This article originally appeared in The HR Examiner, you can read it here.
HR professionals often disagree about the value of “Time to Fill (TTF)” as recruiting metric. The debate is largely pointless — like TTF or not, it's here to stay. TTF endures because for executives outside of HR, it measures outcomes in a readily understandable way. But measuring your recruiting performance by TTF is like measuring your marketing department by the time from wanting a sale until making a sale.
It encompasses too many factors to be a useful guide to action on its own.
Intelligent metrics help you improve performance and allocate resources more wisely. Ideally, they help make your processes more predictable. Before I got into recruiting, I worked for many years as a systems engineer (or more ponderously, “business process reengineering consultant”). I quickly learned that, “Not everything that counts can be counted, and not everything that can be counted, counts.”
Here is how to make Time to Fill useful:
Break down what you are measuring.
The first problem with the “Time to Fill” metric is that it's too broad — it measures the impact of too many different people. It only becomes useful as a way to hold people accountable when you break it down into the components different people directly control:
- Time to Market — The time from when a job opens until it is advertised “in the job market” (Internal approval process is often under the control of Finance or HR.)
- Time from Market to Slate — The time from when a job is marketed and advertised until a reasonable slate of candidates has been developed for first round interviews. (Recruiting time is usually under HR's control). Usually, this metric is the most reliable and accurate measure of recruiting's speed.
- Time from Slate to Hire — The time from when candidates have been identified until a hiring decision has been made. (Interviewing process time is usually under the hiring manager's control.)
- Time from Hire to Fill — The time from when an offer is accepted until the candidate starts work. (Notice period is usually under the candidate's control.)
In my search firm, “Market to Slate (MTS)” is one of our key performance indicators. We have direct control over it. But if you use it, make sure you have a clear definition of what comprises a “slate” in your organization. Is it 3 or 12 candidates? You are welcome to define it any way you prefer, just be consistent.
We define a “slate” as 6 well-qualified candidates who have already passed a rigorous screening process. Read why in the white papers here.
The second problem with TTF is that different jobs require different levels of recruiting effort. The average is meaningless. We'll get to that next.
Use metrics to plan your actions, not just to report on them.
Don't just look in the rear view mirror with your metrics; use them to allocate resources to your future work. If you are going to use any metric, you need clarity about what you are trying to optimize.
In our firm, we optimize predictability. On the first day of a search, we pre-schedule interviews 3 or 4 weeks in advance, and vary the level of recruiting effort we expend to ensure that we keep the search running on schedule. We plan our future work by looking at the level of effort required in past searches (time needed, number of candidates interviewed, etc.) Candidates all respond differently to recruiting approaches, but if you know what level of effort it takes to get in front of 50 people, you can plan your day effectively.
You might use your metrics differently. Perhaps you will want to keep your level of effort consistent, and use your metrics to better forecast when managers will be able to interview, or perhaps you will use your metrics to project the business impact from increasing your recruiting resources.
But the single absolute worst thing you can do is to gather all the information you need to calculate Time to Fill, average it together across different jobs and departments, and go compare it to the Time to Fill reported by some other organization. That's metrics malpractice.
That's like me gathering income data from Bill Gates and reporting that, on average, we are both billionaires — it's accurate, but not helpful.