Why Care About Metrics? Part II

Part I of this series provided a perspective on the importance of interpreting metrics to make business decisions. Essentially, it focused on have a sense of what you want to measure and the decisions you want to make with the measurement prior to collecting the metrics. Part II will review measuring the effectiveness of recruitment programs, the metrics that will help you get there and how to interpret them to make future spending decisions. There are two approaches to measuring the effectiveness of your recruitment programs: Cost per hire and Cost per Advertising Medium. Cost per Hire Cost per hire seems to be the critical measure used by most companies to determine effectiveness of recruiting dollars. Like all metrics, cost per hire is a good measure if interpreted appropriately. Most companies look at an average cost per hire across all hires. I don’t find this to be a very useful statistic. It simply gives an average of all spending across all hires, which doesn’t tell you much about how you should spend your money in the future. A better statistic is the cost per hire for a group of like positions (software developers, database administrators, sales managers, accountants). This information allows you to determine which positions are costing the most to fill and why. For instance it may be that the higher cost groups are harder to fill positions, with skill requirements that are more difficult to find in the marketplace. Hence, the high cost may be justified. Or the issue may not be related to challenging-to-fill positions at all. It may be an internal issue. Possibly the hiring managers are looking for 110% of the skills required and are rejecting too many candidates. Or it could be that the job specifications were not communicated well enough to the recruiter, resulting in mismatched results. These are internal issues disguised as spending issues. Working more closely with the hiring managers to better define needs and required vs. desired skills, may reduce the cost per hire significantly without restructuring your advertising focus. Cost per Advertising Medium With the exponential growth of job boards on the Internet, recruiters want to know where to spend their money most effectively. In determining the effectiveness of Internet recruiting, several factors are very critical — your job ads need to be descriptive, have job titles that compel candidates to read the ads, and be updated on a regular basis. Assuming the above is all done correctly, you now have the challenge of determining from which board the resume came. When polled, most candidates do not remember where they saw the ad on the Internet, only that they saw it on the Internet. If a candidate responds directly from a job board, then the name of the board is often in the response. This provides some help. If candidates send e-mail directly to you, they often do not include the name of the board. Even if you specify in the job ad that they should respond with a specific job identifier, which indicates a code for the job board, only 50% – 75% actually include the code in their response. Sometimes the candidate sees the opportunity on a job board, goes to your site to learn more about the company, and then submits their resume directly on your site. The result: It is difficult to get a true measure of where the candidate originated on the Internet. However, assuming that you are very, very lucky and can get accurate statistics on candidate origination, how are you going to measure effectiveness? I believe that the best measure of effectiveness on a job board is not the number of hires that you got from the board over a period of time, but the number of qualified candidates. Actual hires is dependent on too may outside factors, the most important being how fast you contact candidates once they respond to your ads. If a job board is consistently sending you a flow of candidates with whom you want to pursue further discussions then it is successful. In this case the cost measurement is “cost per qualified candidate.” In making comparisons between boards, it is important to compare apples to apples. A very targeted job board like Oraclejobs.com should not be compared with the larger general boards like Monster.com or CareerMosaic unless you are comparing the costs per qualified candidate for like postings. This can be very tedious but gives you the most accurate results. A word of caution, do not get lost in the numbers. You may have spent five thousand dollars on one site and only hired one person. On paper this looks like a failure. In reality, the person you hired may have been for an extremely difficult position to fill with your only alternative being search fees at four times the amount you spent on the job board. These same metrics should be considered when using other advertising medium like print or career fairs. A print ad one month may produce very little but the same ad in another month may be very productive. To summarize, the key points to consider when measuring effectiveness of advertising dollars are:

  1. Compare the cost per hire for like positions and not an average cost across all positions.
  2. Article Continues Below
  3. Peel back the onion to determine if the costs were justified.
  4. Evaluate the situation to determine if you can reduce costs by streamlining processes rather than changing advertising programs.
  5. Consider cost per qualified candidate, rather than cost per hire when evaluating the effectiveness of particular medium.
  6. Keep perspective, metrics are guidelines not answers.

Karen Osofsky (karen.o@tiburongroup.com) is a co-founder of TiburonGroup.com, an e-recruiting consulting firm that provides outsourced recruiting solutions to rapidly growing companies and new ventures. The firm provides a broad range of recruiting consulting, sourcing, screening, and strategy development services to help companies manage the front-end recruiting process. Tiburon Group is a Certified AIRS Solutions Partner.

Topics

Leave a Comment

Your email address will not be published. Required fields are marked *