KPCB Instacart Redesign
DESIGN / PRODUCT / UX / PSYCHOLOGY / PHOTOGRAPHY
20 min read
Disclaimer: I am neither a designer for Instacart nor affiliated with the company. This project is part of my redesign challenge for the KPCB Design Fellows Application.
In the grocery delivery space, there are numerous services available including Amazon Fresh, Fresh Direct, Jet, and many more. However, Instacart stands apart from the other services because of their mobile and web browsing and shopping experience.
Amazon used to have a standalone app, but it was discontinued, leaving Amazon Fresh to be accessed only through the original Amazon marketplace app and website. Jet does not have a standalone grocery app either. This means that Amazon and Jet customers can only order through their respective marketplace app or website--which includes searching through the hundreds and thousands of other irrelevant products offered. Consequently, the ordering experience becomes overwhelming. Finally, Fresh Direct's current shopping experience on the iOS app presents users with a linear, drawn-out, and cascaded decision-making architecture which fails to cater to the needs of their users. First, the user interface does not feel like an iOS app. Second, the user experience feels restrictive as there is only one option for accessing shop items and one option for browsing through the extensive list of categories and sub-categories. Therefore, this presents an innovation opportunity for Instacart:
In the following redesign challenge, I primarily focus on the user experience design of the iOS app. My redesign does not consist of creating any new standalone features since I do not have access to any user research or other quantitive data from Instacart. Instead, my design decisions strive to refine the current user experience by making it more intuitive, natural and seamless. By deciding to refine the user experience with the limited information I have, I still have the opportunity to quickly prototype my design implementations with walkthroughs with friends and family members who are interested in purchasing groceries the non-conventional way.
With my iOS development background, I am able to draw from my strengths to understand and navigate through any engineering constraints that my design decisions may pose. However, I decided to challenge myself by critically and creatively thinking about how concepts from my Psychology education (e.g., my Thinking and Decision Making course that I am currently enrolled in) can be applied to the tech space. Effectively, to reinforce and enhance each design decision with a human-centered design approach. This way, when I am walking through my redesigned user flows with people, I can better realize, understand, and address their needs to improve the user experience further.
Examining the opportunities for improvement and refinement
When approaching any redesign, it is essential to identify any opportunities where human-facing problems need to be improved and refined. Next, those opportunities should be challenged with interpretive questions in the form of hypotheses.
- What changes would make the app valuable to more people, and more valuable to the people who are already using it?
- Are there any tensions or inconsistencies with the current interface that affects the user experience?
- Is the interface inhibiting users from having the freedom to move throughout the digital categories (a.k.a. aisles) with ease? Does the app allow users to accelerate the process of conventionally shopping for groceries?
- Does the app reduce the headache of the crowded and busy grocery store experience, or does it create a different type of negative stress?
Instacart has both a customer app for people to buy and a shopper app for their delivery employees. Effectively, there is an interaction between these two groups with Instacart at the center. What’s more, Instacart features products from many grocery store companies; therefore creating interdependent relationships with all other players within this system. As someone looking to redesign the user experience for customers, I needed to be mindful of how my changes would affect the different relationships in the system--while being aware of who I am designing for.
Identifying Who I am Solving For
When I first approached this redesign for the user experience, I wondered: if I do not have access to Instacart's user research or quantitive data, then how will I know whether or not my improvements on the user experience will be successful? How do I know that this will actually be something that people find valuable and easy-to-use?
Customer Base and Use Cases
Luckily, I found two people who use the Instacart app: my aunt, Fatima, and my friend Christopher. I also included a third participant in my user research--my mother who does not use the Instacart app for a number of reasons.
Christopher is a vegan and is new to Manhattan. He would prefer if there were only one store to suit his needs; however, that is not the case at least on the Upper West Side. Chris lives in Midtown, and his favorite store is downtown in the West Village--opposite to his commute home. Although he enjoys using Instacart, he complains that finding vegan options takes way to long.
