Big Data, Content and Analytics in HR

To begin at the beginning

As has been stated elsewhere, human resource management is a content and process intensive activity, which makes it somewhat amenable to the deployment of content and process centric IT solutions. In particular, Enterprise Content Management tools that also offer advanced process design and deployment, would seem to be an ideal fit for any significant and continuous human resource activity.

Like many other activities in business, the roles and responsibilities embodied in human resource management have emerged, developed and transformed over the years, and with subjective improvements and innovations the field has become more complex, more varied and more concentrated – in a wide range of aspects, but especially in terms of the explosive proliferation of process, business rules and content.

The reasons why the HR activity has become more intense is manifold.

Of course times, if management consultants are to be believed, are always critical for every business, every market, every department, every role, and for every responsibility, function and process. But sometimes, by providence or serendipity, the hype mysteriously coincides with the reality.

In recent times human resources has been highly affected by the explosion in the digital economy, with web sites such as LinkedIn and Xing appearing on the scene and creating certain degrees of disruption.

In addition, just as it is, and has been possible for businesses to reinvent their past, present and future, without being overly rigorous with regards to the facts, the realities and honesty, so too are people in the jobs market finessing their own curriculums in much the same way that companies embellish their products, inflate their service offerings and exaggerate their achievements. This is one of the problems in human resource management, CVs that do not represent realities or bend realities out of all recognition.

Another problem with human resource management is also to be found in job descriptions and job offers that are badly aligned with prevailing realities. Just as companies find that a hired candidate does not actually align very well to the job or to their own CV in quite such an ideal way, so too are some job descriptions woefully inadequate, insincere and misleading.

So, we have a situation where companies, hiring managers and candidates gild the lily.

Now we probably don’t want to waste time and effort on getting companies to be totally honest about themselves, but as businesses we can take steps to ensure that we interpret job requirements and candidate curriculums correctly by using the power of content management and comfortably contented analytics.

So, what content can we analyse, how do we analyse it and what do we do with the results?

Analysing the Curriculum Base

As Teresa Rees put it “[Shirley] Dex maintains that it is possible to construct and test theories in a range of disciplines using life histories, and argues that they represent a blurring of the traditional boundary between qualitative and quantitative methods.” This, in my view, is particularly relevant and potentially effective when it comes to the analysis of curricula, especially where a vast collection of such content is available, for example in businesses that specialise in providing human resource and project team building, management and dismantling.

Using tools like IBM’s Watson Content Analytics, it is possible to mine a whole range of hidden correlations, from symbolic, through quantitative to quasi-qualitative data. Not only that, but by applying arbitrary predicative analytics, it will be possible to create detailed, dramatic and realistic scenarios based on a whole heap of factors and unsubstantiated correspondences.

This heralds an exciting time for the automated generated of a wide rainbow of Cartesian products, on many plains, in many dimensions and in many interpretations.

By using the power of today’s cheap-commodity computer technology and the vast offer of ‘free’ open software, it will be possible at some time in the future to successfully replicate the skill and art of human resource management practiced in bygone-days but at a more attractive and increased multiplier of the previous cost. So, this will also please suppliers and those who take an undeclared incentivised cut from services and artefacts that are billed for.

Analysing the Job Description Base

As we saw in the example of the curriculum, so too will we be able to mine the vast collections of job descriptions, offers and mentions in order to create an all-encompassing view of market drivers, demands and movers.

In the future, mega-hr-corporations will be using Big Data Analytics, Enterprise Content Management and Content Analytics to suck up the nickels and dimes of the job market, chipping away financial benefit from wherever it may be accrued – from clients and workers alike, especially workers - like ginormous bottom-feeding catfish in a universal sized version of the Everglades or the sewers of a major inter-galactic city somewhere in our dreamy and dystopian inheritance.

Analysing social media and professional social media

Do you have your ducks lined up in a row and hanging from your living-room wall?

If a company is serious about HR, ECM and content analytics, it also needs to think about other things. We need to think of more metrics. Or as Social Media Examiner put it (March 28, 2013):

“Reach. You might want to measure the number of fans, followers, blog subscribers and other statistics to gauge the size of your community.

Engagement is measuring retweets, comments, average time on site, bounce rate, clicks, video views, white paper downloads and anything else that requires the user to engage.

Competitive data may include the brand’s “share of voice” across the web or number of competitors’ brand mentions.

Sentiment. You might want to measure the numbers of mentions with positive or negative sentiment.

Sales conversions. Do you want to measure social media referral traffic to the top of the sales funnel or number of sales aided by social media efforts?”

I couldn’t have stated it better myself, that is, if I had wanted to state it, which I probably didn’t. But it seemed like a good idea at the time.

Correlating results

Now it’s time to pull together all that wonderful analytical magic and coalesce it into a tangible and meaningful whole. That’s the whole idea, the answer is in the whole, their whole, our whole, and (as they say in Ireland) “your whole!”

Wikipedia states that "Correlation does not imply causation is a phrase used in statistics to emphasize that a correlation between two variables does not necessarily imply that one causes the other. Many statistical tests calculate correlation between variables. A few go further, using correlation as a basis for testing a hypothesis of a true causal relationship; examples are the Granger causality test and convergent cross mapping", now we know for an indisputable and unarguable fact that this is arrant nonsense invented by professional statisticians to defend their very shaky and untenable turf.

Believe me, as a fully paid up member of the Royal Order of Gentleman Data Scientists I know full well that correlation is king and that causation is for wimps. It’s been scientifically proven, not only by errant Readers Digest bloggers, but also by the most venerable members of the IT community itself.

This I why and how I can claim, with no element of doubt, shame or certainty, that when we derive nuggets of gold from the mining of CVs, job descriptions and gossip on social media - even the Facebook of the ‘connected and voguish’ professions, that we are deriving a value that surpasses that of the holy grail of all things analytic.

If you don’t take my word for it, just try it in your business, and you will soon become a believer, too.

Just remember, do it right, or all bets are off.

Driving insight and taking decisions

That’s why we can use what we learn in terms of insight to better perform:

• Hiring - planning and decisions
• Profiling – intelligence services, spying and subterfuge
• Insider trading – rationality, pragmatism and coincidence
• Requirements – planning, decisions and execution
• Training - planning and decisions
• Workforce realignment – planning, decisions and actions

And a million other things that take our fancy.

Analysing the analysis, after the event

Moreover, we can also analyse the analysis, to see what worked, what didn’t and in extreme cases, what Big Data driven decisions lead to massive success or fraud, bankruptcy, mass redundancies and jail time.

This part of particularly reliant on what I term the 4th generation Enterprise Data Warehouse.

Conclusions

There is a whole new world opening up in terms of data, Big Data, Big Data analytics, content analytics, human resources, abuse, social media, data privacy violations, Chinese walls and Enterprise Content Management.

Are you going to be part of the success story or will you be left to play catch-up at some time in the future, just when it might be too late to join in the fun or reap the massive benefits.

Many thanks for reading.