I did some competitive analysis to see what competitiors are doing to solve similar problems.
Deep Dive on multiple hours
As I was designing the interface, I did some analysis to identify the most common operating hours for most local businesses and identify split hours patterns.
Key Differentiating Features/Interactions
Designed a smart seller self service tool which pre-fecthes and displays as much information as possible through registration and vetting process.
Any additional information needed by seller is clearly seperated and opens as a model which keep the user within the context and easy exit path in case it was an accidental click. This keeps the users in confident and helps a faster information input.
Self Service Tool
I designed a Pogressive information addition mechanism to reduce ambiguity to the seller and give them complete flexibility.
Example of this interaction model is
Asking seller to add atleast one category and giving them options to add more if they want to.
Starting with minimal requirement and letting the sellers add more photos.
This way I reduced ambuigity for sellers and gave them control of the interface.
Progressive information mechanism
Quick autosuggest and category prediction based on the user input and quick check on the business vetting information.
Introduced a quick way to add operating hours information with intutive prompts and making smart predictions based on user's input
Smart business hours tool
When sellers input the "to" information as "AM" I proposed changing the "from" to have "PM" first for proximety and quick selection.
This way less scanning will be required and quick selections are possible.
Proposed auto suggest for time based on one number input from seller. Example, if seller input 9 we would autosuggest them all permutaions and combinations of 9 with am and pm matches to select from.
At the same time based on the data I identified that some sellers operate 24 hours or have standard 9am to 5pm timings. For quick selection for such sellers I proposed having them always in the choices.
Introduced inlined way of adding, editing and displaying split hours.
Minimal Registration Form
Designed minimal registration form with segmented information chunks so the seller is clear about what information is expected from them and doesnt feel overwhelmed.
Added examples of each information type needed so there is less conginitve load to the user.
Wherever possible proposed pre-filling of information to reduce the data input from sellers.
Average time taken by each seller during usability testing to register this information 5 minutes.
Amazon sells everything from A-Z online but does not have a strong local market. Most of the sellers who sell at Amazon are virtual sellers buying things from one place and selling online through various channels.
Amazon neighborhood was the means though which we wanted to support and integrate local seller into the Amazon ecosystem.
Amazon seller self service tool is a platform to onboard local sellers, give them control to add/edit their business information, add brand and products information etc.
Collaborated with stakeholders to understand the requirements. Based on the press release document which Amazon uses as working backwards and the product requirement documents, we conducted user research with end users to understand who our sellers are?
As a part of research conducted field studies with local sellers from different categories. As a part of field studies did contextual interview 1:1 interview with sellers and a focus group discussion. Based on all the data, created few personas to understand who we are are designing for and designed a high level seller experience map.
I owned the end to end seller design experience which included self sevice tool, seller onboarding, brands and products tool for the seller and was responsible to design both web and mobiole version of the tool.
First I identified all the possible entry point for the seller registration within Amazon ecosystem and created the interface map to understand information flow.
Based on the finalized mockups from stakeholders, I created clickthrough HTML prototype of the interface. Once the interface was ready, I conducted usability testing in coordination with the UX researcher.
8 local sellers from different product lines participated in a usability study. It was a task based usability testing where local sellers were asked to register their business information.
The goal of the testing was to collect the average business information creation time data each seller takes and optmize design accordingly. The average time each seller took to create a fully loaded business information page was approximately 15 minutes which included accessing images from their website or dropbox and uploading them. This was within my design goal hence no major design changes happened except few verbiage changes.
With my design process and approach I was able to achieve the 30 minutes business registering goal.
back to portfolio
I was working on the local seller experience end to end. The biggest challenge I identified during research was local sellers who have brick and mortar setup are not very motivated to go digital due to lack of time and skills
Technology experience ranges from novice to expert
Local sellers have time constraint because they need to deal with many things with in store and dont have time to invest on setting up online store and product information and was looking for a less time consuming interface keeping in mind their skills and expertise
Local sellers have varied product range most of it is by local designer or local brands so my biggest challenge was introducing a tool which is easy to add brand and product information
Local sellers operate different business hours different days and dont have one set of time. Designing a system which can accomodate their business needs was a challenge
Different product to brand ratios
The initial pushback from stakeholders was to retain the registration and business detail page (BDP) information from local team to reduce development effort.
The local BDP page was very overwhelming for sellers and complex to use.
In addition it was not covering all the use cases we wanted to solve for.
I identified too many usability issues trying to use the existing form, like:
Information heirarchy was not clear
Adding business hours was too complicated with multiple form feilds. In addition it did not solve for multiple business hours scenarios
Adding multiple images was not supported in this form
This form did not support adding business categories which was one of major features for our project
Based on my research I identified that many sellers have multiple operating hours based on day of the week. As shown above I had many use cases to solve for like multiple split hours, 30 minutes increments, 15 minutes increments
Complicated taxonomy with up to 6 levels of heirarchy
Each level has up to 20 sub categories
the leaf nodes has up to 50 products
relying on internal brand service tool to store brands information, which does not associate brand to product information
In our legacy category system there was no way to let local sellers add their own business category, but rely on amazon taxonomy which is huge list with multi level heirarchy. Using the taxanomy as is would increase time for local sellers to add business information
To retain existing local forms
Local sellers have multiple business hours which we had to support
Limitations with our legacy taxanomy tool
Project: A multi channel application for onboarding local sellers into the Amazon eco system.
Platform: Web, iOS and Android
My Role: I owned the user expeience of Amazon Local Seller Tool end to end and was responsible to design interactions. I worked along side a UX researcher and participated in the user research of this project. While he was helping me with logistics and recruitments etc, I took the lead on moderating the susability sessions, doing 1:1 interviews with sellers, paper prototype testing with local sellers etc. Handing off the design the developers in a understandable format.
Amazon Neighbourhood Seller Tool
Seller Tool at a glance
I created storyboard from my design ideas and conducted brainstorming sessions with my team members to get some initial critiques and direction for my designs. This helped me validate pros and cons of each choices and helped me choose the right design direction.
My goal was to design an intutive application which enables local sellers to feed in their minimal business information within 30 minutes. (During research many local sellers mentioned their business has time constraint and hence anything above 30 minutes is a too much ask from them. This was major reason for their lack of interest or motivation)
How did I measure this goal?
Through task based usability testing with sellers from different background, skills and expertise and time boxing the entire process.
High Fidelity Wireframes
I then designed the wireframes and visuals design for the seller tools and reviewed with my stakeholders to get signoff on the designs. I designed both mobile and desktop version of the tool.
Through out the high level designs there were many areas which seems minor features like adding hours, categories etc which was real complex given the seller type, their operating hours and our legacy tools which made it challenging. The entire eco-system was very restrictive and we had too many limitations. With the right data and research I was able to push my design ideas.