Before Fatima started using Instacart, she shopped at Giant for meats and Safeway for produce because they were the cheapest at those stores. She found out about the sales from in-mail coupons and befriending workers at both stores. However, she complained that shopping this way was too time-consuming, considering that she is a single mother working full-time and has two kids in middle school. Fatima loves using Instacart because it saves her time (by not adding 45 minutes to her commute) since she can now cross shop and order from multiple stores. However, Fatima would like an easier way to find what is on sale.
Finally, my mother dislikes using digital products to shop because she believes that shopping in person is better--especially since she has the time for it. What's more, she finds the Instacart app challenging to use, mentioning that it does not mirror how people typically shop in grocery stores. Even though she has owned multiple iPhones over the last ten years, she feels overwhelmed with the app's user experience.
Understanding and Solving People's Problems
During our meetings, I carefully watched how they used the app from navigating through it, adding and deleting items from the cart, searching for items, and making purchases. Below are the following people problems that I learned:
- Christopher complains that finding vegan options takes way to long.
- Fatima would like an easier way to find what is on sale.
- Mom would like an experience that mirrors how people typically shop in grocery stores.
From my own observation, I saw discovered other pain points. People can easily accidentally delete an item from their cart by quickly scrolling, tapping to keep the device on, or trying to change the quantity. If someone accidentally deletes something, then it will be hard to remember what was removed because the animation happens too quick and there is no location for deleted items. Therefore, people like Christopher will have to repeat that strenuous and drawn-out way to search for his vegan items.
How do I know that these are real problems?
Interviewing Christopher, Fatima, and my mother and researching on my own regarding how the app interface works, I found that these were problems worth solving. Even though I do not have access to any Instacart research, I can imagine that some users could, in fact, identify with Christophers, Fatima, and my mother’s pain points.
Working towards solving other people’s problems gave me clarity about questions I initially wondered. Learning from Julie Zhuo's talk about How a Facebook Designer Thinks and A Community Built for You, I used questions she posed during the product development cycle while creating my own interpretive questions to make sure that my redesign decisions were aligning--all close as possible--towards solving the people problems I discovered.
- What’s happening before people get in and start interacting with Instacart?
- What are they feeling and thinking?
- What got them to use the app in the first place?
- What happens afterwards?
- After the interaction with the app, how do we want people to feel?
- What do I want people to have taken away from their Instacart experience, and would they be happy to do it again?
Below, I proceed by explaining my three design opportunities that not only serve as the solutions to my hypotheses, but also carrying along the stories and problems of Christopher, Fatima, and my mother throughout my redesign process.
Design Opportunity 1
Consistency between native and web platforms and design guidelines for their respective platforms
By examining both interfaces, there seem to be strong similarities. This design decision is good because it creates familiarity for an intuitive user experience between both platforms. However, too much consistency between the two platforms could cause design guidelines and functionalities for their respective native platforms to be neglected. For example, with further analysis of the cart interface, I observed the following:
- The web interface works because options (e.g., instructions and delete) are neatly laid out as users make choices with great precision with the use of a mouse. The mouse pointer changes events from the arrow to the finger, while hovering over the "instructions" and "delete" options. The change in event signifies to the user that an action will follow if either of the options are chosen.
- However, on the mobile app, this design does not work because of the touch interface. There is no event to signify to the user before an action takes place. The mistake of removing an item (whether accidental or not) is very probable while scrolling, changing quantity, tapping on the item for more information, tapping on instructions, or simply tapping to keep the device on. Therefore, users will not remember what item they accidentally deleted as there is no error prevention.
Since, no undo option is supported on either platform and no warning that an item has been removed as the animation happens too fast, these possible error cases must be addressed as recovering from these mistakes are irreversible on either platform. To fix this issue on the mobile app, users have to leave the cart and re-search the item again.
