31 Data Trials and Triumphs

31.1 Introduction

A data analytics professional’s core responsibility is to provide the data necessary for informed decision-making and to frame their analysis in a way that helps business leaders make the best possible choices.
In this section, we explore the role of data in decision-making, understand why data analytics professionals are vital to this process, and examine how the right or wrong use of data can lead to either project success or failure.

Both data-driven and data-inspired approaches share a common foundation: they treat data as a valuable input to guide and strengthen business decisions.
However, it is essential to remember that while data improves decisions, it does not make them. Human interpretation and context are equally important.

31.2 Data-Driven Decisions

Data-driven decision-making involves using quantitative facts and evidence to guide business strategy.
In this approach, decisions are based solely on data analysis — what the numbers reveal.
The accuracy of data-driven decisions depends heavily on the quality, completeness, and reliability of the data being used.

While data-driven decisions can lead to powerful results, they can also be limited by overreliance on historical data or by neglecting qualitative insights, such as human judgment or context.
In addition, biases in data collection or processing can result in misleading outcomes.

Example: A/B Testing for Website Optimization

Consider a company that sells widgets online. The team believes a new website layout could increase sales.
For two weeks, half of the visitors see the old layout, while the other half see the new one.
At the end of the trial, data shows which version led to higher widget sales.
If the new layout performs better, the company can confidently adopt it.
This is a data-driven decision—based purely on performance metrics and quantitative evidence.

31.3 Data-Inspired Decisions

Data-inspired decision-making expands beyond raw data by incorporating context, intuition, and qualitative insights.
It combines facts with human understanding, drawing on experiences, related concepts, and creative reasoning.
This approach avoids some pitfalls of pure data-driven thinking by balancing evidence with interpretation.

Example: Improving Customer Support

A customer support center collects Customer Satisfaction (CSAT) data using a 1–10 rating scale and written feedback.
The manager reviews both the numerical data and the customers’ comments. They also interview support staff to understand their experiences.
By combining quantitative analysis with qualitative input, the manager identifies areas for improvement and creates a more informed strategy to enhance customer satisfaction.
This is a data-inspired decision, as it merges analytical insight with human judgment.

31.4 A Data Analysis Triumph: PepsiCo

When data is used strategically, it can transform business performance.
A good example of data-inspired success is PepsiCo, which revolutionized its marketing strategy by integrating data analytics into decision-making.

According to Shyam Venugopal (Think with Google), PepsiCo: - Hired analytical talent and built cross-functional workflows centered on consumer data. - Established processes to make decisions grounded in data and technology. - Created a cloud-based data hub that unified information across sources for a holistic understanding of consumers.

This transformation allowed PepsiCo to use both internal and external data to anticipate customer intent, personalize experiences, and adapt marketing in real time.
By adopting a data-inspired approach—grounded in evidence but enhanced by creativity and empathy—PepsiCo strengthened its global brand and customer loyalty.

31.5 Data Analysis Failures

Even accurate data cannot guarantee success if it’s incomplete, misinterpreted, or applied without proper context.
Let’s look at two famous examples of data-driven failures that illustrate how misused data can lead to costly mistakes.

Example 1: The New Coke Launch (1985)

Coca-Cola conducted taste tests with 200,000 people and found that most preferred the taste of New Coke over Pepsi.
Based on this data, the company discontinued Classic Coke and replaced it with New Coke.

However, the data ignored an important factor — emotional attachment to the original product.
Consumers rejected the change, and the company suffered heavy financial losses before reintroducing Classic Coke.
This failure showed how incomplete data can lead to flawed decisions, even when the data itself is accurate.

Example 2: NASA’s Mars Climate Orbiter Loss (1999)

NASA’s $125 million Mars Climate Orbiter was destroyed when it entered Mars’ atmosphere too low and burned up.
The cause? A unit conversion error — one engineering team used pound-force, while the navigation team used newtons.
Although both sets of data were accurate, they were interpreted inconsistently.
Had the teams communicated clearly or standardized their measurement units, the mission could have succeeded.
This highlights that data accuracy is meaningless without consistency and context.

31.6 Key Takeaways

Data enhances, but does not replace, human judgment.
Data-driven and data-inspired approaches both rely on evidence, but data-inspired thinking integrates creativity and context.
Data-driven approaches emphasize accuracy, measurement, and historical insights.
Data-inspired approaches blend data with experience, qualitative feedback, and innovation.

As a data analyst, your goal is not to choose one method over the other but to combine them effectively.
Use data to guide, not dictate, decisions. Stay curious, ask thoughtful questions, and interpret insights with care and fairness.

In short: The best decisions are data-informed—rooted in evidence, shaped by context, and strengthened by human understanding.