Who cares about cost per hire?
Graeme Wright argues that in measuring recruitment performance, price and value are not the same thing.
Nowadays there are few people working in recruitment who do not talk about HR metrics and in particular 'cost per hire' (In the last week alone I have heard from three clients, one colleague and two representatives from the national press about the importance of reducing cost per hire). There are obvious reasons for the popularity of this concept: it is simple to understand, apparently simple to measure and demonstrates a clear link to improvements on the 'bottom line'.
It is easy to imagine a Chief Executive calling for a significant reduction in cost per hire and, equally, the pride on the face of the HR manager who delivers it. However among the many problems with this measure is that almost anyone, if given free rein in an organisation, could reduce cost per hire - but in so doing the organisation could well be creating a whole range of new problems.
I have taken this opportunity to highlight why I think cost per hire is such a poor measure and to champion a different metric - quality of hire. In many ways, it may seem obvious but, given the current prevalence of 'cost per hire' talk, it's worth mentioning.
To give the discussion some context there are four key HR metrics:
- cost per hire
- speed of hire or time to hire
- customer satisfaction
- quality of hire
It sounds like a truism but of these the last is arguably by far the most important - does it matter what the process costs, how long it takes, or whether it meets expectations if the company is hiring the wrong people? So let's consider cost per hire in more detail. As we have said, its key strengths are ease of understanding and measurement; indeed some very sound methodologies have been established to assist in its measurement. To quote the Saratoga Institute.
CPH = (AD + AF + ER + T + Relo + RC) x 110%
NH
where
- AD represents ad costs
- AF is agency fees
- ER is employee referral costs
- T is travel
- Relo is relocation costs
- RC represents the fully loaded cost of all recruiters time (10% is an historically derived figure for all overhead and administration support)
- NH is the number of hires
However this formula also highlights one of the potential weakness of the metric - a lack of consistency in its measurement between organisations or, more likely, poor measurement. How many companies, particularly in the UK, take into account
all the components that would make a accurate cost per hire metric? This is not the only weakness, either.
1) It can very easily be misleading: an aggregate company-wide figure tells you very little and hence the measure should only really be used to measure on a like-for-like basis.
2) It encourages poor practice: f inding great people is seldom easy or cheap. Hiring walk-ins and poor-quality applicants 'no one wants' is cheap.
3) It is often misused: the metric 'cost per hire' is far stronger if it is used in conjunction with 'quality or performance of the hire'. But quality hires are not often low-cost hires (they are, though, the most likely to produce an attractive ROI). If too much emphasis is placed on cost per hire, it would be understandable if an HR department got into the bad habit of 'hiring cheap'.
3) 'Hiring cheap' can send out a cheap message. Potential candidates often judge a company by the quality of its communications both in terms of design and content. So being 'cheap' may send a negative message to applicants and so, in trying to cut cost of hire, a company may inadvertently be reducing its ROI by scaring applicants off.
4) It does not tell the whole story . There are many aspects of the recruitment process that are not measured by this metric: quality is the most obvious, but time to hire and customer satisfaction are also important. An emphasis on cost per hire could easily mean a company ignores declining customer satisfaction, applicant frustration, declining ROI and decreased retention, for example.
5) There are more appropriate measures. Foremost amongst them is quality of hire. Arguably (and it is a very strong argument) this is THE key measure of the effectiveness of a recruitment programme. Cost per hire, time to hire, numbers of hires and satisfaction are largely irrelevant if not balanced against quality of hire (although we should also bear in mind that quality of hire is not just the quality of the individual, it also encompasses the quality of the total process used to hire them). One of the reasons this is not as widely used as the key recruitment index is that is not the easiest metric to measure; indeed it is not itself a single metric at all but a number of interconnected indices:
- Match against original role description / competency specification
- Performance
- Appraisal and assessment
- Salary (initial and rises) and promotion
- Retention
Each of these components can be measured in short-, medium- and long-term contexts and - importantly - can be linked to an idea of return on investment. While measuring these factors may appear more complex than simple cost per hire, they are the decisive indicators of true recruitment performance. Significantly, the trend towards e-recruitment will make data on these variable easier to capture and correlate. But that's another story. Conclusion
I would suggest that any talk of 'cost per hire' is heavily qualified and emphasis is put instead on those measures relating to quality of hire. Only in this way will employers get an accurate idea of the true return on investment from their recruitment pound.
Graeme Wright is a media and research specialist in the knowledge workshop
at work.