Deleting an item from the cart, then tapping on the marketplace's cell and hoping to search for your item again.
What’s more, the mistake of removing an item from the cart (or shared cart) becomes a more significant problem when users are ordering groceries from different marketplaces. Now users have to fulfill an additional step of changing marketplaces before searching for an item that they probably do not remember deleteing. Naturally, users would think that tapping on the marketplace table view cell will segue them to the respective marketplace. However, that is not the case as they will be presented with options for the next delivery times.
Design Opportunity 2
Improve cart access
There are only two ways to access the cart: tapping the cart icon in the upper-right corner (which is difficult to reach on phones with larger screens) or using a non-intuitive gesture (which is not always recognizable) that requires users to swipe-from-the-right in any part of the app. However, the latter option does not always work because most of the design decisions in the other app screens implement a table view interface with horizontally scrollable rows.
For example, when at the “Home” screen, the gesture is only recognized when a row contains a coupon or promotion. What’s more, coupons and promotions are only presented three times: before the “Your Items,” “Trending Near You,” and “Popular in [Insert your city]” rows. Unfortunately, some of these sections are also horizontally scrollable which means that the gesture will also not be recognized. Moreover, if users attempt to swipe from the right to access their cart in any of these horizontally scrollable tableview rows, they will either scroll through the presented items or be segued over to a new screen that presents them with more items.
To add, the "Coupons" and "Your Items" tab-bar screens also implement a horizontally scrollable table view interface. Meaning, the gesture will also not be recognized.
The technical reason that the gesture is unrecognized with horizontally scrollable table views is because the gesture cannot override the horizontal scrolling of the table view. However, most users do now have a background in iOS development, so how would they know. Therefore, the current design implementation does not work.
Design Opportunity 3
Searching and shopping for items is a lengthy process
If a user accidentally deletes an item that they don’t need, then retrieving that item becomes difficult because there is no location for deleted items. Therefore, the search process of looking for that item will have to be repeated.
Searching by selecting the "browse" option from the tab-bar seems like the most natural choice for users, as it resembles shopping through aisles. In the app, aisles are broken down into categories. However, shopping through these categories requires users to filter through each sub-category one-by-one to find exactly what they want. Unfortunately, the filtering moves users through a drawn-out sequence of view controllers. The problem with this approach is that when users decide to go back after adding an item to the cart (or not finding an item), they are brought back to the category (or aisle) they had initially selected—not the sub-category. The flaw with this design implementation is that it is counter-intuitive against how people typically shop in aisles.
Tapping back after filtering brings you back to the main category not sub-category
Imagine for a second that you walk down an aisle and find what you want (or don’t find anything). Then if you want to look for something else, you would have to go back to the beginning of the aisle to start this process again. Effectively, the app does not follow the normal behavior of how people typically shop in a grocery store aisle.
Based off of my analysis and assumptions, below are the following goals I am striving for with my redesign:
- improved user experience with mobile app that matches the user experience with other mobile apps that they have grown accustomed to.
- ease and flexibility of quickly searching through an entire store based on user's needs without feeling overwhelmed.
- non-destructive shopping experience that allows users to easily retrieve recently deleted items.
- a digital shopping experience that mirrors the physical shopping experience and reduces real-life shopping time in the process.
Design Process (Wire-framing)
To start the design process, I created three wireframes that solves the problems for my three design opportunities: improved removal of items from the cart, accessing the cart, and shopping for items.
From left-to-right: Wireframes for Design Opportunity 1, 2, and 3 designed in Sketch
The three wireframes above are each very similar to their respective original designs, their difference being primarily in their functionality.
The first wireframe incorporates the slide-to-remove gesture that users have become accustomed to in iOS table views (e.g., Apple Mail app) because of perceived affordance. This design decision serves two purposes: it eliminates the redundancy of two visible actions (which can serve the same purpose) in the small table view cell and prevents error-prone conditions.
