article by Dr. John Sullivan & Master Burnett To the average human resource professional, it might seem as if everyone in the corporate world is on the verge of information overload these days. No matter where you look, data is being collected, manipulated, and pushed back out. From computer generated reports sent via e-mail to status updates sent to your Blackberry, massive volumes of information are swarming around. Most of this information is utterly useless; after all, it does nothing more than tell you about historical performance according to some oversimplified formula that in reality tells you little about how to do your job more effectively. That said, information in general can be very powerful ó if the type of data collected and the method used to collect it are accurately aligned with the type of decisions you will need to make upfront. Unfortunately, most measurement systems start with what data is easiest to collect. This article is going to focus on world-class measurement related to making corporate employment sites more effective, but the concepts presented here could be applied to almost any topic. Reporting versus Operational Metrics The problem with most corporate recruiting metrics is that they are designed to report up in an organization, not power the day-to-day decision-making process where small corrections are needed to keep a fast-moving organization on track. These types of metrics are commonly referred to as “reporting metrics,” and are most often reported out on a periodic basis versus a timeline associated with operational actions or event-dictated responses. They are great at telling you that something isn’t working, but horrible at telling you why it isn’t working, or where your process to achieve certain results went awry. This results in most reports being deemed not relevant by those who need the information most. Before you go saying that’s not true, think about how many reports get circulated in your organization, and then think about what percentage of them you actively refer to when making a decision. Most data suggests that in an average organization, only 17% of the information made available via standardized reporting gets used! Making Web Measures World Class and Usable The keys to making web measures world-class are pretty much the same as those required to make any set of measures world class. They include:
- Making the information provided actionable. This edict cannot be repeated enough. To make information valued in an organization, it must be presented in such a way that the recipient can immediately realize how to make use of the information ó and how to act differently.
- Keeping the information relevant to the ultimate goal. Many metrics exist that are interesting but not relevant to any particular business goal. Since business professionals are often evaluated based on specific objectives, getting them to pay attention to information requires that whatever information you want them to look at be relevant to some aspect of the job they are being held accountable for.
- Putting context around the information. By itself, most information is useless unless you provide something to compare it against that gives it context. In an ideal world, all corporations would compare their metrics to that of a select group of competitors, but we all know how difficult getting such information reliably can be, so you may have to settle for industry numbers.
With these three keys in mind, let’s look at the most common employment website numbers. The Most Common Corporate Employment Web Measures When corporate employment sites became popular in the late 1990s (yes, it was about one decade ago) the leading business professionals in recruiting knew that measurement would play a key role in making this new mechanism a key tool. To that end, they approached other web professionals and adopted a series of metrics similar to those that power most websites, including:
- Volume of unique individuals visiting each page of the site
- How long they stayed on the site
- What page brought them to the site, if they started their session on any page other than your home page (also known as a landing page)
- What other sites are linking to your pages
- What sites visitors were coming from
By themselves, each of these measures tells you something, but none rise beyond the level of basic reporting metrics. In short, they fail to demonstrate the keys to being world-class on the three counts listed above. Unfortunately, many organizations are stuck at this level. (It is important to note that these statistics do not include those that an applicant tracking system would provide. Those metrics are outside the purview of this article, because in theory, if not in practice, an ATS does not constitute an “employment site” but merely serves as a tool to help make better usage of the information collected by the site. For lack of a better word, most corporate recruiting sites “s**k” and have become nothing more than bland front-ends to ATSs and RMSs, but again that is not the topic of this article. For more on that topic, please see this article.) Transitioning Into World Class Starts with a Goal To move beyond the archaic world that is common web measurement, we must first realize what goal the corporate employment site helps to achieve. Luckily, the answer to that question is commonly known. We know that the site exists to drive qualified applicants into the ATS so that they can become candidates. In the world of sales, the act of taking a visitor to a website and making them a participant in the end goal is referred to as “conversion.” Therefore, we can extrapolate that the most valuable reporting web metric would be conversion rate, i.e. the percentage of qualified web visitors who made it through the process to become applicants, as compared to the average or the rates of other known firms. With this in mind we can now turn our attention to what information might help us make better decisions to impact this metric. Does the following make sense?
- What paths, i.e. what pages, do people visit that most often result in conversion, and in what order do they visit them in?
- What paths do web visitors take that least often result in conversion?
- How does the user experience or information presented differ between the two?
By looking at the paths that regularly do not lead to conversion and attempting to fix them, you can positively impact the end goal. Improving the conversion rate of a corporate employment site is not always an easy job, especially when your employment site is well integrated into your corporate site, because politics often come into play. Some things to keep in mind:
- One of the best ways to determine what might be “offramping” visitors from your site is trial and error.
- Don’t change a bunch a variables at once; it makes it nearly impossible to determine which changes had the desired result.
- The most common problem is that the variables present in high conversion paths are missing altogether from lower conversion paths.
- Your employment home page or landing page is the most critical page under your control.
- Building a corporate website that isn’t accessible to major search engines is akin to locating your business on an island without major access routes!
What’s Next? Web Measurement in the Modern Era (The Amazon Story) As a professional discipline, web measurement has come a long way since the advent of the Internet. It has become such an advanced science that there is little commercial sites like Amazon.com don’t know or are unable to predict about the visitors that come to their sites. In fact, the approaches pioneered by Amazon.com have had such a profound impact on the discipline that any article on this topic would be remiss not to credit them as one of the key innovators. Many of the early features they debuted have already made their way into the recruiting world, and some of the features they are testing now most certainly will as well. Some of the Amazon.com features that have recently transitioned to recruiting (or are likely to in the near future) include:
- Monitoring what type of product visitors routinely seek out for the purpose of recommending other similar products
- Identifying trends among products specific visitors viewed, but never purchased
- Identifying trends among products that specific visitors routinely recommended to others
- Identifying trends among the type of products that get purchased as gifts for other people (not Amazon.com customers)
- Combining web usage and consumer data from multiple sources to mass personalize the landing page to include products and information most statistically relevant given your past behavior
This last feature, while insanely complex, wraps the future up into one tiny activity. For years, corporations have been amassing data that was relevant for short-term reporting purposes at the time, but the data now exists in such volume that data miners can accurately extract behavioral models relevant to each individual. In short, were corporate employment sites or commercial job sites to partner with search engines and financial services providers, they could predict with a fairly high degree of accuracy how likely you would be to accept a specific offer for a specific opportunity. Are we there yet? Not quite. Is it possible? Absolutely. Will it happen? Your guess is as good as mine! Conclusion The art of making a corporate employment site more effective isn’t an art at all, but rather a science. Determining what copy will instill an urgency to act on the part of the visitor is a science we know how to do, a science some companies do very well at on the sales side of the business. Candidate acquisition is a lot easier when everyone you want comes to you, but that is a complicated goal, and one that cannot be achieved without constant measurement and adjustment to operational activities. If having a world-class employment site is a goal for you, as it should be, this article should help focus your efforts. I would wish you good luck, but then again luck has no place in science!