Automating a data importing process with user-defined rules
Overview
Acoustic is a marketing software provider that empowers marketers with tools to streamline their work. Acoustic users regularly update information about their customers by importing data. I was tasked with streamlining this process through the design of a new feature.
Impact
Through multiple iterations and user research, I designed automated data importing, which reduces the number of imports and creation time by over 50%.
Problem Breakdown
Manually importing data takes too much time
Keeping databases up to date is a hassle. Since marketers receive new data regularly, a lot of time is spent setting up individual imports. Additionally, the import process is five steps and takes manual effort to complete. Having users complete these steps for every import is redundant and inefficient. Overall, frequent imports demand significant time and effort.
Five steps with multiple fields and inputs for users to complete per data import.
Design goals
Exploration
Cutting down the number of imports
After exploring a couple different approaches, I created a way for users to automate their workflow. Users were now able to create automated jobs, which created manual imports automatically without future user action.
Time scheduling
The need to define numerous variables, such as “start date,” “end date,” and “import frequency,” made scheduling difficult and inflexible. Marketing needs are often unpredictable, and this system couldn't handle urgent or precise data requests.
Wildcard characters
After exploring a couple different approaches, I created a way for users to automate their workflow. Users were now able to create automated jobs, which created manual imports automatically without future user action.
Decisions
Setting up automated jobs with folder monitoring
After exploring a couple different approaches, I created a way for users to automate their workflow. Users were now able to create automated jobs, which created manual imports automatically without future user action.
This approach does not use wildcard characters and require users to remember file names. Files that meet the defined criteria are imported automatically, which allows for urgent requests.
To determine which files within the monitored folder to import, the automated job followed a set of rules centered around file names. Users customized and specified these rules with flexible options and text inputs.
Viewing automated jobs
To accommodate the addition of automated jobs, I added a tab to the already existing data overview table. I split the jobs into imports and automated jobs. Since automated jobs and data imports were so different, the information on the tables needed to change as well.
This view allows users to switch between two tables of manual imports and automated jobs.
After clicking into an automated job from the tables above, users can view more detailed information about each job.
Pausing and editing automated jobs
Since automated jobs are continuously monitoring folders for files to import, I created different ways for users to pause the automations just in case. The first way to pause was from the data overview page and the second was from the detailed view.
I added an edit button next to the pause button in the detailed view to allow users to go back through the set up flow to change up the rules or name. Users could only edit automated jobs that were paused first.
Creating statuses
I designed status tags to show users if automated jobs were paused or active. These statuses looked and behaved the same as the statuses for data imports: In progress, completed, and failed.
Design Validation and Handoff
What questions do we need to answer with testing?
To uncover what pain points occur within my newly designed flow, some user testing was needed. After communicating and working with a UX researcher, we landed on a couple questions to target with our research:
While analyzing research sessions, I found low engagement with the new feature. This prompted a user-friendly rebranding and the design of a simpler onboarding process.
I provided engineers with detailed, flow-organized handoff files with documentation for each screen and worked closely with them to make sure they understood all parts of the design.
Final screens
Set future imports
Complete control
New data overview
Learnings
I took the time to thoroughly understand the design system and all of its components, which helped me tremendously in this project. By using existing design components, I was able to speed up my design process and ensure design consistency. Additionally, I gained valuable experiences in the end-to-end design process, from conception to launch.
Product delivered
The product I created has been delivered to be developed and I was lucky to be able to collaborate with so many different teams. I was able to get the perspectives of writers, researchers, developers, and stakeholders which really shaped the product. Tying these learnings together was something I wont forget and will remember in my future work.