With Instacart's existing design, users can remove items with both the "remove" and "quantity" options. However, both options are rather close to each other, and one is absolute--the "remove" option. Therefore, implementing the slide-to-remove gesture will protect against user mistakes. Furthermore, as of version 6.1.3 of Instacart, the app supports much of the functionality of iOS 11 (e.g., haptic feedback and drag-down to dismiss view controllers presented modally). Therefore users already expect specific and familiar features to be supported.
Rather than distracting users by presenting them with an undo, warning, or confirmation option before they commit to the removal of an item, their items will be safeguarded in another "cart" after the removal process. The second wireframe has an additional table view row option--"Recently removed from cart." This option only consists of items removed during that shopping session. After each new session, this safeguarded cart gets emptied and populated with recently removed items from the latest session. Finally, the safeguarded cart can only be accessed through the table view row--which means that it is hidden away from the user and out of their way until they need to use it, but at the same time, easily accessible through the "Your Items" from the tab-bar.
The second wireframe provides an improved solution for accessing the cart. I removed the "By category" table view row from Instacart's original design. Originally, this row provides access to the cart; however, tapping on the row segues the user to another screen to view the cart. If a user taps on a tab labeled "Your Items," they should not have to be segued again to see their cart items.
My redesign adds a horizontally scrollable table view which allows the user to scroll both left and right to view their cart. The advantage of this design allows users to view their frequently bought items and suggested items in the "Buy it again" category and the "You may like" category, respectively--all while seeing their cart. Moreover, the "Buy it again," "My cart," and "You may like" horizontally scrollable interface will now support 15 items rather than 10. This way users will only be automatically segued when needed. However, there is still the existing option to tap "View more" if necessary.
Finally, the third wireframe refines the searching for items process. Using the "Browse" option from the tab-bar would seem like the most natural option for users, as they do not always know what to search for or they may be interested in discovering new options. However, when users search for items through the browse option, they are led down an extensive decision-making architecture. Below is the current architecture for searching and filtering through spreads in the pantry at the Whole Foods Market.
Instacart's current decision-making architecture
The problem with the current decision architecture is that if users want to search through the sub-categories in the pantry for specific items, (e.g., organic, kosher, or items they may not know the name of), they will have to follow the architecture above. Meaning, users can only search through one sub-category at any given time. Now, when users find their item (or do not find) after filtering, tapping on the back button brings you back to the beginning of the pantry only to search through another sub-category again. What's more, some stores have categories with nearly a dozen or more different sub-categories. Unfortunately, this filtering experience does not translate to how we typically shop for groceries in real life.
Assume you are in an aisle (in our case, a pantry), once you find something you do not go back to the beginning of the aisle. You stay where you are and continue to search.
What if I just want to shop for organic and/or kosher goods throughout the entire pantry? Currently, there is no opportunity to do so. Furthermore, some stores that call themselves organic, do not always carry organic goods, so how do I solve this problem?
By condensing the decision-making architecture and allowing users to search the entire aisle by filtering through all the sub-categories in that aisle--all at once. Doing so, we achieve the following decision architecture:
Revised decision-making architecture
This implementation not only reduces the number of steps for users to search, but it also prevents them from getting lost in a series of cascaded view controllers. Now, you are probably wondering well not all filter options apply to the same sub-category. Not to worry, as an engineer, I know that if you search through a list and do not find what you are looking for, you can decide not to do anything. Effectively, if a specific filter does not apply for a given set of sub-categories, then we will only get back items in those sub-categories which do pertain to what was initially searched. This idea also applies to filtering by the "Brands" parameter.
Assume you choose to filter through Spices and Spreads for both "organic" and "kosher." However, there are no "kosher" Spreads. The items you will be presented with will be "organic" Spices and Spreads and "kosher" Spices.
Perfect, you can now shop through all aisles for only organic and/or kosher products!
