Experiences with Data Entry
With SaaS applications becoming more advanced and processing billions of data points, the input of the data is just as important. When it comes to the government, it means a lot when you receive Federal Funding from that data. I have the privilege to design TAP (Transportation Application Platform), which is a product built by Transit Labs that uses data to pioneer a smarter public transportation system.
I wanted to write about data entry because it was a huge friction point for transit operators, it’s becoming prevalent everywhere, and a challenge of experiential value. The part that I’ll focus on is consolidating complex input matrices using input masking patterns. I’ll give you the quick scenario so we can jump into the fun.
All public transit data is reported to FTA (Federal Transit Administration) through the NTD (National Transit Database). There are many levels of how federal funds are apportioned to transit systems. There are new standards emerging on collecting monthly data, but let’s talk about the large annual report due at the end of a transit system’s fiscal year. For a large report you will need to submit various sets of data and the product will validate over 800 data items. You may see text inputs, lots of numeric data inputs, checkboxes, radios, and drop downs.
Planning & Research
Let’s take a look at the S-10 form on NTD’s site – A matrix of inputs. Keep in mind that for each type of transportation mode (Bus, Train, Trolley) you will need to submit that particular mode’s data point, potentially within the same data item. Imagine you had 3 different writing utensils and at your final exam your teacher told you to answer the each question with every utensil. It’s a very rudimentary example but I hope you get the point.
NTD S-10 form
This just allows the operator to enter the data for just one mode at a time. The hurdle is facilitating a condensed method for submitting a data point for each transportation mode. A few years ago I remember an experiment that used input mask patterns to make form entry a lot less painless. With having that knowledge I knew that it could be extended on a larger scale, turning entry from months into weeks.
Plotting the course.
The initial concepts focused on business hierarchy, layout, typography, weights, and depths. I refine how the input element layout will work before I add pagination of transportation modes and sub-sets.
Please keep in mind that transit operators have a complete understanding of transit acronyms.
High fidelity refinement
Now let’s add some pagination to quickly determine the status of a mode for a particular data point. While we are on the topic of mode selection, I do want to add that in the left sidebar there is a “drop-out” menu to globally change all the inputs to a specific mode for quick viewing or entry.
As for the thin bars and ticks on the right of the sidebar, they are quick visual indications for the status of a data point. This color pattern is applied to typography, vertical indicators, mask titles, input pagination, input border, and input text. Remember to repeat a specific pattern in your application to drive behavior change.
Green = Successful
Red = Validation Issue
Grey = Incomplete
With the method of masking the data points for each mode, colors, and dynamic data – the operator is now able to clearly identify what data point they are submitting, visualize their progress, and have historical data to guide there submission.
Since each data point has historical data, the table for each data point changes depending on the data point that you are focused on. In the end, this is still a lot of data entry but the nifty thing with this project is that there is some form automation and verification functionality. I hope this can inspire more efficiencies.
Please let me know your thoughts on data entry.