You are responsible for hiring in your company. It’s a tough job, but you’re experienced and do it well. Lately, however, you’ve heard whisper of new trend in HR – big data – and wonder how it can help you. For the uninitiated it’s a complication that confounds at every turn.
But don’t worry, there’s no need to hit the panic button. Maachu is
the mentor, the wise Svengali to light the way. We are experts in the
use of big data for recruitment and have gathered our thoughts to give you some key pointers on how it can be leveraged during the recruitment process.
You may well be wondering what big data has got to do with recruitment. After all, in the not too distant past a pen, paper and filing cabinet were key recruitment tools. The world today, though, is a more complex and sophisticated place and big data provides companies with vast amounts of information that can help them make smarter and quicker business decisions.
For example, if you’re in the cut-throat retail world you need to be able to analyse customer purchases, supplier pricing and consumer behaviour, all in real time. Big data tools give that information to you in an instant – Excel just doesn’t cut it!
OK, so recruiting isn’t quite there yet, but change is coming and it’s coming fast. No longer is it good enough to just search the CVs on your ATS. There may well be unemployment problems in many countries, but there’s also a global talent shortage and finding the perfect hire for your business requires skilful analysis of numerous data points and metrics.
Just looked at LinkedIn. The social media network for professionals has successfully acquired career data of over 200 million job seekers worldwide. It’s a great big data resource. And even traditional career sites now give you some user data. We’ve taken it a step further and asked candidates about their visa requirements, industry skills and current and desired salary amongst other things.
Well great – but how can big data make your job easier?
Lately, I’ve been working a lot with the data we’ve collected from over 15,000 candidates and I’ve come up with a few ideas:
1) You can see what skills are available on the market
When working for one client, we were asked to find programmers in Singapore. Fairly simple task, but the customer also wanted their hire to possess a detailed combination of skills and experience. Without going into too much detail they were looking for one of those elusive ‘purple squirrels’ and could quite easily have ended up with no candidates.
To help them understand this, we ran a survey with our programmers and asked them: “What computing environments are you familiar with?” and included all of our customer’s requirements and more.
Here is a drill-down of the search results:
- Programmers in Singapore with required background: 170
- Who specialize in Java: 42
- And are familiar with Hibernate: 20
- And know GIT or SVN: 4
- And use JBoss: 1
So, out of a pool of 170 programmers, only one had the original combination of skills.
Armed with that information, the employer was then able to decide whether to wait for the right person, or train one of the candidates from the larger pool.
Of course this could be gleaned from CVs, but how much time would it have taken to arrive at a conclusion that was instantly obvious by looking at available data?
2) You can see what you need to pay to get someone to move
Another challenge faced by employers is that once they have found the right candidate, they need to make a financial offer that encourages the person to move, but isn’t too much for the role.
Without data, this is impossible . Salary surveys are not precise and, according to Harvard Business School, you will benefit from coming up with the first figure (called the ‘anchor’ of the negotiation).
So how can you make the best offer for your company?
Well, with our candidates we’ve made it simple. We’ve asked them what salary they want to move and you can decide whether that suits you. But say, for whatever reason, that figure is unavailable for the candidate you’re interested in, what then?
Well data can still ride to your rescue. From a vast candidate pool, we have real salary data which can help you come up with the right pay for roles – including years of experience, and location.
For example, if we go back to our programmers, I can quickly see that in Singapore they are, on average, looking for 14.9% increase on their current salary. But, we can break it down further. A candidate with under 5 years’ experience wants a 19.5% increase, and those with over 5 years’ experience would move for just 12.5%. This information can really help you when you’re at the negotiating table.
And the figures are somewhat different for, say, bank trade processors who look for an 18.4% increase on average (broken down, 0-5 years wanted a 21.8% boost; 5+ years just 13.1%).
Now, you may decide to offer more or less than the market depending on the urgency and your budget. But the point of big data is to make your decision with ALL of the available data – and what I’d like everyone reading this to know is that this data is now available!
3) You can decide whether to pay a premium for a skill
Now imagine you’re looking at a few financial researcher candidates, but one shines above the rest. He or she, however, has a master’s degree and wants significantly more than the other candidates.
Should you pay more for a candidate with an advanced degree?
Well, the data is your friend once again, though it gets a bit more
complicated. First, we look at the salary data and compare those with
the qualification and those without.
It looks like financial researchers with a master’s degree are paid more, but how can we know for sure? Maybe the higher salaries just reflect experience.
Well, we can use a technique well-known in the data world – regression analysis. Now, if you’re not mathematically inclined it might seem like magic, but it’s an established method for determining whether an input matters or not. In our case, we need to know whether having a master’s degree affects salary.
I’ll spare you the gritty details here (though happy to share them – just email me) but it turns out that regression analysis shows that – surprise – having a master’s degree does not affect the salary for financial researchers, whereas years of experience affects it a lot (which we would expect). This is not to say that a person with a master’s degree does not expect more, but according to our data, the market does not pay a premium for that qualification.
If we change the job role, though, the story is different.
This time round we are looking at financial risk managers and a
glance at the graph below looks like risk professionals with a master’s
are paid more – but what does the regression analysis say?
Well, here the analysis not only tells us the market pays more for the master’s but it pays $35,335 more on average! That number is certainly skewed by the relatively highly paid people in the 5-10 years’ experience range, but still this would be very good to know if you were negotiating a salary for a risk professional.
Harnessing the power of big data
Fundamentally, big data in recruitment is about providing highly targeted and detailed information that can help you to:
- Ensure your requirements are realistic
- Make an irresistible offer (and one that won’t blow your budget)
- Know whether you should pay more for a certain skill or qualification
There is a lot of data out there. The volume is expanding and take up by recruitment professionals is on the increase. However, very few people actually know how to interpret the facts, figures, graphs and statistics, rendering all this valuable data useless.
At Maachu we can provide you with the data and the intelligent tools that allow you to painlessly harness the true power of the now available recruitment data, and help you to make sound sourcing decisions.