Who is Our Audience?

As we worked to narrow down the problem statements to a specific audience, pain-point, hypothesis, and desired outcome, much of my early time was spent helping my product partners analyze and diagram current workflows to help identify areas of impact and audience overlap. These processes spanned many systems, teams, and technologies, and analytics were difficult to come by. I interviewed stakeholders and Operations staff and compiled diagrams to help in planning sessions. There was high ambiguity and a risk of trying to do too much at once, but I supported conversations with visuals and helped them narrow down our direction.

Client First

While some of the problem areas focused on internal audiences, we ultimately decided to focus on our client experience first. We needed to design and deliver a considerably more user-friendly experience to help our clients discover our existing integrations which represented a substantial investment. We also had to do it with as little impact as necessary to our internal operations processes, and of course, we needed to deliver as soon as possible. Analysis of our target users gave some clarity on who we would be serving and what their needs were.

New Client Experience—Same Operations

My Product partner and I brainstormed our approach to combining new with old and I put together diagrams to help him describe our tactics to the development team. This helped them see how we would initially leave Operations alone but begin building a new experience and possibly a new source of truth for any potential enhancements to our current data. Eventually, we would also need to build a new internal tool for Operations to manage all of the data in one place. See my Third Party Management project for more about that.

Lo-Fi Wireframe Explorations

How might this fit into our existing product navigation? How might we structure pages to allow us to scale the content and functionality? What types of information might we include based on our understanding of competitor marketplace UIs? These were some of the ideas I mocked up in response to our competitor explorations, stakeholder interviews, and internal user interviews. These mockups were more of a dream-big exploration, but they helped us to look more critically at existing data and give stakeholders confidence that we were hearing them even though we might not build certain features for a while. They would also help shape our roadmap until we could validate with user research.

Current Experience

We already had a “Marketplace” product, but very few clients could find it or even had access to it. There was no question that we would need to rebuild it even if we delivered parity to the current functionality. We knew the content wasn’t providing enough value and that search/filter capabilities were sub-standard. It also didn't follow our internal design system, adhere to established guidelines, or match the rest of our product suite. We would also need to use the data we currently had but consider the need for enhancements to content and information architecture.

MVP Explorations

Next I needed to demonstrate a possible MVP design using our design system components, patterns, and standards. I proposed using our sidebar navigation as a filter which hadn’t been done before, and I worked with the Design System Team to introduce the Tile List which combined aspects of our Image Gallery, Data Table, and KPI Tile components. The inclusion of a vendor logo for each integration was an enhancement I pushed for as it acts as a visual identifier and makes the layout much easier for users to scan.

Feedback is a Gift

At this point we needed to do research to get feedback from clients on our initial designs and to validate their current practices and preferences around integrations. Remote, moderated user research sessions with ten client HR admins from multiple industries gave us the feedback we needed confirming that in the future we would need more robust filtering capabilities like product, industry, and company size as well as the ability to sort by popularity, relevance, or alphabetically once we had the necessary data.

A review of our integration names, categories, and descriptions also uncovered inconsistencies in purpose and relevance to helping a user discover and understand the data. Our research also confirmed the additional content that we would need to include to help users make integration selections with greater confidence and without generating a support call. We were able to justify additional projects to refine and improve our existing integration marketing content while developers began building our first iteration of the new search experience.