Data and the myth of actionable insights

Management

Author Karl Dinkelmann at Nexus Data debunks one of the biggest myths about data analytics and AI

Countless companies spend millions on data analytics and AI projects without ever seeing a return. These initiatives promise to deliver actionable insights. But for business leaders concerned with the bottom line, such investments are often coupled with confusion.

Instead of driving real results, data teams and CFOs end up speaking different languages—leading to misalignment, wasted resources, and ultimately, failed initiatives.

The most recent reliable research on the success rates of data analytics projects is a 2018 Gartner report. This report predicted that by 2022, only 20% of data analytics projects (one in five) would generate value. Our recent observations indicate that success rates have declined as more companies experiment – and fail to generate value – with data analytics and AI-led initiatives.

Generating value from actionable insights

While having dashboards, reports, and AI-led initiatives in place is essential to business success, many leaders wrongly assume that these will somehow automatically lead to growth. The hard truth remains: insights on their own do nothing but create more noise.

Think of insights like a high-performance car. If it just sits in the garage, it has no value. It’s only when you drive it and use it to reach a desirable destination that it becomes a worthwhile investment.

The idea that actionable insights alone produce value is the biggest myth in data analytics. Without the utilisation of these insights in business processes driving value-adding decisions within the window of maximum value, across a broad range of employees operating at the coalface of the business, the value of actionable insights is severely limited.

Determinate versus indeterminate value

For insights to translate into value, it’s vital to differentiate between determinate and indeterminate value and for business leaders to rethink their approach to data analytics.

Indeterminate value is vague; it sounds good in theory but lacks a clear link to financial or operational impact. This could play out in several ways, one being the implementation of a new AI-driven dashboard to track customer sentiment. While the plan to deliver “actionable insights” may sound impressive, a CFO will ask: "What’s the actual business impact?"

On the contrary, determinate value is value you can measure and track—cost savings, revenue growth and efficiency gains. The above scenario could be more firmly tied to business outcomes by embedding the AI-driven customer sentiment insights in customer service processes and retention campaigns to reduce churn by 15%, adding $50m in retained revenue. Would you agree that this facilitates a more robust conversation between data and business leaders?

The biggest mistake in data analytics is focusing on insights that create indeterminate value instead of embedding insights to drive actions that deliver determinate value.

Speaking the same language

Let’s face it: finance and data leaders rarely speak the same language. As my colleague and co-author, Zjaén Coetzee likes to say, “Business people don’t speak Data, and data people don’t speak Business.”

However, by becoming fluent in business value as a common language, companies can bridge the gap between different departments and align to achieve collective success.

Data leaders, if you walk into your CFO’s office and say, ‘We need $10 million for a data project,’ you’ll probably get a blank stare. But if you say, ‘We need $10 million to reduce operational costs by $50 million over three years, and here’s the tangible evidence from a proof of value we executed,’ the conversation changes entirely. While the same amount of investment is required, the latter scenario speaks the language of value, while the former doesn’t.

Therefore, it is vital that data leaders remind themselves of this fundamental fact: businesses exist to drive value, and actionable insights on their own don’t produce real value.

Why 80% of data initiatives fail

Companies don’t need more insights—they need the right insights that translate into measurable business impact. Insights uncoupled from value-driving actions create more noise in an already noisy system. Every data project must start with a clear financial objective.

Four out of five data and AI-led initiatives fail because companies get stuck at the insights stage. Businesses treat insights as an end result rather than a starting point for value creation.

Moreover, after investing millions in dashboards, AI models, and predictive analytics, it’s often true that data teams operate in siloes, generating reports that are not clearly aligned to any specific strategic outcomes. Meanwhile, executives face "analysis paralysis"—drowning in insights but struggling to take action.

And this is exactly why we have developed a proven methodology for business leaders to address this issue.

Data-confident leaders

To avoid the pitfalls that befall so many companies, as a business leader, it is your responsibility to become data-confident.

Data confidence is a key differentiator between companies that generate substantial value from data analytics and those that do not. The most successful companies integrate data analytics into their core strategy, with a strong agenda led by C-suite executives.

These data-confident leaders identify the right initiatives, ensure value generation, and sustain that value by recognising and improving it. They avoid hype and distractions, maintaining a sharp focus on value creation that guides every decision.

Instead of being data-driven with value as a secondary consideration, aim to be value-driven and enabled by data. This mindset switch, which can be brought about by becoming data-confident, solves many challenges.

The RAPPID Value Cycle

In my 20 years in business, I’ve found that what sets these top-tier companies apart from the masses, is that they have embedded data-led decision-making into their operations. In other words, their executives—including their data leaders—speak the same data-confident language focusing on value to achieve shared success.

The top-tier companies that successfully translate data into business impact have shifted their focus from insights to execution.

The RAPPID Value Cycle is a proven approach to generating measurable value from data. At a high level, the various components of this methodology ensure that every insight leads to structured, measurable action.

Recognising value
Approach to delivering insights
People and data value culture
Platform optimised for value
Investing in insights
Data trust

Let’s take a real-world scenario: By implementing the RAPPID Value Cycle, a mining company was able to identify and address overspend on overtime of more than $5 million per year. By delivering valuable insights highlighting specific departments and teams that were overspending, supervisors could take value-driving actions, based on those embedded insights, to curb the leakage.

Through this shift of approach, from delivering insights to driving business impact, the RAPPID Value Cycle enforces structured execution—ensuring data analytics isn’t just a reporting function but an operational driver.

That’s the difference between insights and impact. After all, the only data that matters is the data that improves your bottom line. Everything else is just noise.

Dispelling the myth

We have addressed the biggest myth in data analytics: that actionable insights alone don’t produce real value. Without data-confident leaders applying these insights in business processes, potential value remains untapped. True value comes from driving value-adding decisions across a broad spectrum of employees – turning strategy into action every day.

The RAPPID Value Cycle was developed to turn data into results, insights into action, and analytics into business growth. It provides a practical roadmap for business leaders who wish to become data-confident.

The real question isn’t whether your company is investing in data. It’s whether those investments are delivering the real business results needed to drive long-term success. If they aren’t, it’s time to rethink your approach.

Karl Dinkelmann is a seasoned data analytics expert and business strategist and the co-founder and CEO of Nexus Data. For more information about his book, Drive RAPPID Results from Data, visit rappidvaluecycle.com.

Main image courtesy of iStockPhoto.com and NicoElNino