etc.

UNDERSTANDING THE PROBLEM

Before Nexev hired a product design team, a beta release of the new platform had been implemented based on the blue-sky concept executed by the founder and developer. These mockups were created without any usability testing and had little consideration for the technical and product limitations on the scope of work.

I conducted research interviews with the primary users to uncover any pain points that they were experiencing with the beta release. My research encompassed:

  • Understanding the user goals and needs
  • Uncovering pain points with the existing user journey
  • Determining the success of the task measured

GATHERING INSIGHTS

I developed a list of usability issues in order of severity, based on Jakob Nielsen’s five-point rating scale:

  • 0 – Not a usability problem at all
  • 1 – Cosmetic problem only (need not be fixed unless extra time is available on project)
  • 2 – Minor usability problem (fixing this should be given low priority)
  • 3 – Major usability problem (important to fix, so should be given high priority)
  • 4 – Usability catastrophe (imperative to fix this before product can be released)

PROTOTYPING THE SOLUTION

Based on the above problems identified, I worked towards addressing these pain points by coming up with potential solutions:

  • Redesign the nav bar, concentrating main features into five tabs.
  • Redesign buttons and icons, enlarging the touch area to improve tapping accuracy.
  • Redesign the event creating page, reducing the number of steps to minimize time to completion.
  • Redesign sort and filter features to a clearer and more intuitive way.
  • Redesign event cards on the homepage, showing only the necessary information to make it clearer and cleaner.
  • Add details in the event category by referring to current event planning competitors.
  • Rephrase description texts in the create event form to a simpler and more understandable way.
  • Enlarge the type areas.
  • Change the current font to a more common font used on mobile apps - Open Sans - to increase the app’s credibility.

I brought my ideas to the founder and developer. After a discussion with feedback, we chose to skip the wireframing process according to the business needs.

FINAL PROTOTYPES

I proposed changing the brand colors to a more youthful and energetic style. After several iterations and continuous discussions with stakeholders, I opted for a bolder design direction and created the final prototypes for the event creation task flow.

WHAT’S NEXT

I will work very closely with the front end team to spec out any missing interactions that were not covered in the mockups during the whole development process.

After the implementation of the new rebuild of Nexev, I will validate all the design decisions based on the performance of key metrics such as active users, number of events, and retention rates.

Some key takeaways from this project are:

// Involve engineering upfront

This helps to reduce any rework later on as an understanding of the technical limitations upfront will help to inform your design strategy.

// Trade off between quality and cost

Sometimes real UX projects won’t be as complete and comprehensive as ideally planned, considering the business needs, time and cost. As a designer, it is critical to balance between cost and quality constraints, and adjust the design process accordingly.