The strongest diversity hiring initiatives begin at the top of the funnel, before prospects even apply. Most channels (eg referrals and inbound) can be less diverse, so if you’re not actively sourcing and nurturing a diverse talent pool, you may not see a diverse pipeline, a diverse set of interviews, or a diverse team. It's hard to assess the effectiveness of diversity-focused hiring strategies and fix potential problems in the hiring funnel without being able to measure diversity from the very beginning of the hiring pipeline. Our goal at Gem is to help teams track and analyze those efforts.
How do I use this information?
Gem offers race/ethnicity insights for talent teams to track how their work affects racial/ethnic diversity through the recruiting funnel, from first outreach through conversion to hire. Use these insights to:
- Catch whether certain groups are disproportionately dropping out of the funnel at a certain stage
- Forecast how many reachouts are needed to convert one underrepresented hire
- Understand via top of funnel metrics why the team did or did not hit diversity hiring goals
- Inform hiring managers about diversity breakdowns in historical data
- View difference in pipeline between focused effort on diversity vs sourcing through unfocused methods
- See whether initial outreach is alienating a specific population
In Gem’s Outreach Stats, talent teams can view how their efforts are faring across different racial/ethnic groups, from first touch through to replied, interested, and even converted to process. Greenhouse customers with Pipeline Analytics enabled can analyze conversion rates through the funnel for the same racial/ethnic groups, and even secondarily group by gender or other attributes.
This kind of information can shed light on whether teams are reaching out to a diverse talent pool to build a diverse pipeline, or whether there are systemic biases that might show up as some segments of candidates get stuck at certain stages of the funnel, or don’t even respond to outreach to begin with.
How does Gem get this data?
Prediction is determined by a candidate’s first and last name via a model trained on large datasets. You can expect our race/ethnicity accuracy in range of 75%-95%, determined using confusion tables with large datasets of self-ID (eg US Census data). Gem’s race/ethnicity diversity is broken into four categories with one undetermined category: Asian, Black, Hispanic/Latino, White, and undetermined.
Import Self ID
If you collect self ID data with Greenhouse's EEOC form, you can grant Gem permissions so you can overwrite predictions in Pipeline Analytics. Ask your CSM for more information.
Why is this data shown in aggregate?
Because this information is for directional guidance on pipelines rather than tracking candidates on an individual basis, race/ethnicity predictions will be viewable only in aggregate in Outreach Stats and Pipeline Analytics. The race/ethnicity prediction will not be displayed on candidate profile, and is not editable. Aggregate stats are not visible if the sample size is too small.
Interested in tracking diversity across your entire funnel, from sourcing to hire? Contact firstname.lastname@example.org to learn more about custom fields & advanced reporting. Click here to learn more about DEI in Gem.