Mr. Fix-it: Fixing the Broken Hiring Process with Prescriptive Analytics and Machine Learning | DCR Workforce Blog

Mr. Fix-it: Fixing the Broken Hiring Process with Prescriptive Analytics and Machine Learning

When I was growing up, every neighborhood had a Mr. Fix-it type of guy (back then it was usually guys; dads more specifically). Ours was my best friend’s dad. If something was broken, he could find the trouble spots and fix them. We brought him our toys and bikes, mostly. He could build things from scratch, too, if need be (like her go cart).

It’s kind of like the hiring process. As the HR and procurement industries have raced to get the best talent at the best prices with the best insight and data, the process got a little broken. There were too many arbitrary decisions, the process was chaotic and the data (if even available) wasn’t being used.

“First things first” fixes

Then came the basic Vendor Management System (VMS), which helped fix many broken issues. It helped identify things at-a-glance such as:

  • How many contractors you have or how much you’re spending on contract labor.
  • If you have data on how quickly your staffing vendors respond to open positions.
  • How long your contractors are staying and if they’re leaving their assignments early or staying indefinitely.
  • Whether all your contractors compliant.
  • If they’ve taken and passed your required drug and background testing done.
  • How much of your total non-employee spend is on overtime hours and if those hours are really needed.

As time progressed and more contingent workers were hired, sometimes the system was circumvented “in the interest of time.” But as most managers agree, when objective data is used, without unconscious bias, better hiring decisions are made.

Companies that rely on innovative breakthroughs in the VMS sphere are leading the charge with best-in-class contingent workers. How are they getting them? By using a VMS that uses prescriptive analytics and machine learning. Companies are slowing down and instead of chasing the fast hire, are looking more in-depth at finding the right hire.

With that comes the need to dig deep into the skills and potential the candidate brings. And how better to do so than to dig deep into what the organization really needs and build that into the front-end of your VMS. Does it take time? Sure. Is it necessary to create the best requisition? Yes. Can your VMS learn how to find the best candidate through the mounds of data? If it’s Smart Track it can!

Future-forward fixes with machine learning

As companies demand better workers by relying on objective data instead of human bias, DCR created within Smart Track a cutting-edge machine learning datasets and our algorithm that powered our innovative Match Index Intelligence with prescriptive analytics, so hiring managers could take one glance and see who the best candidates for the requisition are.

DCR’s SIMSIM (Smart, Intelligent Machine, twice over) Engine uses cognitive machine learning to power many innovations in Smart Track. We’ve always provided past and current insights. Now our workforce insights help users predict future events with recommendations that a decision-maker can use to capitalize on business opportunities or minimize risk through our prediction and recommendation model.

The SIMSIM Engine provides recommendations on a wide range of business topics, including candidate source, candidate referrals, freelance and candidate usage, candidate rankings, SOW and contingent worker usage, candidate tenure analysis, supplier scorecard and strategic resource analysis.

It sounds impressive because it is. We’re teaching Smart Track to think like a human with tons of experience in workforce management, but without the added caveats that impact actual human thinking. We use machine learning algorithms and data science to provide a tool that our clients can use to make candidate decisions that are both objective and based on their own client-specific experience.

Not a quick fix

This isn’t something you can build overnight! Anyone who says otherwise does not have the power of resources that Smart Track has. DCR has been pulling data from numerous sources including top career sites and the company’s own hiring-related historical data for numerous years. We make sure that each client’s data is private, so hiring managers are only seeing matches that are made based on external market data and their own historical data and behavior. And our multi-tenant system ensures that there’s a brick wall between different clients’ data sets – each one is independent, and no two clients’ data gets mixed with one another’s.

Building a feature such as our proprietary Match Index Intelligence requires an immense amount of data to build – we’ve been feeding our engine market data for more than a year on a rolling 12-month period. To date, DCR is the only VMS provider to provide such a value-added feature to our clients.

And this learning is continuous – so, with each run, Smart Track learns more and provides even better matches. For example, trying to search for or match a candidate to a requisition using standard word search is tedious and often yields poor results. Hiring decisions require better cognitive analysis than that. Smart Track uses Natural Language Processing (NLP) algorithms to analyze job requirements, candidate qualifications and historical hiring manager selections to learn what type of candidates are really the best match for a position.

Because we use data mining and machine learning, Smart Track learns from the content itself. This is already built in, so no one needs to fit in a list of synonyms, abbreviations, acronyms or misspellings. Rather Smart Track reads the resumes and formulates a “term space” that identifies the relationship between every term in every resume in the collection of resumes. This term space is like the human brain. Every term in our brain has some context. So like the human brain, the engine maps how terms are used interchangeably, and concludes that the terms are related, or synonymous, with one another. Once it creates the term space, it organizes the terms into concepts of related terms.

With resumes and job descriptions, skills and requirements are represented as concepts. While using a word search or Boolean keyword is limited to just one way to express that concept, Smart Track’s machine-based learning allows for the concept to be expressed and understood without limiting the results to just one expression of the concept.

Mr. Fix-it triumphs again

We often hear from our clients that they spend a lot of time and effort reading through hundreds of resumes to determine who the best candidate is for their requisition. Smart Track’s Match Index Intelligence solves this pain point by identifying which candidates are the best fit for the position, saving them hundreds of hours of time, so they can concentrate on other initiatives.

The Match Index Intelligence pre-pairs candidates to requisitions by matching supplier-submitted actual resume content with the requisition details, using NLP, data mining and machine learning to mirror human decision-making without the inherent bias. It then gives a ranking to each candidate and buckets them into groups for hiring managers to easily understand who the best match is for the job. The Candidate Match provides the match percentage and the rate, so managers have everything they need to make a data-driven decision.

The prescriptive analytics and machine learning in our proprietary Match Index Intelligence is only one of the exciting features available in Smart Track and ONLY in Smart Track.

If your hiring process needs fixing, email us at, so we can discuss your program bottlenecks and help you find a solution.

The content on this blog is for informational purposes only and cannot be construed as specific legal advice or as a substitute for competent legal advice. They reflect the opinions of DCR Workforce and may not reflect the opinions of any individual attorney. Do contact an attorney for advice specific to your issue or problem.
Ozzie has more than 30 years of experience in human capital management as an expert in managed service programs and vendor management systems for the non-employee workforce. With a focus on strategic solutions by analyzing client pain points and issues, he has achieved results-oriented first-in-class programs.