Forget that Great Barrier Reef job, the sexiest jobs in the 21st century will be in data science. This was predicted by Thomas Davenport and D.J. Patil in the Harvard Business Review. They should know. They’re data scientists. They crunched the numbers and they came out sexy. There’s been a lot of Big Data talk over the last couple of years. It’s been heralded as a management revolution. The technology research house, Gartner, expects the market for Big Data and analytics to generate $3.7 Trillion in products and services, and 4.4 million new jobs by 2015. That’s a lot of sexy jobs, and filling those jobs is clearly going to need some brain power to be applied to talent analytics.
If Big Data is sparking a revolution in general management, then Predictive Analytics may well be the “next big thing” in talent management. Google has certainly reached this conclusion. It used to be the case that hiring at Google was a relatively intuitive process reliant on famously quirky brain teasers and many, many different points of view. It was not unusual in the past for a potential recruit to undergo interviews with over a dozen different people before a hiring decision was made. And of course, for a long period of time it was also known that nobody was hired without the nod from the Google founders, Sergei and Larry. This felt intuitively right, but then the Google talent team hired Laszlo Bock to set up a new function called ‘People Operations’. His mission was to “apply the same discipline and rigor to people operations that we use to manage Google’s business operations.” The desire was to ensure that all people decisions were informed by data and analytics. When they looked at the hiring process there were some unsettling surprises.
As Laszlo commented in a recent New York Times interview: “We looked at tens of thousands of interviews, and everyone who had done the interviews and what they scored the candidate, and how that person ultimately performed in their job. We found zero relationship. It was a complete random mess.” The Google team also recognized that brainteasers didn’t predict anything. The only purpose they appeared to serve was to make interviewers feel smart. Google’s job candidate interview methods are now far more data-driven, with a heavy emphasis on structured behavioural interviews (where you ask people to speak to their own experience, rather than respond to more hypothetical questions). The company also conducts far fewer interviews per candidate, after discovering that four interviews were enough to gauge whether an interviewee is going to be a good fit for a position. These changes have been made out of necessity not simply analytical virtue. Google now employs over 50,000 people around the world, and receives somewhere between 2 and 3 million applications a year, from which it makes thousands of new hires. Nevertheless Google’s data driven approach has begun to shake up the way HR thinks about talent metrics.
The use of measurement to determine the effectiveness of different employer brand marketing activities has been relatively piecemeal in most organizations. Only 25% of the participants in our employer brand practice benchmark survey claimed to be measuring the external impact of their employer brand activities and very few claimed to be connecting external and internal measures to evaluate the all-round heath of their employer brand. This reflects the generally poor state of talent analytics within most organizations. In Deloitte’s 2014 Global Human Capital Trends survey HR / Talent Analytics were reported to be one of the most important trends in people management. It was also the trend which people felt organizations were least ready for.
HR has always collected a lot of data, but has tended to be poor at analysing and applying it. One of the key reasons for this is the complexity of HR data. Bersin Deloitte’s research on HR systems found that the average large company has more than 10 different HR applications which generally makes data linkage a highly time consuming and expensive task. The second likely reason is expertise. Data analytics demands a very different skill set from conventional HR, and finding the right talent to perform the job is difficult. A recent survey involving 300 IT professionals revealed that 55% of data analytics projects are abandoned, and one of the most common reasons for failure was managers lacking the right expertise to “connect the dots” and form appropriate insights. The third and most important reason from my perspective is the silo driven nature of many HR departments. As Martin Moehrle, the Global Head of Talent at UBS puts it:
“HR tends to be a collection of independent processes rather than an integrated suite of people management services. It’s easier to run things separately rather than manage a complex web of inter-dependencies, but this fragmentation is sub-optimal in terms of supporting business performance. One of the most significant divisions lies between those who bring people in and those who cultivate people once they’ve joined the organization. This is a missing link which the integrated talent approach might begin to close. Bringing these different elements together is vitally important if you want to build a strong employer brand”.
So what needs to be done to get your employer brand metrics into better shape? Here are ten steps you should be considering.
1. Establish robust track to hire metrics, so you can establish the relative efficiency of your media and content investments (in other words, your cost per hire by source of attraction). The latest campaign tracking methodologies not only enable you to establish how many applicants and hires are being generated from different media investments, but also track the further visits potential applicants make to your social sites and other relevant points of reference before they commit to an application. Your Facebook careers page may not be dominant source of hire, but it may prove to have a significant influence over application and acceptance, so it’s important to incorporate this extended tracking into your recruitment metrics dashboard.
2. Clarify your target profile and EVP. This should determine the key capabilities and culture-fit characteristics you’re looking for in potential candidates (quality of hire), and the primary employer brand associations you want to establish (brand image).
3. Benchmark the strength and depth of your current talent pool. Both internally (general bench strength) and externally. How many accurate contacts do you have access to via your ATS database and Linkedin? How many people are actively engaged with your social and professional network pages?
4. Evaluate your current external brand image among key target populations and the potential gaps between current and desired image. You can use student surveys like Universum and Trendence if Universities represent an important source of talent, and brand image surveys among key target populations or wider applicant surveys for experienced hires. Note that the latter method will not give you awareness or engagement, but it can give you an indication of primary image associations.
5. In addition to applications / cost per hire, link your marketing media / content investment with longer term trends in the strength and depth of your external talent pool and employer brand image. Linkedin research suggests that a stronger employer brand image will reduce your overall cost per hire, over and above the efficiency improvements you make in selecting the right media.
6. Establish a new Joiner Survey timed to allow assessment of both the candidate and orientation experience. Three months is a better time for this kind of survey than 3 weeks, as this will also enable you to pick up any significant gaps between employer brand promises, employee expectations and the emerging reality of the employment experience.
7. Incorporate an ‘employer brand index’ into your employee engagement survey to track perceptions of your organization’s delivery against your EVP promises, and how strongly these perceptions correlate with engagement and retention.
8. Develop a quality of hire measure. Many companies determine this through a combination of hiring manager satisfaction, first year attrition, and first year performance evaluations.
9. Link your quality of hire measure back to source of hire to determine the most effective quality hire media. You should also track the relationship between your quality hire metrics with new joiner survey data to track the effect of improvements in candidate / new joiner experience on overall levels of quality hire retention and performance.
10. Finally, you should try and identify the linkages between your employer brand marketing investments, employer brand reputation and experience measures, and business performance measures like productivity, customer satisfaction and sales.
If you can establish these full-cycle metrics and linkages you will be in much stronger position to shift the overall focus of your employer brand marketing activities from short term cost considerations to the longer value you’re adding to business performance. As a result you’ll also be more able to secure the necessary investments required to make a lasting difference. Sexy indeed.
This blog contains extracts from Richard Mosley’s latest book ‘Employer Brand Management: Lessons from the World’s Leading Employers’ (Wiley) which is available to order on Amazon.