“Anyone can make the simple complicated. Creativity is making the complicated simple.” That quote, attributed to jazz bassist Charles Mingus, perfectly encapsulates a constant tension in the design and building of Textio: how can rich and complicated data sets be made easy to use?”
Textio is a product where the writing experience is paramount. If the interface was too noisy, you’d be distracted by the guidance and that might negatively impact your writing. Too calm, and you might miss recommendations that will help you write more effective job descriptions.
Think about traffic signs: some signs want to tell you what lane to be in very soon, and some want you to know how long it is to the next rest stop. Some want you to know what the speed limit is, and still others want to make sure you know what highway you’re on.
A speed limit sign would be akin to a simple spellcheck squiggle: a basic device to convey a single bit of information: you’re going too fast. Or, in a text editor, this word is not recognized in the dictionary. But driving requires a huge swath of information delivered in different ways. This is more akin to Textio, bringing rich data sets, and using sophisticated coloring and interface to signal you to different sorts of guidance about your writing.
The on-canvas highlights are a good example of underlying complexity expressed in a simple interface. If you see a green outline, for example, that means Textio has uncovered an opportunity to replace your original choice with a word or phrase that will attract more qualified people to your listing. Imagine this as a supercomputer predicting a chess game: but instead of tracking each move and predicting the next, Textio looks ahead hundreds of thousands of moves. That means those green outlines are only the most potentially impactful change: the things you didn’t write, but could have.
Or think about the words that Textio doesn’t highlight. Just because a word or phrase isn’t called out doesn’t mean it’s not evaluated; if Textio showed all possible guidance at all times, every word in your document would have some kind of markup, which would be overly distracting. Textio’s guidance is always focused on helping you improve your writing in the most contextually impactful way, by making choices of how much of its findings to reveal. It’s all relevant to your document at the moment of your writing process.
So, Textio strikes a balance between what to show and how to show it, but what about the data driving the recommendations?
A simple-looking highlight doesn’t explicitly say so, but Textio arrived at suggestions through analzying millions of job descriptions (over 300 million and counting, at a rate of about 10 million per month, as of this writing), and their outcomes. How fast was that position filled? How many applicants were there?
Location matters, too. Did you know a Project Manager job description set in Dallas will have different guidance than a Project manager job description set in New York City? The data under Textio’s platform measures the hiring variables and data by location, to help you fill jobs faster around the world.
Think about those surprisingly accurate quizzes that tell you what region you’re from based on how you use simple words. In the United States, “soda” vs. “pop”, or “tennis shoes” vs. “sneakers” can help pinpoint where a person was brought up — because people in different areas use language differently.
So, of course, it makes sense that regional language use translates to job descriptions as well.
You could imagine a scenario where every time Textio offered suggestions, you’d see a table that showed how many millions of documents were analyzed to reach the guidance. Where the suggestions are stacked all way up to the nav bar, a rich data set of how Textio phrase suggestions were derived. No doubt, some number-crunchers would love that, but for most people this kind of overkill would be a distraction from the task at hand: helping you to write more effective words. It’s Textio’s job to get out of the way so that you can work.
The choice of how much data we should expose is a topic constantly under exploration in the Textio design and building processes. Working on new features is akin to architects deciding how much of the floor should be made from glass: it might be an interesting novelty to look and see the strength of the foundation, but it won’t make walking around each day any easier.