3 Key Traits of Usable Data Products


3 Key Traits of Usable Data Products

In case you missed it, earlier this month MetroStar Systems partnered with George Mason University and Zoomph for GMU’s annual Usabilathon. The Usabilathon, an initiative led by GMU’s Human Factors and Applied Cognition program, provides graduate students with an opportunity to hack client driven solutions.

Usabilathon participants were challenged to build a product that converted data captured by Zoomph (a social media analytics platform empowering brands to instantly and contextually pinpoint relevant audiences and trends) into actionable insights. Through their creations, our usabilathon groups unveiled the following traits that make data products usable.


Simplicity is key when converting big data into an easy-to-use product. Here are two key pieces to help ensure simplicity:

1. Before you define your data, it should be translated into Plain Language.

2. Segmenting data, or skimming metrics down to the bare essentials, are other ways of ensuring that your product appeals to target audiences. Remember that more data does not always mean more value. Rather, relevant data is what resonates best.

Think of it this way: though impressive in size, your warehouse of data is useless if a viewer is unable to pinpoint the value of one metric over another. Our usabilathon groups advised presenting a limited scope of metrics that are specific to your user’s mission (e.g., if you’re aiming to provide useful social media insights to customer experience specialists, hone in on sentiment, geolocation, affinity, and emoji analytics that reflect customer perceptions of a brand).


Data products need to offer more than just descriptive analytics to have a long shelf life – they must offer additional predictions, optimizations, or solutions to particular problems, as acknowledged by Wall Street Journal.

Having understood this, our usabilathon groups followed a systematic approach to building out the wheels and cogs of their project:

1. Identify the mission

2. Identify potential use cases

3. Define personas of an ideal user

4. Build wireframes

5. Performed user testing

6. Adjust wireframes according to user testing results.

Step 5 was critical to ensuring users understood the overall end goal of their product. Without exception, our groups found that users cared more about the results that systems delivered than the bare data it offered. All product owners had to confirm that their systems routed their users to a certain result via intuitive design or step-by-step functionality.


Since 2007, the Internet has seen a 9,900% increase in the use of visualized information, according to Piktochart. Visuals appeal to one of our greatest senses (a human takes 13 ms to process an image, and just 100 ms to attach meaning to it).

Our usabilathon groups experimented with many different styles of data visualization. The key takeaways were:

• Simple is still better. Rather than cobbling together various visuals and analytics all on one page, simplifying things with categories, filters, or other divisions in content works best.

• Use vibrant visuals. Color-coded maps, emojis, and other vibrant visuals help users process data more efficiently.

• Stick to one screen. Though one group fearlessly tested a product that required multiple screens for a full, immersive experience, they found that it was highly impractical for everyday use.

View a full recap of our Usabilathon.