honestbee: A Usability Case Study

Update as of September 2017:

At time of publication, I was not affiliated with honestbee.

While on-demand delivery has been around for quite some time in Singapore, it was usually constrained as templated forms hacked together on the websites of restaurants (if they even had a website). With the rise of concierge services handled through mobile apps, it is inevitable that new domains and verticals will be added.

One such vertical has been on-demand grocery delivery service, a familiar concept in the states (think AmazonFresh) but not so in Singapore. Currently, there are two main competitors: RedMart and honestbee.

honestbee

honestbee holds itself to be Asia’s leading online lifestyle concierge and delivery service. They currently operate in Singapore, Taipei, Hong Kong, Jakarta, Kuala Lumpur, Bangkok, Niseko and Tokyo. It has also ventured to other services besides groceries — laundry, health, and home & auto. For this case study, we will look into the user flow of just online groceries to keep other variables constant.

Disclaimer: I am in no way affiliated to honestbee. This case study is a way to get my thinking hat on and attempt to come up with solutions.

Design Process

I will be using IDEO’s human-centered design thinking process for this study.

IDEO Design process

Empathize

User Personas

Honestbee being a marketplace has two sides of users: the buyer and the concierge, otherwise known as a “bee”. In this case I have decided to focus on just the buyer. After looking through honestbee’s range of grocery store offering and its items, I came up with the below user personas.

Jobs to be done

Keeping in mind these user personas, here is the primary motivation of a prospective user.

When I want to buy an item, I want to be able to find it easily so that I can purchase the item and have it delivered to me.

Guerrilla Usability Testing

I had 5 participants complete the following task:

I would like to buy a laundry detergent. How do I proceed?

Here is the set of micro-tasks associated with this task:

• Search for the laundry detergent
• Add the laundry detergent to cart
• Choose a replacement for the laundry detergent
• Verify shipping address

All 5 users tested were from the age range of 20–34, are tech-savvy, and have never used the honestbee app or other grocery delivery app before.

Define

After getting feedback from the users, I proceeded to categorize them, and putting similar issues together under themes to figure out prevalent pain points.

Affinity Mapping

In order to verify which pain points to tackle, I used a 2x2 matrix with the x-axis charting business importance and the y-axis charting importance to the user.

Affinity Mapping using themes

This matrix allowed me to narrow down to the top 4 issues that matter most to both the business and to the user.

1. Search and Filtering Results

Store screen — Users were tasked with buying a laundry detergent

2. Adding To Cart

Adding an Item

3. Finding a Substitute

However, this process needed so many taps just to finalize the substitute. For example, 4 out of 5 users kept returning to the previous original item to check on its description before deciding on the final substitute.

4. Check Shipping Address

Ideate

With all these input in mind, it’s time to start sketching out solutions. First, I did several lo-fi UI sketches:

I should continue improving my sketches :p

before polishing them. Below are the Hi-Fi mockups done in Sketch.

Search and Filtering Results

Objective: This will help to improve discoverability of Search and help users find what they want quickly and easily.

Adding to Cart

Objective: Increases buyer confidence that item has been successfully added to cart.

Choosing a Substitute

Users previously had to toggle between the screens showing the original item and the substitute item.

In the suggested version, a panel of suggested substitutions will be revealed when user taps on “Choose a substitute”. User can also tap on “See more” to show a fuller list of suggested items and carry out a search if needed.

Check Shipping Address

How many screens does it take to change a shipping address?

In the above flow, the user is required to have known changing the address can only be done on the home screen (as shown in the second screen above). This was not the case, however, as indicated by the user testing feedback where only 1 user knew how to change the address successfully. Hence, I have decided to add an edit button to the address on the checkout screen itself. Tapping the edit button will enable the user to jump straight to the map to locate the desired address.

Change the address on the checkout process

Below is the final flow for changing the shipping address.

Less screens now!

I have also included another entry point for the user to add an address under You > Settings.

Add an address under You tab

Objective: To reduce friction in changing address and speed up checkout process.

Prototype & Test

I handed my prototype to several more users (Would absolutely love for you to try the prototype here!) and here are my results:

Additional musings:

• It is interesting to note that changing the shipping address would actually clear out the entire cart, causing the buyer to have to re-add items to the cart. I assume it is due to concierge efficiency and catalogue accuracy. Concierge efficiency because the “bee” or the concierge will simply go to the store nearest to the user to pick up items. Catalogue accuracy because stores in other locations might stock different items in various quantities. While this reasoning is valid, it does not agree to the mental model the buyer has. As a buyer, I want to be able to get the items I want regardless of where I get it from. I should not have to re-add my items when I change my address.

Search results can only be filtered by Department

• Search results could be better refined and given more filters such as brand, popularity and price. Users were able to locate their preferred item but they spent most of the time scrolling down
• More product details would be good for easier decision-making such as nutritional information and ingredient list which are missing currently

Final Thoughts

It would have been great to get real data on how customers are actually using the product since the above suggestions were made hugely based on my assumptions and the limited feedback obtained from user testing.

Nonetheless, this was my first attempt at doing a usability case study and I hope to cover more startups based in Asia through either onboarding teardowns or more case studies. Would greatly appreciate any feedback 💖

Feel free to reach out to me at stph.sutanto[at]gmail.com or at LinkedIn — I’m always down for a conversation!