Watch your (gender) tone

At Textio we often get requests for a “watch list” of gender-biased words and phrases to avoid, so that people can make sure their job listings don’t unintentionally skew the gender balance of their talent pool. Trust me, we’d love to help you with that—but it won’t fit on a sticky note posted on the side of your monitor. Even if it did, you’d be constantly crossing out old phrases and adding new ones.

If you compare the text from millions of job ads to their hiring outcomes (like Textio’s predictive engine does every day), you’ll come up with tens of thousands of different phrases that can impact the ratio of men and women who will apply to your job. And because the way people use and react to language is always evolving, the words on that list are always changing. Those blue and purple highlights in the Textio document you’re writing today may light up a new set of phrases a few months from now, depending on what the real-world hiring data says at the time.

That’s where Textio’s tone meter comes in. But before explaining how it works, let’s talk about how Textio finds and measures gendered language in the first place.

Gendered language isn’t always obvious or intuitive.

Gendered language isn’t always obvious or intuitive. There are some types of phrases that seem to fall into loose groups or categories, like growth-mindset language. But as we’ve pointed out before, there are plenty of seemingly random examples like exhaustive, enforcement, and fearless that are statistically proven to skew your talent pool (toward men in this case), and it’s highly unlikely you’re going to hear about those at your company’s unconscious bias training workshop.

Instead, Textio’s predictive engine uses a combination of natural language processing and data mining to find them and bring them to your attention while you are writing. These are all of the words and phrases that are statistically likely to create an imbalance between the number of men and women who are inspired to respond to your job ad today.

Yes, there is some deep and complex psychology in how people react to these phrases, but that’s beside the point. And no, these aren’t inherently bad words, and people who use them are not bad people — it’s just math. The data speaks for itself, and it’s always on the move.

That’s why we designed the tone meter: it always knows the current data. The tone meter operates independently of your Textio Score. You could potentially have a very high-scoring document that will draw more qualified applicants in less time, but if the tone of your writing is highly biased, then most of those candidates will come from one gender.

Reading the tone meter is super simple:

• The white triangle marker indicates the tone of your listing
• The colored bands represent the degree of gender bias (e.g. the light blue band means “slightly masculine”)
• The central gray section means neutral or balanced (no bias)
• The outer ends of the scale mean your writing is extremely biased toward either women (dark purple) or men (dark blue)
• The dark gray vertical line indicates the average tone of all similar jobs in Textio’s data set, so you can see where you fall in relation to your competition

The best strategy for most writers is to simply set a goal of high-scoring, neutral-tone job postings. If your tone meter is skewed toward the left or right, then look for the blue or purple highlights in your writing. Hover on those highlights with your mouse, and Textio will provide options for replacing that particular phrase.

Once you have reached an even number of gendered phrases, or if you have no gendered language at all, then the tone meter will stay in the neutral gray zone. Content that is neutral in tone tends to recruit the most diverse candidate pool on average, and Textio’s tone meter helps you stay on top of that language as it changes. That way the sticky note posted on the side of your monitor can just say “Watch your tone!”