Why condense the Decision-making Architecture?
By understanding how the Decision-making as a Constrained Optimization Model works. I will explain below.
The objective function is defined as the decisions to be optimized. Decisions can be anything between deciding how to find ingredients for a dish you never cooked before, figuring out what is the best way to know what stores around your area offer only organic, kosher, or vegan options, or discovering the best sales from your local markets.
Constraints are defined as factors which can cause influence. These factors can be cognitive (e.g., attention span, memory ability, etc.) or material (e.g., money, time, preferences, etc.).
Note: Both decisions and constraints can be spontaneous as they are always subject to change depending on what happens in our day-to-day lives.
Using this model allows for brainstorming more accurate scenarios for evaluating not only how people make decisions but also why and when people choose between maximizing decisions and satisficing decisions. These scenarios can then be used to learn use-case scenarios better and analyze results while conducting usability tests and user research. Most importantly, the model provides us with powerful ways to influence the decision-making process to better help and guide others.
Original User Journey: Typically users would have to allocate specific time in the day to use the app.
New User Journey: Now users can easily and quickly use the app without the need o remembering to set time in the day. Instead, the Instacart experience seamlessly integrates into their lifestyle.
We have now encouraged users to not stop by the corner shop coming home from work on the NYC subway because they can now quickly and easily search through store categories and sub-categories on the app. Therefore, allowing them more time to spend with their families.
By condensing the decision-making architecture, my "Browse" redesign provides users with a larger probability of maximizing their decisions while taking into account a wide range of constraints. For example, there is less memory strain in having to remember which sub-categories to filter through. Effectively, this means that the chances of users having to modify their order will be reduced, which in turn will reduce confusion between Instacart's customers and shoppers.
Design Process (Hi-Fidelity Designs)
After that long and thorough explanation, I will now walk you through the flow of my final mockup redesigns.
1. Improved removal of items from the cart
Less mistakes, more time for shopping. But just in case you make one, we got your back.
Cart before deleting. Cart while deleting. Cart with deleted items.
Let's observe the new flow from left-to-right. The first screen shows my shopping cart. The second screen is deleting an item from the shopping cart using the familiar swipe-to-delete gesture. The third screen shows the cart consisting of "Recently removed items." -- which is now accessed through the "Your Items" tab-bar option. The screen also presents items that were recently removed from other stores. Remember that recently removed items only get saved for the current shopping session. Meaning, if a user has made a new order and has not added and deleted anything during their new shopping session, the safeguard cart will be empty. Another feature to note is that recently removed items are classified depending on the store. Therefore, if a user is cross-shopping between two stores, they will not have to go back and change the store to re-add that grass-fed steak from Whole Foods.
Note that there is no "Remove" option for items in the shopping cart's table view cell of the first wireframe. This is to guard against user errors such as accidentally tapping the option while scrolling, changing quantity, tapping on the item for more information, tapping on instructions, tapping to keep the device on, or changing quantity. As a result, the interface is cleaner and not as redundant.
2. Accessing the cart
Cross-shop in real time.
In the above redesign, users now get an at-a-glance view of their cart and also have the option to tap on the "Your Items" tab to see their recently removed items from the cart--which only contains items from their current shopping session.
By examining the mockups from left-to-right, users can also view their recently bought items and suggested items. Most importantly, their shopping cart is in between both categories for two reasons. First, the cart access is now closer to their fingers--regardless whether the user is left- or right-handed. Second, the horizontally scrollable interface allows for cross-shopping between both categories--all while viewing what is in their cart. Therefore reducing the cognitive load on the users of having to remember what is actually in their cart. Finally, my redesign enables users to now have the option to access their cart from the table view of the "Your Items" tab-bar. This way, users will not have to re-orient the device to access the cart by tapping the icon in the upper-right corner; however, the option is still there.
From left-to-right: Instacart's "Your Items" screen and my redesign for the new interface (prototyped in Principle).
