Project Overview
Café Blanc is a Korean inspired coffee shop located in Coquitlam, British Columbia. It has a charming ambience, a majority student customer base, and once upon a time— a horrible website. During my time as a design lead, I worked on branding, social media, while creating a website. The one I inherited struggled to make digital conversions without relevant copy, clear branding, and order links. My goal was to redesign it in a team to increase revenue and help the café thrive in a menacing pandemic.
Discovering the Problem
Almost all of the café’s sales were made from walk-in customers, who could no longer be relied on as the sole source of revenue. In addition to failing to maintain sales, a lack of traffic meant losing ground to similar stores with stronger web presences. Of the visitors, few were ordering— dismal statistics cited an annual bounce rate over 83%. The cafe's sole strong funnel of visitors were Instagram campaigns which accounted for 58% of visitors.

A delightful fix of caffeine

Role: Lead designer, team of 4
+ Initiated a brand refresh & created an identity system
+ Conducted user research and applied findings to create a website using double diamond methodology
+ Designed and shipped the website on Figma and Zeplin
+ Outcome: Currently at developer handoff stage
September 2021 - Present
A brief look at the final solutions

Defining the Scope

To see how Café Blanc could differentiate themselves, I evaluated the existing websites of several local competitors and identified it focused on specialty drinks and mini-cakes. Café Blanc's menu is mid/high-priced for the area while uniquely consisting of comforting fare made with quality ingredients. I recognized that the brand identity should thus reflect this with a design centered around quality, simplicity, and refinement.
To better understand how I could drive sales by using the website to acquire new customers, I organized patterns found in user feedback and my heuristic evaluation into 3 main categories:
The pre-redesign website: Unpersonalized and cold

The owners asked to keep the original icon from the logo (the coffee bean), but otherwise gave me creative control. My user research cemented the importance of a brand persona and a UVP, so I advocated for a brand refresh and came to two questions relevant towards business metrics and the user experience:

How might we rebrand the café to better align with their products?
How might we capture potential customers and increase orders?

Defining the Solution

Due to its proximity to Simon Fraser University and the high percentage of Instagram referrals, the target demographic of the café is university students. After interviewing a mix of customers and non-customers from the ages of 18-55, I found several commonalities in website content of interest. When asked “In what scenario” and “How often” they visited a café website, most responded they would do so to look for a menu. Seeing that accessible links were a priority, I created a sitemap to make sure sufficient CTAs were present. I then made a user journey map and created a set of design and brand principles to unify the design system with emotional guidelines, intending for all components to reflect “friendliness,” “purposefulness,” and “comfort.” 

Developing the solution

In the process of planning, I discovered the voice of the café— the comforting tone of copy I wanted to write, and the artistic but approachable designs I wanted to make. Higher prices made it necessary to emphasize quality, while playing to the cafe’s unique value proposition as a comfortable place to study. I tried to convey the calm ambience customers described through hand-drawn illustrations and slightly rounded corners. A friendly color palette of warm browns and beige was incorporated into the art and photography, while the copy was written in a similarly approachable tone. The designs themselves used DM Sans and Roxborough CF, typefaces that appear approachable because of their double story letters and exaggerated circular counters. 

The landing page is the first thing visitors see; playing an important role in establishing a brand’s context, aesthetics, and tone. I tested my first prototypes by asking users to rank taglines and complete timed tasks like ordering from UberEats. Finally, I would follow up with an Quick-Exposure Memory test, asking them what they remembered from the landing page, and if they could choose 3 adjectives that best described the visual and emotional impression they got:

Both qualitative and quantitative research contributed to my final design; the images from design 2 were organized in a more interesting way, while the text was enclosed in a container so it didn't look decorative. The photography and design system highlight brand keywords and specialties— reflecting the wholesome and local nature of the café with a human touch. 


Making changes based on insights

Some of the older interviewees found design 2 and 3 slightly distracting, and preferred a minimal design with one picture, tagline, and order buttons. However, I chose not to include CTAs on the hero image because they are already accessible on the Google page and Instagram; meaning that users chose to click in and ignore them to seek more information on the business. Instead, I had a fixed navigation bar with the order links, and provided them farther in where more trust is established. Every other page integrates CTAs in an appealing way rather than forcing them on the user early and leaving a bad impression.

design goal : aesthetically pleasing but not extraneous

My early iterations were problematic because of color choice, poor information hierarchy, and undifferentiated content. Later, I improved on them with visual hints like borders and order icons that suggested the polaroids afforded clicking:


Initial Order Link Design
Final Order Link Design

The new CTA page above better employs hierarchy by prefacing the links with text and changing the visual rhythm to move from left to right. The size of the type goes from the heading to the images, drawing a user’s eye to the clickable visual links. Beyond the initial issues with the polaroids, users found them a fun and skeuomorphic way to represent ordering: adding delight and personalization to the experience.


The design rationale behind this was Fitt's Law; which states that the time it takes to reach a target area is a function of distance and size to the location. I employed this principle by conveniently placing order links at the bottom of the screen, closest to the user’s thumbs. 

Putting it all together: the final desktop design

Old design on left, new design on right

Edge case: error state

Delivering the solution

Unused hours draft designs that automatically detected the user's current time to display the cafe's status

Once a developer was onboard I had to re-evaluate the designs and we didn't ship most of the animations, contact form, or Instagram integration. As much as the time feature that displayed the cafe’s status was well-received by testers, it would invite issues if it wasn’t updated properly or if there was a bug (wrongly suggesting the café was open or closed). Meetings are extremely important, and earlier communication would’ve saved me a lot of time. I learned that continuous dialogue between designers and developers ensures unity and understanding— so go be nice to your dev and buy them a coffee.
As Google Analytics is connected to the domain, potential metrics include bounce rate, average session duration, and percentage conversions to ordering apps. While I don't presently have any numbers to present, I am immensely happy to have been part of creating a scalable design system with my developer in Zeplin— one that I think will grow far beyond my designs with future employees.

Takeaways

copy IS UNDErRATED
This project helped me realize how much of a difference copy makes, and how challenging it can be to write. Consistency between writing and visuals establishes the “voice” of the brand, making both crucial to a digital experience. Good copy can elevate an experience from good to great, and tone should be decided early in the design process (this is how I have been justifying the hours spent on A/B testing). Long story short— hire good copywriters, test your copy, and always make multiple versions of designs and text.
Test often- Fail fast to learn faster 
By running usability tests early, I was able to pivot quickly and scrap what wasn't working. Within this, users also had conflicting opinions. While some would hate a design, others would love it. I tested a range of people while keeping the targeted demographic in mind, allowing me to directly apply feedback and build on top of failed iterations. The design process is not linear, so always prototype, test, and try new things.
Other case studies
SPC REDESIGN (ux & product)Bored museum (branding)