Are Managers an Unnecessary Filter in Data Interpretation?

Decisions based on intuition or emotion can lead to uncertain outcomes, particularly when you consider the wealth of data your website accumulates. Embracing data-driven decisions is not just advisable; it’s imperative, especially when guided by someone skilled in data science.

Data, in its essence, reveals the unvarnished truth.
A data analyst’s role is pivotal in ensuring that this truth is interpreted accurately and faithfully.

However, what happens when the data’s journey from collection to interpretation is marred by unintentional bias? When it’s twisted or contorted to align with a preconceived narrative, emotion, or intention?

Imagine the introduction of an additional layer of interpretation at a critical juncture – a manager attempting to decipher the findings anew, right at the culmination of the process.

A simple unnecessary filter – skewing and twisting the results.

A Data Scientist’s Role In Marketing

Data scientists in marketing possess the unique ability to sift through numbers, trends, and patterns to extract actionable insights.

Their expertise not only lies in handling complex datasets but also in employing advanced analytical techniques to predict user behavior, optimize marketing campaigns, and enhance customer experiences.

This direct interaction with data allows for a nuanced understanding that is critical in the fast-paced digital marketing realm.

If you’re collecting data in Google Analytics, Search Console, or tools like SEMRush, it’s easy to assume those tools present the truth, a fully rendered representation of the data. But it’s simply not true.

The job of a data scientist/analytics is to dig deeper, collect the raw data and transform it into something tangible. Ultimately presenting recommended actions after taking in all other variables.

So, where does a manager fit in?

Understanding the Data Science Process

Let’s first look at what typically happens when data needs processing…

  1. Data Collection: Gathering raw data from various sources.
  2. Data Cleaning: Pre-processing and cleaning data to ensure accuracy.
  3. Data Analysis: Applying statistical methods and algorithms to understand trends and patterns.
  4. Data Interpretation: Translating analysis results into actionable insights.
  5. Data Visualization: Creating visual representations of findings to facilitate understanding.
  6. Decision Support: Providing recommendations based on data-driven insights.

At each of these stages, the data scientist’s work is methodical, precise, and aimed at uncovering the truth hidden within the data. This process is a science, requiring a deep understanding of both the technical aspects and the business implications.

What is Managerial Filtration?

When managers act as intermediaries between data scientists and decision-makers, they can inadvertently inject themselves into this process at various stages, potentially skewing the immutable output.

For instance, a manager might prioritize certain data sources over others, influencing the collection stage, or push for a particular interpretation of the data that aligns with their expectations, affecting the interpretation and decision support stages.

This managerial filtration can lead to a diluted or distorted view of the data, compromising the integrity of the insights derived from it.

Hidden Costs of Managerial Gatekeeping

Imagine an e-commerce website and the marketing team is tasked with improving the website’s conversion rate. The data scientists within the team are analysing user behavior data to identify friction points in the user journey that might be causing potential customers to leave the site without making a purchase.

The team discovers several key insights:

  • Pages with slow loading times have higher bounce rates.
  • Product pages lacking clear calls-to-action (CTAs) have lower conversion rates.
  • And a streamlined checkout process could significantly improve conversions.

However, the manager overseeing the project prefers to be the gatekeeper of information and the primary decision-maker.
Instead of allowing the data scientists to present their findings and recommendations directly to the team, the manager decides to report the data themselves, filtering and interpreting the findings according to their understanding and priorities.

There is no room for emotion. No room for desire. It’s pure science.

Based on the manager’s presentation, the team allocates resources to redesign the homepage and launch new marketing campaigns.
While these initiatives might have some positive impact, they do not address the critical issues identified by the data scientists.

The site continues to suffer from high bounce rates and suboptimal conversion rates because the root causes – identified through data analysis – were not adequately communicated or acted upon.

How Can We Remove The Filter?

Promote Direct Communication

Establish direct channels for data scientists to present their findings to decision-makers and action insights. This can be facilitated through regular insight-sharing sessions or integrated digital dashboards that provide real-time access to data insights.

Educate and Train

Invest in cross-functional training for managers and data scientists. Managers should have a basic understanding of data analysis principles to appreciate the value and limitations of data insights. Similarly, data scientists should be aware of the broader business context to tailor their analyses and recommendations accordingly.

Implement Collaborative Decision-Making Processes

Engage your data scientists in key decisions, strategic campaigns and important facts about the business or direction that the wider teams and stakeholders are taking. While this won’t affect the results, it does mean that the recommended actions can be provided in context that fits the goals, budget, or limitations of the business.

Advocating for a Streamlined Approach

To harness the full potential of data-driven insights in marketing, organizations must reconsider the traditional hierarchy that places managers as gatekeepers of data.

Encouraging direct communication and collaboration between data scientists, senior decision-makers and other teams can enhance the accuracy and relevance of insights.

This streamlined approach not only accelerates the decision-making process but also empowers data scientists to take a more strategic role, ensuring that their expertise directly informs business outcomes.

Fostering this kind of action-based data science, allows for faster processes, and ensuring that your raw data isn’t being filtered by unnecessary steps or interpretation – leading to improved outcomes and higher chances of success.