3. Shopping for items
Not just for the power-users, but the ones that want fat-free organic chili-cheese nacho pretzels. Don't worry--you'll find them.
The revised "Browse" interface removes Instacart's original table view header from the top of the screen which had contained sub-categories. In my redesign, sub-categories are now grouped together in a horizontally scrollable table view cell. Moving and grouping the items allows for easier access and less scrolling, respectively. My design implementation is non-obtrusive as it allows for users to still browse the current way by scrolling past the new categories table view cell. Furthermore, the "Shop by categories" table view cell follows the same horizontally scrollable design as the other cells in the table view. Therefore, signifying to the user how they should interact with it. I believe that this will be a well-received design feature for users because it is much less overwhelming than the existing "View 100+ more" option.
Users are free to choose as many sub-categories as they like. They also have the option to select all categories. If they choose at least one, an option above the tab-bar will be presented. Tapping "apply" will present the user with items from the selected sub-categorie(s) (just like how Instacart's current design presents users with items from one selected category). However, from this screen, users will now have the option to filter through one sub-category or all sub-categories they had selected. Tapping the "Clear all" removes all selections and hides the pop-up button.
From left-to-right: Instacart's sub-category interface and my redesign for the new sub-category interface (prototyped in Principle).
Note two things in my redesigned gif above. First, tapping on a sub-category now gives users visual feedback as there are selected states. Second, "coupons" now come first in the my redesigned "Shop by categories" table view. After conducting interviews and user flow walkthroughs, people like Fatima really wanted to see what was on sale. Unfortunately, Instacart's original design had the coupons hidden all the way at the end of the scrollable table view header--away from their users. With my redesign, users now have the opportunity to search by coupons. Who does not like a good sale?!
Recap: Revisiting my Hypotheses
It is important to note that my redesign does not add any new standalone features. Instead, it further refines Instacart's existing features. More specifically, it introduces new ways for users to shop in a more efficient, flexible, secure, and powerful way. However, users are still free to interact with the app using Instacart's original design implementations. Below, I address the solutions and their respective value proposition for each of my hypotheses:
1. What changes would make the app valuable to more people, and more valuable to the people who are already using it?
Safe and flexible digital shopping experience
Add items to your cart without worrying about what will happen when they are deleted--you can easily get them back. Found cheaper yogurt at Whole Foods, add it to your cart--don't search for it again.
2. Are there any tensions or inconsistencies with the current interface that affects the user experience?
Eliminate existing barriers by using familiar interactions
Navigate carts and shared carts with confidence without worrying about about accidentally deleting anything. No more modifications on placed orders because you thought you had added something.
3. Is the interface inhibiting users from having the freedom to move through the digital categories (a.k.a. aisles) with ease? Does the app allow users to accelerate the process of conventionally shopping for groceries?
Browsing experience that offers flexibility and ease-of-use
Search the entire "aisle" of your store using the new "Browse" option through the tab-bar. Find everything all at once--no more running into rabbit holes looking for that one thing.
4. Does the app reduce the headache of the crowded and busy grocery store experience, or does it create a different type of negative stress?
Shopping cart experience that reinforces recognition rather than recall
Reduce cognitive load of having to remember information from one part of the screen to the next. Browse and cross-shop items faster than you can actually do it in real life. Want to add everything to your cart and compare later? You can do that.
Reflections, Results & Takeaways
Why did I choose my design decisions?
Without having any research, data, or access to Instacart's engineering, design, and business teams, it is difficult to design significant changes that alter the core functionality of the app. Perhaps the current implementation helps meet business objectives for stores by presenting users with an endless selection of items to encourage impulse buying. Maybe the coupons and "featured items" throughout the app allude to said assumption.
