The Metric CEOs Get Wrong — and How DataEQ Fixed It

Key Takeaways:

• Many CEOs rely on flawed sentiment metrics derived from automated systems that miss tone, context, and nuance in customer data.
• DataEQ’s hybrid model combines artificial and human intelligence to ensure sentiment insights are accurate.
• Accurate sentiment analysis can help boost results across CX, compliance, risk, and reputation.

More than ever, business decisions need to be made quickly, and that can only happen with real-time feedback and hasty data analysis.

The issue is that a decision is only as good as the data behind it, and there’s one specific metric many CEOs are unfortunately getting wrong: customer sentiment.

Nic Ray, CEO of DataEQ, argues that the core problem isn’t access to data. While available data is a common issue, it comes down to how it’s interpreted.

Most sentiment analysis tools rely on AI to scan massive amounts of unstructured information, but DataEQ believes AI alone lacks the nuance to separate constructive criticism from sarcasm.

In an interview with DesignRush, Ray explains the solution to this issue, and how truly accurate sentiment insights have delivered real-world results across several industries.

Who is Nic Ray?

Nic Ray is the CEO of DataEQ, where he leads the company’s hybrid AI-human intelligence approach to structuring customer feedback. Previously, he was part of the founding team at Quirk, a digital agency acquired by WPP. He now oversees DataEQ’s international growth across Africa, the Middle East, and Europe.

Companies have come to rely on sentiment scores and feedback to establish brand awareness, customer satisfaction, and overall public perception.

There’s plenty of solutions aimed at providing that data, but most are powered by AI alone, and according to Nic Ray, that can lead to misinterpretations.

In other words, CEOs and decision makers will be making important calls based on inaccurate data.

“AI alone struggled with context, sentiment, and nuance, elements critical for decision-making,” Nic tells DesignRush.

“This led us to develop our hybrid intelligence model, the EQ Engine, which combines AI with human oversight to structure unstructured data at scale.”

What began as a tool for marketing is now used across functions, from customer service and compliance to risk monitoring and operations.

Trump and Brexit Help Make the Point

One of the earliest DataEQ validations applied to politics — not commerce.

In 2016, the company accurately predicted two of the most unexpected events in recent history: Brexit and Donald Trump’s election victory.

While the consensus was that Brexit and Trump would both fail, the analysis by DataEQ of social sentiment painted a different picture — one that proved correct.

“These predictions weren’t about politics for us; they were a proof of concept,” Nic says. “We designed them as a marketing exercise to demonstrate the power of accurately structuring millions of unfiltered public opinions from social data at scale.”

The company has since expanded its impact.

In the financial sector, DataEQ has worked with regulators and top banks to map out social sentiment against compliance frameworks.

DataEQ has helped institutions align more closely with customer expectations while meeting oversight requirements.

Why Most CEOs Misread Sentiment

According to Ray, a fundamental issue is the tendency to chase volume over value.

Businesses get flooded with online mentions and often interpret spikes as signals, when they are most likely noise.

Even more critically, many companies reduce sentiment to a single metric, ignoring the difference between service-related complaints and overall commentary about the brand.

“A single sentiment score doesn’t provide the full picture. Operational feedback is vastly different from reputational conversations, and these should be analysed and reported on separately.”

The result? Businesses act on fragmented data or miss key issues altogether.

Worse, flawed sentiment insights can lead to misaligned strategies, wasted resources, and missed opportunities.

The Hybrid Model

The DataEQ solution is rooted in what Ray calls ‘hybrid intelligence.’

The company uses AI to process large volumes of unstructured data, then applies human judgment to validate context, tone, and intent.

Ray believes this two-layer system dramatically increases sentiment accuracy, and by extension, business relevance.

“AI brings the scale, but human expertise brings the context, interpreting tone, nuance, and intent that machines alone often miss,” Nic says.

The approach enables DataEQ to triage customer service issues more effectively, combining the large data restructuring from AI with the human nuance and understanding.

In other words, DataEQ believes this turns sentiment into a reliable metric for companies and CEOs, not a mere guess.

From Crisis Management to Brand Turnarounds

DataEQ’s methods have already delivered measurable outcomes in the real-world.

During the Cape Town water crisis, when public trust was fragile and misinformation rampant, local government officials used DataEQ’s platform to guide their communication strategy.

“Our sentiment analysis surfaced public concerns, misinformation trends, and shifts in sentiment, enabling authorities to adjust messaging on the fly,” Nic says.

In another case, Telkom South Africa used structured sentiment data to overhaul its customer experience strategy.

This led to a 38.2-point improvement in Net Sentiment over three years, and that metric became a core KPI tied to executive decisions and business performance.

Metrics That Matter

Ray believes the days of relying on vanity metrics, like engagement counts or follower growth are over — and honestly, it’s a sentiment we’ve seen shared by other agencies we interviewed recently.

CEOs and decision makers should be more concerned about metrics that actually matter, including:

• Net Sentiment Tracking: a continuous signal of brand health.
• Sentiment Drivers: to understand why customers feel the way they do.
• Resolution Rate: to track how well customer issues are handled.
• Retention and Growth Signals: to surface at-risk or high-potential segments.

“If you misread how customers feel as your starting point, every downstream insight is compromised,” Nic says.

For CEOs and other leaders, this means building strategies on a foundation they can trust and use, not just numbers that look pretty on a dashboard.

The Future of Sentiment and AI

Ray sees a growing reliance on predictive AI in customer data analysis, but he is keen in reaffirming a clear caveat:

It must be built on context-rich data, and that can only be achieved by keeping human judgment in the loop.

“Huge value lies in context. Organisations that build on human-informed data foundations will gain a sharper, more dependable lens on customer behavior.

That’s the only way to convert insights into real-world impact," he added.

The DataEQ perspective is clear: The companies that succeed is those that treat sentiment strategically, not by using AI as a replacement — but as a complement to human judgement.