Remote Frog

Everyone wants a measure of Quality of Hire. Lots of us claim to have it. And almost nobody defines it the same way. This is because the question is reasonable, but the shortcut is dangerous. Organizations want to know whether hiring is working. And they want to reduce this question into a single score that captures the value of a person (or a set of persons), a role, a team context, and a future contribution all at once.

We aren’t measuring persons

The first thing worth stating is obvious but easy to forget. QoH is not a persons’ measure, but a limited attempt to estimate whether a hiring process is producing the outcomes the organization hoped for. This reframing matters because we’re not ranking humans. In the worst case, we would be grading the appropriateness of the hiring team’s processes for a given role.

We use it to improve the process, not to make people more productive

The best use of quality-of-hire data is not to label individual hires after the fact. It is to learn which sourcing channels, assessments, interview patterns, or role definitions are associated with better outcomes over time.

That shifts the question from Which person got what score? to What are we learning about how we hire? Much healthier. Also more useful.

Also, and this is not negligible. You can improve what you measure, and QoH is not about “improving people” but hiring processes. If we measure facts we want to improve -not persons’ traits- we can deploy actions in our hiring pipeline with the goal of improving these results. And we can play to evaluate the impact of these actions (accepting changes will also be influenced by a myriad of other things, a good number of them you have no agency over)

What’s QoH composed of

Let’s start with the most popular factors, from my conversations in People Analytics circles. We must add retention to the recipe, but retention alone is not enough. Hiring managers’ satisfaction matters, but alone is not enough. Early performance matters, but performance ratings alone are not enough.

(*) Another question would be how do we factor retention, manager satisfaction and early performance in, as there are tons of different ways and of course, each one has the potential to add a different spice / bias to the result. Special note to performance: we need to be strong to resist the temptation to reduce it to a number that can be misused.

We can fine-tune what QoH means for us by including a small set of super-simple outcomes that reflect what success in the role actually means. It may depend on the company, the role and the moment, but could include

• Time to effective contribution. Better if we can define what effective contribution is for a role. Problematic if we want to unify this indicator across different roles in a company.
• Performance after a meaningful ramp period. Gosh! This is something I read in a source -let me not disclose where. And… after careful and long deliberation, I still don’t understand what this exactly means. But I left it as a valid reminder and a note to myself 😀 No matter how many times you see something in documentation or references and how relevant it looks: If you don’t know what something is (aka if you cannot define it in words your teammates understand), then skip it.
• Retention at a relevant milestone. Better if you can define why this milestone is relevant.
• Manager or peer assessment, used carefully and always knowing this is a subjective measure that won’t be used for Performance Reviews.
• Role-specific signals tied to real output. We tend to directly think of outputs, but the process matters too as output can depend on the team or projects the new hire is taking part on, their team leader instructions… I’ve also experimented with other proxy measures like communication frequency, flows and impact, over different systems like GitHub for developers, project management systems, internal documentation, Slack channels…- (See Communities Analytics here)
• The perceived fit and engagement for the person we hired.
• …

Avoid false precision trap

Putting everything into one decimal-heavy score can feel rigorous. But often it is just making up uncertainty. Be sure we always keep in mind how that number is exactly calculated so that we don’t attribute magical powers to it.