Nonetheless, my redesign does not replace those decisions. I noticed them in my design analysis and designed around them. Furthermore, I focused my redesign on alleviating any negative connotations of looking through a 5-inch diagonal screen to buy groceries by making it easier for users to find and purchase groceries. From my redesign, I learned that grocery delivery services are very different from food delivery services as users tend to purchase more items on the former.
Also, grocery-delivery services provide the option to purchase from more than one store. Therefore, I made myself aware and knowledgeable about the two different types of services and how to effectively cater to the needs of customers who are ordering groceries on the app. For example, that is why I created a cart for recently removed items and classified them by stores (for users who decide to cross-shop). This was, in fact, one of the most crucial suggestions I noticed when doing walkthroughs with other people. Nothing is worst than a user who gives up and decides to stop by the corner market on their way home because the app frustrated them.
Thinking about the platforms
When designing for both mobile and web, it is essential to understand how to maintain consistency between the two platforms. However, taking into consideration their native differences and how that changes the way users interact on both platforms is critical. Some users might find using a mouse and keyboard easy for shopping because they usually shop on Amazon's website. But they are open to a more lightweight and mobile way of shopping on their mobile devices. It is also important to consider those factors and allow the similar shopping experiences to be created on both platforms by using their native functionalities to enhance the user experience.
Each platform has its advantages regarding functionality, and it is sometimes difficult to remember that when designers are striving to create a seamless experience that feels the same on all platforms. However, by aligning our focus on designing a seamless experience while leveraging native functionalities, designers will be able to create value for their users better.
What I learned
- Researching other grocery delivery services such as Amazon Fresh, Jet, and Fresh Direct to observe their advantages and disadvantages provided me better clarity for understanding how and why to choose my redesign approaches.
- Awareness is key when analyzing and solving design pain points. Not all delivery apps can be approached with the same solution. Although I knew some differences between grocery and food delivery apps, I learned more about their stark differences while interviewing and walking through user flows with family and friends. The way I use a service is very different from the way someone else uses it. For example, some people wanted to be able to cross-shop between stores to find the best deals. This epiphany inspired my "Recently removed from cart" feature.
- How to be resourceful about learning how to hone my prototyping skills--from watching YouTube videos, tutorials, and walkthroughs to learn new techniques and strategies. I also learned where I need to improve in my Principle and Framer skillset.
- Understanding how decision theory can be applied to understand or influence real-world decisions of others.
- Research tradeoffs based on my design decisions: By condensing the decision-making architecture for the "Browse" category, I assumed that my redesign would improve people's decision processes. My assumption was based off of the normative model which identifies how to approach the best decision given a set of choices. However, what affect does my redesign have on the engineering algorithms? Can users search through all sub-categories or is there a limit because of the system performance? In fact, is my assumption true? I am very curious about conducting A/B usability tests to see how my redesign can be further refined.
- Engage with different user types to identify gaps: Although my hypotheses helped me further refine some of Instacart's features, I would love to have collaborated with the other designers and engineers at Instacart. There is no denying the fact that Instacart's users in different cities around the country all have unique needs. My redesign may work for users in Manhattan, but what about Boston, MA or Boulder, CO? I am interested in conducting usability tests to observe similarities and differences between both of our design layouts by analyzing conversions, growth, and engagement.
- Design. Build. Test. Learn. Repeat.: For my redesign, I did my best to constrain my design decisions around what I would think would be real-life constraints at Instacart. The purpose was to prevent myself from being romantic and falling in love with my designs. Effectively, I became more critical of myself. In the real-world, how would my redesign do with constraints and objectives from the business and operation team?
- Feedback from actual Instacart Shoppers: I am eager to see how my redesign will further improve the shopper's experience. Throughout my redesign, I was striving to improve the shopper's experience by reducing order modifications. This way shoppers would have more opportunities to fulfill other orders rather than tending to order modifications. However, I would like to be able to test my assumptions through Instacart's Shoppers Advisors Program and learn from the shopper's feedback.
I designed using Sketch 3 and prototyped using Principle.