Data requirements are not standard across the legacy admin systems used by our enterprise clients. A repeatable, non-technical data configuration solution was needed in order to prevent one-off, non repeatable custom development work and to enable external users to own a portion of this work without internal support.
In order to improve how we handled data configuration, I needed to focus on the following:
- Bringing the data requirements phase to the forefront of setup
- Applying data requirements to fields without custom code
- Smoothly assigning fields to form templates
The result was a field management UI for visibility into the exact data being digitized, a drag-and-drop field placement tool for faster template definition, and a backend staff tool for non-technical data configuration - plus a longer term vision for making the process even simpler in the future.
As the company's sole designer, I vetted and championed the problem with a junior PM and owned the end to end UX of this project. I worked side by side with a scrum team containing 7 app engineers.
The setup process dangerously lengthened time-to-value:
- Requirements gathering focused on Captricity needs (forms to templatize), not customer needs (the inventory of data needing to be extracted, digitized and sent downstream), creating constant rounds of rework for each new use case
- The focus on non customer-centric slowdowns in setup forced customers to focus on Captricity-created problems, creating disenchantment before the project could make it to production
Data configuration requests were not yet productized:
- All requests were custom built in a way that was not-repeatable
- Only internal solutions architects could complete this work and the team did not have the capacity to deliver efficiently for a growing user base
Prototypes and Testing
Prototyping out of the box enrichments:
- Design 1: In-product. Engineering capacity was low and the frontend work was decidedly too daunting to be prioritized at the start of the project
- Design 2: Google Sheets integration. The proposed flow was less conventional and only accessible for staff users. But it was already familiar to staff and straightforward for engineers to build so it was the lowest risk at the time. It would serve as a test to prove if we should continue growing standard enrichments or whether we instead should choose to provide ways for customization to be owned externally
Testing the new user flow of data configuration to template definition
- I shifted the setup flow to be customer-centric and data configuration-centric and validated the improvements with the professional services team
- I conducted a handful of iterative co-design sessions with the implementation team so that they could start to adhere to my new vision for their process as early as possible. This proved to be helpful by the time the tool launched
Final Design and Resolution
Data configuration with enrichments: Google Sheet integration for repeatable, code-free data configuration. Spreadsheet format enables fast bulk configuration of hundreds of fields at once.
- Setup time is decreased and internal staff have more time for higher value tasks
- Digitization costs are lowered because standardized enrichments enable higher AI contribution
Field management: An editable in-product inventory of all fields and their statuses.
- Transparency into project scope for customers and partners who previously did not have definitive access to this information
- New access enables external users to participate in setup more efficiently because feedback rounds are now rooted in an up-to-date single source
Field placement/template definition: A modular, intuitive, and fast UI for mapping data requirements to forms, decoupled from the actual task of data configuration.
- Data configuration happens first, so at the form template-level, the user can focus on placement. This simplified task enables external-user autonomy because they don’t need to understand the greater configuration requirements. The tool is faster and more intuitive, contributing to the decrease in time-to-value
Default view of all fields for an entire account
Breadcrumbs navigation to individual template filter, bulk select options
Template search functionality, individual cell loading
Save all/bulk change loading
Success with a young engineering team:
- Alongside this project, I helped lead the effort to formalize a demo day for beta testing. This helped create a regular culture of user testing with the engineering team
- I created a reusable table UI design pattern. This went on to be reused for a variety of other prototypes and in-production pages. The small engineering team lacks confidence in frontend work, so this was helpful in mobilizing and prioritizing other UI improvements
Marketing UX improvements for prioritization:
- I had been sitting on the field management improvements for some time, unable to get them prioritized. This was the first project where I was able to convince stakeholders that these seemingly un-flashy ideas were essential to bringing value to users and the business. It was a nice lesson of the importance of fighting for good UX