Changing stubborn opinions with visualizations [case study]

Posted on May 29th, 2015 in Charts and Graphs - 28 comments

We, analysts take pride in the fact that we tell stories. But what if you have a boss, client or colleague who wouldn’t buy the story?

This is a problem we face often. Let’s say your boss has stubborn opinion about something, like more advertising leads to more sales. You know the data doesn’t support this theory. But how do you change your boss’ mind?

Here is an interesting way, showcased in NY Times recently.

family-income-vs-college-enrollment-nytimes

Changing stubborn opinions with visualizations – 3 step process:

Assuming we are talking about ad spend vs. sales example:

  1. Ask your boss to draw a line that (s)he thinks to be true.
  2. Then show the line from original data (or observations).
  3. Tell your boss how accurate / wrong his(her) line is.

This technic can be very persuasive if you make it interactive.

For more on how NY Times implemented it, check out this page:

How family income affects children’s college chances

Your thoughts please:

Here are 2 questions for you:

  1. How do you change someone’s mind using data?
  2. How would you implement these concepts in Excel?

Go ahead and share your thoughts in the comments section.

I have a stubborn opinion that only 1% people who read articles on Chandoo.org leave comments. Change my mind. 🙂

 

Written by Chandoo
Tags: , , , , , ,
Home: Chandoo.org Main Page
? Doubt: Ask an Excel Question

28 Responses to “Changing stubborn opinions with visualizations [case study]”

  1. Duncan says:

    I went to the NY Times site and took part in their survey. So far 58,474 people have tried it. They told me, "You drew a more accurate picture of reality than about 88 percent of people who have tried so far.""

    It's a good exercise by the way and it's something that data analysts and data scientists do every day: anticipate what they might find and/or use the actual to explain how right or wrong they were!

    Thanks for sharing Chandoo!

  2. While data can change minds, sadly most people will never use data to change the gut feeling. I've often had discussions with people who are gut feelers and dismiss data because it isn't real or it will cost money or their jobs.

    Oh well, that's why people like us exist, right?

  3. Tim Rodman says:

    I really like this article.

    This line approach is awesome for two reasons (in my opinion):

    1. You aren't asking your boss to give you a number, but draw a line. Drawing a line is a much more visual and engaging activity.

    2. The "guess" vs "reality" results are more educational than if you just compared two numbers. If you just compare two numbers, then your boss is either "wrong" or "right". If they were wrong, then they will argue with the data. If they were right, then they didn't learn anything. But with the line, it could reveal that they were "wrong" in some areas, but "right" in others. If your boss sees that they were "right" in some areas, they will be more open to being educated in the areas where they were "wrong". You have to massage the egos of those in power a little if you want to them to listen to you.

  4. Isaac says:

    Great topic!

    Let's suppose that you have drawings of "stubborn opinions" from several bosses. How could you incorporate those all into the chart in an aggregate format so that you don't make any of the bosses look particularly "stubborn"?

    I recently made a presentation -- and before "revealing" the results for a particular item to attendees, I asked them to draw lines on blank pages that I provided in advance of the presentation.

    And I collected their drawings.

    So, the next question: How can I aggregate or illustrate these opinions to compare them with the actual data?

    • jason says:

      You could print the actual results on to their drawings (if the drawings would fit in a printer and are somewhat close in scale)) then you could scan them into a pdf and present it electronically (no names of course) or print them and hand them out.

  5. Isaac says:

    Re: aggregating the data

    I followed the link in the article to the NY Times and took the survey -- and it had an example, aggregating the data from other respondents in a purple haze. I like it 🙂

    Any suggestions on creating something like that with Excel?

  6. Ilyas says:

    Really great approach of justifying your idea.

  7. mark says:

    Ok, great article, nice way to aggregate the data. Now I just have to learn how to do this. There Chandoo, I have added my comment, just for you!

  8. Doug says:

    Excellent article. The only danger with this approach is if your boss draws a line totally opposite of the actual result. You may not like their reaction.

  9. Duncan says:

    There is an equally important aspect to this which is the future. It is typical in the business world for forecasts to follow the Hockey Stick Effect. Ask your boss or your accountant to create a forecast or budget, especially for a company currently in trouble. They will typically draw a line that starts at x,y that will then fall, ie things will look worse in a year's time, then the curve might flatten but then it will rise and rise.

    In the end the curve looks just like a Hockey Stick: no! not an Ice Hockey stick but the kind of Hockey that they excel at in India!!!

  10. lalitha says:

    Very Informative post

  11. JP says:

    Very nice article. Yeah, it'd be nice if we can easily do this in Excel, not sure though how we can create that interactive line input.

    Hey Chandoo, comment done! 🙂

  12. Kent says:

    Good article. Fortunately for me, our leadership is data driven and therefore we are a very successful Company when it comes to uncovering root cause and resolving issues permanently.

    Chandoo - - Are we getting close to the >1% mark yet?

  13. Kit says:

    Your site, tutorials and articles are brilliant. I only wish I had more time to learn how to implement all the ideas. Thank you for sharing your tribal knowledge.

  14. Anthony says:

    As a teacher and life long learner, I love your data ideas, slicers, pivot tables, dash boards, and awesome ideas related to data. Keep inspiring!

  15. AdAdilson says:

    IIcouldnt agree more with you. Data always must tell anan history or we are losing our time with fancy design that is not produtive.

  16. Justin says:

    Excellent article and link! I will be using it in my Managerial Decision Making classes when we discuss biases.

  17. victoria says:

    Hi Chandoo,
    I love your articles, but normally do not leave comments. I am writing to change your opinion about 1% of your readers leaving a comment. Your articles are always useful, practical, enjoyable and fresh. It is like unwarapping a box with a present. You know that something exciting and always fresh and new will be there. Also they help me to broaden my knowledge, develop my skills and make a song out of dull work. Thanks heaps!

  18. Julie says:

    Great article. I love your courses. I recently was trying to figure out why our productivity had declined. I thought I knew what it was and made several dashboards and could not prove my hypothesis. Turns out it hadn't declined, we just had added more hours to the productivity ability of the department. DATA is always right, you can't argue with facts.

  19. Tjawiee says:

    Done it, awesome information. Thank you. 🙂

  20. Lalitha says:

    Thanks a lot for the useful information.

  21. Anji says:

    Hi chandoo,
    I would like to know how this can be done using excel.... you are too good please keep sharing your knowledge with us....

  22. William says:

    Someone go do this on Excel and share your workbook with the community.

  23. Dave Paradi says:

    Here's a video of how you can do this in PowerPoint: http://screencast.com/t/gskX1Wamc

    Dave Paradi, PowerPoint MVP
    http://www.ThinkOutsideTheSlide.com

  24. Marla says:

    Very compelling; thanks, Chandoo.

  25. Fred says:

    Especially interesting is how their program automatically analyzes the guessing and returns text. Visualizations are great, but I can see the value of some text summaries:
    •About 66 percent of people drew a more accurate picture of reality than you did.
    •You underestimated the chances of college enrollment for the very poorest children. In reality, about one in four children in America’s poorest families go to college. (You guessed around 10 percent.)

  26. Chris says:

    Hi, I was challenged by the 1% statement, so here I am with my first ever reply. Q how will you determine the really true fact: how many readers do you actually have? My interest was captured by a previous statement about people's gut feeling. In my field too many decisions are made by leaders with too little information bec obtaining the facts is impossible. Here is a challenge for me. Thanks Chandoo, please keep up the daily thoughts.

Leave a Reply