Six Big Mistakes Data Professionals Make When Delivering Results

May 02, 2023

You’re getting ready to share exciting analytic results! You’ve worked long and hard on the analysis. You even have cool, 3D graphics to make the findings compelling.

But when the time comes, you get blank looks. Your audience doesn’t seem to grasp how important this is. How does this happen?

Don’t the results speak for themselves?  No. They don’t.

And, in case you are wondering, complex methods or fancy visualization formats don’t necessarily fix the problem, either.

Some of the biggest mistakes people make when delivering results have nothing to do with the:

  • Quality of work,
  • Appropriateness of the methods,
  • Attractiveness of the presentation, or
  • Talent of the analysts.

No.  When the audience gets lost, it is most likely because the data professionals focus primarily on their own work, rather than what the audience needs. 

Below are common—easily fixable—mistakes data professionals make when delivering results.

  1. Jumping in without orienting to the original question

Analytic teams often forget that, while they have been buried knee-deep in their work, other teams have been focused elsewhere. They may not remember what this analysis is about, or if they do remember, they have been distracted by one hundred other things. 

A meeting about results needs to begin with a reorientation about what the purpose of this project was, and what question is being answered.

  1. Skipping context

In addition to not reminding everyone about the exact question being asked, analysts often miss the opportunity to further ground the project in clear statements about its relevance to the audience.  This can include the problem being solved, the decisions being made, or the actions being taken.  It can also include context about how the results fit into other initiatives that are important to this audience.

For example, instead of simply saying:  as a reminder, our goal was to see if we could predict turnover, the analytic team can elaborate with other pertinent information. Given the historic rise in turnover after the end of COVID lockdown, leadership identified employee retention as the top goal for 2023. You might recall that turnover in Q4 reached 17%, compared to only 7% the previous year. You might also recall that turnover seems worse for these three locations. National statistics show that the labor market remains tight and will likely stay that way through 2024. The cost of excess turnover to the business is estimated at over $4M annually. So, the purpose of this project was to identify factors that might predict turnover ahead of time, so we might have opportunities to intervene proactively.

This reorientation might only take one or two minutes, but it pulls the audience fully into the purpose and relevance of the full project.

  1. Pointing out what is wrong first.

Data scientists are trained to identify potential errors, biases, flaws, and limitations. It is a critical aspect of their training, leading them to be extra cautious about making definitive statements.  As a result, they have a tendency to lead with the reasons their conclusions might be wrong or suspect, believing it will help the audience interpret the findings correctly.

However, when the first statement is hesitant or highlighting problems, what business professionals hear is uncertainty. Instead of giving the message that the analysts are being appropriately cautious, it undermines confidence and the business audience tunes out.

       4. Getting bogged down in the details or jargon

It’s natural to get excited about our work and want to share details about how it was done.  But giving details about methods and nuances of the analysis simply distracts from the overall results.  Similarly, even basic statistical terminology may confuse or alienate a non-analytic audience, distracting them from what’s important.

Instead, data professionals should review language in advance, identify any terminology unique to analytics, and select clear, lay vocabulary to substitute. Taking an extra moment to translate will go a long way in engaging the audience.  

  1. Focus on a finding instead of implications

It’s not uncommon for data professionals to prioritize statistical findings as their main focus in a presentation. After all, that’s been their goal all along! So, they spend the majority of a report on what they found out, rather than how it matters to the audience.

For example, the analytic team may report that they found eight significant predictors of turnover, listing them in order of impact, and explaining the model accounts for 40% of the variability in turnover rates. All of which might be true, but it doesn’t really help a business audience grasp what it means.

An alternative is to take predictors that might be actionable variables (such as performance, lack of a recent raise, or having low employee engagement in their department) and model the differences in turnover. This tells the audience that compared to high performers who had a recent raise and work in highly-engaged departments, those not getting a raise and working in lower-engaged departments have five times the rate of turnover. And those at greatest risk account for 23% of employees, so they can be targeted. This tells a story they can grasp and apply.

  1. Discouraging meaningful feedback

Finally, it’s common for data professionals to wrap up their presentations with Yes or No questions to encourage acceptance, such as: Did that answer your question? Make sense? Did this meet your needs?  Such questions are what I call “begging for yes.”

Unless the findings are completely off base, the audience almost always says Sure, or Yes. This response allows the data team to declare the project done and move on.  However, it does not encourage meaningful feedback to improve projects in the future.

Instead, at the end of a presentation, the data team can invite useful feedback with open-ended questions, such as: Can you share your initial impressions now that you’ve see these results for the first time?  Or posing a question that allows more than a Yes-No response, such as:  I’m curious, how well did this project meet your needs?  That way, the audience can share the parts that did and did not meet their objectives.

Being audience-focused helps you avoid these mistakes.

Nothing listed here is difficult to avoid, it just takes awareness and a few minutes of preparation. An underlying principle that can help you prevent these missteps is being audience focused rather than analytic focused.

We begin by reminding and reorienting the business team with clear statements and impactful context.  Then we tighten our messages, using clear language, focusing on overall conclusions, and avoiding jargon, excessive detail, or pointing out limitations at the beginning.  We also translate the results into useable, understandable implications and invite open feedback.

By using these simple guidelines, you dramatically increase the likelihood that your next presentation will deliver real value and highlight the capabilities of the analytic team.  Plus, now that you have engaged your client in meaningful dialog, you can begin exploring what comes next to continue the cycle of informed decision-making using analytics.

For more about ways to apply Analytic Translation in practice, see the resources and training here.

Wendy D. Lynch's headshot photo.

Wendy D. Lynch, PhD

Wendy Lynch is an experienced sense-maker and data scientist with over 35 years of research experience, primarily in business settings. She has played the role of Analytic Translator for hundreds of companies, from start-ups to Fortune 100 corporations. Her expertise is both in data analytics and effective communication, combining the two into a framework for optimizing the value of analytics in a business setting.  Connect with her through LinkedIn or email.

Newsletter: A Matter of Translation

Making sense where business and analytics meet.
Training professionals to speak both languages.

We won't send spam. Unsubscribe at any time.