Talent intelligence is transforming workforces, and early adopters are already pulling ahead

The illusion of intelligence: Employers aren’t as data-savvy as they think

I’ve been digging into why employers are struggling with hiring despite gathering data, using the best analytics tools, and leveraging AI. Here’s what I uncovered. 

You’re probably using the wrong metrics

One big reason hiring tech keeps missing the mark is that employers have been measuring the wrong metrics for years.

Here’s what traditionally sat at the center of hiring dashboards. 

• Speed and efficiency metrics, e.g., time to hire or time to fill.

• Volume and funnel metrics, such as the number of applicants, conversion rates across different stages of the hiring process, and offer acceptance rates. 

• Cost metrics – for instance, cost per hire and agency spend.

• Early retention metrics, like 30-, 60-, or 90-day turnover.

• Exit metrics, such as average tenure or qualitative information from exit interviews. 

While these metrics are good for looking back and evaluating your people strategy, they’re all reactive, which means they don’t effectively prevent mis-hiring.  Critics also argue that they fail to capture “quality of hire” (QOH), which will be at the top of employers’ recruiting agendas.

Personally, I don’t agree with the QOH hullabaloo: by the time you can measure it, a mis-hire has already happened, which means time and money down the drain. Matt Charney, a talent acquisition analyst, bluntly notes another problem with it in his blog: 

“Quality of hire’s such a poorly defined metric that even the people who swear by it (mainly management consultants and product marketers) can’t tell you what it means,” he writes. 

Again, when you don’t understand a metric, it’s hard to apply its insights in any meaningful way. 

The only “predictive” metrics I’ve seen used are credentials and background data, such as degrees, years of experience, job titles, previous employers, and employment gaps. Recruiters use these data points to assess whether a candidate is a good fit for the job. 

But there’s now growing evidence that these metrics are poor predictors of future job performance. For example, an article in the Harvard Business Review showed that college education has little correlation with job performance. 

So the problem isn’t a lack of data – it’s a lack of useful data. 

Let’s face it: Recruiters aren’t data scientists 

Today, recruiters are handed multiple dashboards, color-coded charts, and detailed reports from their applicant tracking systems (ATS), HR information systems, and analytics platforms. These show trends like where hires are coming from or how long a role sits open, but they don’t explain the “why” behind them. 

To understand these analytics properly, McCartney et al. (2020) say that HR analysts need a number of skills, including “consulting, technical knowledge, data fluency and data analysis, HR and business acumen, research and discovery, and storytelling and communication.” 

But the truth is, while many recruiters have different strengths, from sourcing to negotiating, it’s unlikely that they’re equipped with all of the above. And this is part of the problem. A 2025 white paper notes that about half of HR professionals report being above average in generating data, but only about a third can measure their impact and translate it into constructive changes. 

And keyword matching tools, ATS filters, and resume screeners can only take you so far

The good news is that hiring teams don’t necessarily have to crunch the data themselves anymore. There’s a host of tech and AI tools that promise to not only do the analysis for you, but also take action on your behalf. 

For instance, ATS filters automatically pick out resumes that match your role description using keyword-matching tools. And AI tools can shortlist candidates not only based on their resumes but also by analysing facial expressions during interviews. 

The problem is that these tools have been shown to distort outcomes. Amazon (in)famously scrapped its AI resume screener after it was found to prioritize male candidates over females. HireVue also faced backlash for its facial expression analysis tool, raising concerns about fairness, accuracy, and discrimination – for instance, against neurodivergent candidates.

Even simpler tools like keyword matchers have shown cracks. If a role description asks for “project management” and a strong candidate uses a different term like “project coordinator,” they might never make it past the filter. 

Additionally, these tools can’t track context or nuance. For instance, an ATS filter might reject candidates with the right skills just because they’re from a different industry. 

That’s why AI alone can’t be trusted to drive decisions or change. You (the human) need to understand how and why it’s making decisions. However, this is where the “black box” problem comes into play: Most AI tools we’ve used so far are opaque and don’t explain their decisions.. 

Here, your data is being translated into action, but there’s no way to judge if these actions are right or wrong. 

Separating signals from noise: What talent intelligence is and how it’s different

Talent intelligence cuts through the overwhelming, often irrelevant, data we have and focuses on the metrics, insights, and advanced tools that truly make a difference. Here’s how it’s different from traditional reporting and analytics. 

