Introduction
I joined a newly formed team with an ambiguous mandate. Through conversations with my product manager, we defined that our goal was to bring a new user group into the ecosystem of the company.

I was used to designing for primarily two user groups: merchants and consumers. The new user group were suppliers and they had goals, motivations and pain points that were all foreign to me.

The three main user group categories.

The first thing I needed to understand was how suppliers interact with our existing users. I had a variety of open ended interviews with merchants and supplies to gather information.

Other than ordering from suppliers, merchants also needed accurate product information. We discovered some pain points the merchants were experiencing:
We decided that we would design experiences for both user groups but to focus on merchants because they were our primary user group.

The most approachable problem to address was the lack of product data. The first milestone of the project was to create a catalog that would surface accurate product information directly from suppliers. The next milestone would be to create an integrated ordering workflow.

We needed to understand the browsing and purchasing interactions between merchants and suppliers now. The main difference we were aware of at the time was ordering in bulk.

The sitemap showing common pathways between the main sections of the catalog.

The initial structure of the catalog was intended to be simple to avoid over-engineering a solution for use cases we didn’t have.

The product page for the catalog was vital. We had to understand all the nuance and details in order to move forward.

There were primarily 2 types of products for the first iteration. Single products, and products with variations.

Product information model. I still think 'Matrix product' is a weird word.

Product pages are extremely common on websites, however the goal wasn’t to make a flashy product page because those are intended for consumers. Merchants cared far less about the aesthetic display of the product, and considerably more about the purchasing information.

The research helped me prioritize prime real estate on the page layout to fit the most important information.

Purchasing information was extremely difficult to gather. We knew which fields were most important, however it would require a series of phases for us to gather all of that information. I designed the product page given what information we could obtain and put thought into how the layout would evolve over time.

The product page designed to fit the information model so that it could apply to any product in the catalog. I hate looking at this backpack.

We assumed merchants just cared about the products, but in the research we found that the supplier was also a significant factor. Merchants developed relationships with their suppliers. There were intangible costs to doing business with certain suppliers, and these costs factored into the decision making.

Suppliers had the most accurate product data. Allowing merchants to import this data into their point of sale system solved a lot of meticulous data entry issues.
Browsing and searching

"Why can’t you just make it work like Google?"

Taking a step back, it was also important to answer this question: How exactly do merchants find products? We spent time to learning about browsing and searching behaviours.

Search UX and autocomplete behaviour is an infinite rabbit hole to design but we needed something reliable for this part of the catalog. I created interaction designs that illustrated how nested information within a product could also be searched.

Different states of the auto-complete feature when searching the catalog.

Browsing was primarily performed within a supplier's product offering. The direction of the catalog would begin with highlighting the few suppliers that were integrated, and creating an experience to browse across multiple suppliers when the catalog started to grow.

First version of the landing page. We had very few things to highlight so it was a bit of struggle.

Grid view intended to show more products.

While the merchant experience was being created in the catalog, I was also designing an administrative tool for suppliers. It allowed suppliers to upload product data, verify errors and configure settings about their appearance in the catalog.

Multiple methods to upload data were a requirement for suppliers.

The ability to review upload attempts and adjust was necessary because suppliers were working with large sets of data and a strict schema.

Lessons learned

Some thoughts

Don't underestimate starting something new

Starting something from scratch involves more work than you originally think and trying to learn new domain knowledge is a gradual process.

Documentation is underrated

Previous underrated by me and my 40 IQ brain. Since there was a significant amount of research, I learned the importance of documenting everything because it’s easy to lose track of all the information. It was also funny to re-read my horribly inaccurate assumptions.

Products become great through iteration

Great products aren't created in a vacuum. They have to be tested, examined and iterated on until they become meaningful. I had to internalize the idea that the products I create are never finished; they are just in another stage of their development.

Research everything

Use research to make your life easier. There was so much ambiguity with the project, and we received so many conflicting directions about where the project should go. It’s important to do research to substantiate which goals are reasonable.