Predictive Analytics: The HR and Recruiting Data Science Maturity Model

In 2016 the buzz around artificial intelligence (AI), data, and people analytics was without a doubt the biggest trend in conferences, webinars, articles and debates, and other HR industry news.

The chatter was heightened by the increase in private investment and enterprise software companies’ focus on introducing a multitude of new ways for data to transform HR. From hiring to on-boarding to compensation to culture, the ways HR and the people business as we know it is becoming smarter.

One of the biggest claims we heard in 2016 was how HR will become automated. However, it didn’t happen on any grand scale, which has led many to question the viability of AI in HR.

Data around HR processes exists, but without a way to monitor and balance scorecards, the data is not actionable. The science of AI and real-time data are still in the early stages for most organizations. And it may be years before we will automatically improve business operations, balance scorecards, and mature to utilize people data appropriately.

In 2015, data-driven modeling became mandatory, but the solutions were limited to “show me the money” in a bunch of pretty graphs and colors. The next steps for HR to become data driven will be handled in phases, and for large organizations, each phase may take years.

The HR and Recruiting Data Science Maturity Model

Most organizations will follow a maturity model to record, analyze, and implement automation in the next few years. In 2017, the industry will begin to see the results of leading organizations moving into Phase 4 to a greater extent.

Phase 1: Getting Started

Identify opportunities and new data sources. Senior leadership help define and develop the team and business case.

What Metrics Matter?

1. What are our KPIs?
2. Which systems / locations / departments do they come from?

How do we monitor these metrics?

1. Do we need better systems?
2. Do we need a partner? Can our partner handle this?

Phase 2: Sophisticated Analysis

Develop business analysis to measure business operations. Begin analyze data required to test hypothesis and begin advanced data modeling.

1. Let’s look at our KPIs.
2. Collect data for actionable decision making.
3. Create hypothesis: What changes will have the greatest business impact?
4. Deliver immediate value and guide data strategy.

Phase 3: Business Operations

Earliest steps to increase impact on business operations, create or enable “data products” for longer term value or to replace legacy systems.

1. Decisions are made based on data.
2. We have a foundation for organizational intelligence.
3. Begin experiments to validate hypothesis.

Phase 4: Automation and Data Science

Full capability of data science and earliest use of AI supported throughout the organization and strategically embedded in business processes and senior decision-making.

1. Build, implement, and deliver solutions based on valid hypothesis to improve business operations.
2. Continue monitoring, analyze macro trends from collected KPI data.
3. Begin new experiments to validate new hypothesis.

Where Will We Go From Here?

The implementation and analysis of the actual data science to improve the HR function is still early on for most companies. Even leading enterprise vendors are still early in product development to make the most of HR data beyond reporting. However, in 2017, the industry will begin to see signs of automation improving HR, most publicly in improvements in the way companies recruit and hire new people.

2017 Prediction: Data-driven HR functions have landed; AI will begin to drastically improve recruiting.

AI Chat Bots Will Enter Recruiting to Create Transparency

With the incredible rate of consumer adoption of platforms such as Slack, Facebook Messenger, WhatsApp, Snapchat — outpacing other forms of media — messaging bots become an opportunity leading organizations and early adopters this year will rapidly expand the way people interact with their company through AI messaging bots.

Bots enable people to learn more about opportunities, conduct traditional HR process, and make career decisions without job applications and pre-hire recruiting qualification calls. Companies can host automated conversations in leading messaging platforms like Facebook Messenger and Slack to reduce application drop-off and increase pre-qualification through short, iterative conversations that do not require company resources.

AI Will Reduce Cost per Hire

For the most part, recruiters cannot take advantage of the millions of candidates in their ATS. Outdated resumes, unqualified applicants, and millions of records impede your team’s ability to produce new hires from existing data.

AI and data analysis will reduce cost-per-candidate and time to fill with real-time data enrichment to update candidate resumes, and improved matching technology will answer questions often left to a recruiter, such as:

• Who in our database is best for this role?
• What is the realistic offer that will be accepted?
• Who is the most likely person to accept an offer?

The largest recruiting process outsourcers are all taking steps to better understand who resembles the best person for your team based on data gathered in Phase 2 of the maturity model.

Attention to existing applicant databases will likely increase due to shrinking recruiting budgets and the decline of the Job Board era. In 2016, we saw the final decline of the job board era. The acquisitions of Monster, Simply Hired, and LinkedIn were all data acquisitions and suggest a new paradigm for the already fragmented HR tech market.

Recruiting Decisions Will Be Measured More Like Marketing

It’s easier to source candidates from the web, but it’s getting harder to get them to respond to you. Most recruiters point to the candidate driven market and that passive outreach is less and less effective.

Marketing and recruiting will continue to team up to bolster employer branding efforts and usher in new methods to increasing applicants. With recruiting beginning to look more like marketing, so too will the performance metrics blend. If marketing is responsible and accountable for customer acquisition costs, so too will recruiting.

Marketing and sales departments have been retooled ahead of HR to focus on making customer data actionable. HR leadership should take a serious look at how they can model employee acquisition and retention after marketing’s performance metrics.