Combines the right internal and external data points 

Volen Vulkov, Co-Founder of Enhancv, does a good job of breaking down the data that feeds TI, calling it the “central brain system that connects and powers internal people analytics to skills and hiring market dynamics externally through AI technology.” 

“While people analytics focuses on looking inward (e.g, who among your high performers are at risk of leaving) and sourcing intelligence tracks external hiring pools, ‘talent intelligence’ creates a living, dynamic map that links the two,” he says. 

While TI doesn’t ignore key talent acquisition metrics like retention and performance, it places greater emphasis on predictive insights and largely disregards unreliable performance measures such as credentials, employment histories, and job titles.

Here’s what TI typically looks at. 

It’s forward-looking, not backward-looking 

While traditional analytics look backward – why a hire didn’t work out, or what can be done better next time, talent intelligence looks ahead. 

It forecasts the skills needed based on your business growth plans and encourages proactive sourcing, internal mobility planning, and up- or re-skilling initiatives before roles even open up. 

As Lacey Kaelani, CEO of Metaintro, says, “A lot of companies claim they're doing 'talent intelligence,' but they're just using AI to run reports faster on what already happened. Real talent intelligence predicts what's needed before gaps happen.”

And Sasha Berson, Co-Founder and Chief Growth Executive at Grow Law, echoes this. “At my company, it’s all about using both internal and external data to figure out not just who we should hire, but who we should promote, retrain, and where we might be at risk or even ahead of the game,” he tells TestGorilla. 

So, when it comes to candidate selection, TI focuses on holistic candidate insights to make better hiring decisions so you don’t have to look back and wonder what went wrong. 

It evaluates everything from candidates‘ technical expertise to their soft skills, cognitive abilities, behavioral attributes, and cultural alignment – so you can select those who have the best shot at performing and staying with you long-term.

TI uses tech and AI platforms that spill the TEA – they’re transparent, explainable, and auditable

Talent intelligence doesn’t leave you to interpret static dashboards and complex reports on your own. Nor does it blindly take action, leaving you out of the equation. Instead, it taps into the latest AI hiring tools that surface insights transparently, explicitly explain their choices (such as why a candidate was rejected), and let you have the final say. 

Further, the best tools undergo regular checks and audits to ensure they’re free of bias or flawed decision-making.

Beyond the offer: How TI supports the entire employee lifecycle

Instead of looking at things retrospectively, talent intelligence tells you what to do now – not just to hire better, but also to improve the overall employee experience and retain your best people. Here’s how TI touches different parts of the HR lifecycle. 

Analyzing skills gaps and building precise talent pipelines

One thing I’ve observed over time is that the smartest recruiters don’t sit around waiting for roles to open up and then scramble to fill them. They look ahead and build candidate pipelines in advance so they can hit the ground running as soon as a position is posted. 

They do this in four key steps. 

1. They dive deeper into business objectives. This step is non-negotiable, according to Josh Tolan, CEO at SparkHire, who says: 

“Far too many organizations fail to realize the direct correlation between the hiring process and the ability to meet business objectives. People are at the center of your success. Talent Intelligence (TI) teams need to pivot their focus[...] to identifying metrics that can proactively move the needle on key business challenges.”

Looking at company objectives and key results (OKRs), annual plans, and product roadmaps, and speaking to management are great starting points for aligning recruiting with real business needs. 

2. Then they look outward at emerging market trends, which skills are rising in demand, what competitors are doing, and what the labor market looks like.

3. Next, they use workforce analysis platforms like Eightfold AI or Horsefly Analytics to study their own company’s existing skill set and identify where they’re falling short. 

4. From there, sourcing becomes so much smarter; rather than wasting hours trawling platforms that use unreliable Boolean searches, like LinkedIn Recruiter, TI bets on ready-vetted sourcing platforms like TestGorilla Sourcing, where millions of skills-tested candidates are readily accessible. 

This ensures a targeted approach, allowing recruiters to nurture the right people early on rather than wasting time creating a pool of mismatched, unverified employees. 

Now that’s what I call intelligent sourcing. 

Using advanced AI scoring tools for screening

Unlike AI screeners, which rely on keyword matching, talent intelligence uses these smarter tools:

AI resume and interview scoring 

With these systems, recruiters and hiring managers set clear and fleshed-out skills-based job criteria that reflect what actually matters for the role. From there, AI does the heavy lifting. For example, TestGorilla’s AI tools can evaluate resumes and video interviews, and assign a score from 0 to 5 based on how closely each candidate aligns with the criteria you defined. 

