In 2012, Target famously got in trouble when it was revealed that, through big data, it knew a teenager was pregnant before the girl’s father did.
What happened was that through the girl’s purchases and Target’s data algorithms, the company correctly determined she was pregnant and began sending coupons for cribs and baby clothes to her house. Her father saw these coupons and went to Target to complain, only to discover that his daughter was indeed pregnant and she was hiding it from him.
It makes for a kind of funny, if not a bit scary, tale of both the power and accuracy of big data, which companies have primarily used in marketing. But what if those same processes were used in recruiting?
In other words, what if your online activity was used to determine what job you should have? What if, rather than having to rely on a resume and an interview to determine if a candidate is the right fit, companies used your browsing history to figure out if you were really right for the position?
Sounds crazy. But it seems like that’s exactly what Amazon is doing. And, frankly, if they don’t, someone else will.
What Amazon Is Doing
The tangible evidence that Amazon is trying to accomplish the aforementioned goal is limited to a job advertisement posted 146 days ago for a position in Edinburgh, Scotland (although rumors have run amok). The position is called “Software Development Engineer/Machine Learning Engineer” for the company’s “Intelligent Recruiting Systems Team”.
Here’s the first paragraph of the job advertisement:
“Join our Intelligent Recruiting Systems team to invent the future of recruiting at Amazon. We’re inventing new systems to automatically source candidates and match them to jobs using machine learning and Natural Language Processing at massive scale. Our objective is to deliver an outstanding candidate experience while helping Amazon to grow by enabling great hiring decisions.”
The advertisement goes on to say it is looking for engineers who can “optimize systems that extract semantic data from unstructured text, store and process data at massive scale and apply machine learning to make great decisions.”
The job is not a low-paying one, as the position calls for someone who has at least a master’s degree, and preferable a PhD in computer science or engineering. Clearly, for a company like Amazon to make that sort of investment (probably around a $200,000 package), it is making its “Intelligence Recruiting Systems Team” a big priority.
Aside from that one job description, there’s no other evidence online about Amazon’s “Intelligence Recruiting Systems Team”. A Google search comes up blank and even a search of that term in Amazon’s job portal proved fruitless.
There’s the possibility that Amazon didn’t find the right person it wanted or scrapped the project for some other reason. Or the opposite, Amazon found exactly the type of person it was looking for, and in Scotland right now some crazy project to completely change the way the world hires is underway.
Either way, there’s a very strong possibility that this approach will be common in the not-so-distant future. Most companies are already using data to make hiring and other people-related decisions (Google being the best example), but that is data that is being collected once a person applies for a job.
But soon, it will extend well past that. With the data companies collect already, they can accurately determine your level of income, interests, sexual preference, medical history and just about anything else. Why wouldn’t they create a profile of the exact person they want, and then actively reach out to someone who matches it?
Take a practical example. I, Paul Petrone, use Amazon to primarily buy books and graphic novels, including books on how to write better. Using other information, I’m sure Amazon could determine my salary expectation, location, age, and, perhaps eventually, how strong I am at my job.
Then, say, Amazon is looking for someone to create content for its graphic novel entity, comiXology, which they determine I’d be a good fit for. Why wouldn’t they reach out to me? And since they know so much about me, they’d probably know the exact salary to offer, what benefits I’d seek, etc.
In other words, they could probably figure out exactly what to pay me, and not a dollar more. That information alone is worth a lot of money to companies.
Amazon might or might not be able to accomplish the goal I just described – although, frankly, I wouldn’t bet against them. But either way, someone will, and it likely will become a popular idea within companies.
It will be interesting to see if there is backlash from the public and governments. Perhaps some will love it, seeing it as an easy way to move up the career ladder. And perhaps others will see it as a violation of privacy. (Quick side note: my not-so-bold prediction is that the people who get jobs using this new system will be the ones who love it and the ones who don’t will hate it.)
There are privacy laws in place, but like most laws, they do not keep with new technologies. Google, for example, doesn’t look at candidates’ search histories when they hire. But that doesn’t mean there aren’t ways around the law or new ways to use the data that aren’t covered in the law.
The point is, the way we hire in a few years will be drastically different than the way we hire today, as companies will be armed with more information about candidates than ever. The HR teams and recruiters who accept this and figure out how to make it work for them will be highly successful; the ones who don’t, won’t.
VoiceGlance is a cloud-based hiring tool used by forward-thinking companies to hire smarter, instead of harder. Learn more here.