Impressively, these tools show the “why” – with a brief explanation for each score so recruiters can instantly see which skills were demonstrated, where gaps exist, and how decisions were made. 

But, most importantly, if AI misses context or you don’t agree with the decision, you can override scores. This is critical because talent intelligence isn’t about handing everything over to a machine. It leaves space for human intelligence and judgment, which is irreplaceable. 

Skills testing 

According to one study, 70% of candidates admit to lying in their resumes. So, even the best AI tools can’t prove that candidates truly have the skills they claim. That’s why talent intelligence combines these data points with something more concrete: candidates’ skills test results. 

TestGorilla’s platform offers hundreds of expert-created talent assessments that evaluate candidates’ skills, traits, and cultural contributions. This additional data is critical to making the right choices before an offer is made.

Tolan sums it up: “Instead of just measuring QOH post-hire, focus on building a process that can precisely predict a new employee's success, performance, and retention.”

Making offers candidates can’t refuse

Ever since I can remember, comp decisions have stood on shaky ground. Recruiters would rely on a candidate’s previous salary, a loosely defined budget for the role, whatever seemed “market-aligned,” and just shoot their shot. 

Instead of anchoring offers to the past, however, talent intelligence grounds them in real market data. 

For instance, recruiters can look at pay benchmarks on websites like Payscale or Salary.com to understand what similar roles pay. You can also tap into the Bureau of Labor Statistics for geographic wage data, which helps you account for factors like the cost of living and labor supply. 

TI also factors in what candidates want, which is crucial in today’s war for talent. For example, some employers now put candidates through a Motivation test to determine whether a job offer aligns with their expectations. 

Once you have all this data in hand, it’s much easier to make fair offers that meet candidates’ needs, whether that’s through fixed pay, variable bonuses, or even benefits. 

Promoting reskilling and internal mobility

According to the World Economic Forum, nearly four in ten employees’ existing skill sets will change or become obsolete in the next few years. As a result, employers are now prioritizing candidates with transferable soft skills such as creative and analytical thinking and a willingness to learn. 

However, few seem to be thinking about how to reskill the people they already have and deploy them into new positions as things change. Berson sees this as a major issue. 

“Talent waste is the biggest problem I see. A lot of companies I work with have amazing talent already in-house, but they're spending crazy money trying to bring in new people with the same skills[...] it’s way better to invest in internal mobility, upskilling, and thinking about your organization’s future design. You know, we’ve saved a ton of money just by finding internal candidates who were 80% ready, and all they needed was a little targeted training.”

That’s where TI comes back into the equation. It encourages employers to dive deeper into internal data, such as skills inventories, skills assessment results, and performance signals, to match team members to emerging roles that align with their abilities. 

Helping with employee engagement and retention 

The job of talent intelligence continues long after you match person X to role Y. You can now use natural language processing, sentiment analysis, and other AI tools to analyze anything from employee engagement surveys to manager 1:1s and internal communication between team members. 

Berson has seen the benefits firsthand. “Gone are the days when we had to depend on quarterly reviews and managers’ gut feelings. Now, we use AI to track real-time data, like engagement levels, project performance, and even Slack behavior. It flags people who might be ready for more responsibility or who might be thinking of leaving, well before it shows up,” he says. 

These real-time, objective signals can help you catch and resolve problems proactively rather than waiting for them to surface in performance interventions and exit interviews. 

Talent intelligence is a new north star for modern HR

Despite an abundance of data and fancy analytics tools, employers still struggle to make good hiring decisions. Much of this stems from using the wrong signals, looking at data retrospectively, and assuming AI can make all the decisions for you. 

Talent Intelligence changes all this. It’s forward-looking, skills-centered, and focuses on transparency, making it a strategic capability that supports HR far beyond just recruiting. 

For Vulkov, talent intelligence is his “new north star.” He breaks it down simply:

“Segment first, then forecast, then take targeted action based on it, and finally, leverage AI as a strategic collaborator – not a black box.”

Contributors

• Josh Tolan, SparkHire, CEO

• Lacey Kaelani, Metaintro, CEO

• Sasha Berson, Grow Law, Co-founder and Chief Growth Executive

• Volen Vulko, Enhancv, Co-founder