cplysy

So your organization gave you a Tableau license. Or perhaps you attended a half-day PowerBI workshop. Now you’re ready to use your new tools and make some of your data interactive.

But should you?

Not everything should be interactive!

Tools like Tableau and PowerBI give you the ability to pack a lot of data into a small space. This can be incredibly useful when you have a lot of data that you want to open up for exploration.

Good interactive visualization can reduce the overwhelm felt by the data user. It does this by layering the data and presenting just little bits at a time. Because when you have thousands upon thousands of rows of data you need to simplify the experience.

But when you only have a little bit of data, creating an interactive dashboard does more harm than good. Instead of opening up a large dataset, you end up hiding data from your dashboard user. In order to find all the data that could easily fit on a single page, they need to click, and click, and click.

Example: The National Health and Nutrition Examination Survey (NHANES) Interactive Visualizations

NHANES is a big multi-year multi-million dollar CDC study that has collected lots of data over the years. And according to their website, they have some new interactive visualizations.

Their visualizations are some basic Tableau dashboards showing the prevalence of high total cholesterol, hypertension, and obesity in the US across ten points in time. The data is also filterable by Sex, Age Group, and Race and Hispanic Origin.

The initial visual we are given is a line graph showing high total cholesterol representing all adults 20 and over.

For all its options and buttons, this is a small set of data for a dashboard. You find that out quickly if you download the CSV file.

The data for all three topics, sex, and age groups fits within 6 columns and 360 rows of data.

The race and hispanic origin data was not found inside the CSV, but it really doesn’t add a ton of data to the overall dashboard. Especially since the data is not broken down for every age group and there are some missing values due to low sample sizes for those groups.

All in all, the data represents hundreds of rows not thousands.

An alternative: create a series of charts.

The alternative I would pitch would be pretty simple. Instead of creating an interactive dashboard this data could be presented in a simple series of charts. Then you add narrative around each giving additional explanation and context.

For instance here are the charts for all adults 20 and over for each of the three topics (high cholesterol, hypertension, and obesity).

These same three charts could easily be visualized in a single chart. We can even use some subtle interactivity through a tool like Flourish. Here is an example of the same data represented in the above three dashboard views but in a single chart.

One chart will not replace the full dashboard.

That would just be the initial chart.

I would then split up the three topics. And for each topic, I would show the data split by sex, age, and race.

In total, it would be ten charts with supporting narrative. None of the data would be hidden, and you wouldn’t need an explanation about how to explore the data. The page would feel more like reading a news article.

The Takeaway.

If your interactive dashboard would need 100s of charts to show the same amount of data, then a dashboard is the right call. But if your interactive visualization could be replaced using a handful of charts, you should just use the handful of charts.

It’s late August, and are into another hot week here in North Carolina. So yet again I was outside with the hose watering our little vegetable garden. And in that meditative moment I thought about The Lorax.

I am the Lorax. I speak for the trees. I speak for the trees for the trees have no tongues.

The Lorax by Dr. Seusss

Even as a kid I really felt that story. And I think it connects with what we do when we report, but maybe not in the way you might think.

By now you might be sick about me talking about my book (The Reporting Revolution, now on Amazon). But honestly, even though it does have almost everything to do with my consulting work, the book was still a passion project. I believe that if we want our work to stay relevant we need a mindset change when it comes to reporting.

We don’t need to speak for the data.

When a lot of researchers and evaluators talk about reporting, they’re really talking about documenting.

Don’t get me wrong, there is a real need to document our work. To write down everything about our methods, our analyses, our findings, and our recommendations. This is why I don’t despise 100 pagers. They have a role to play, it’s just not as a report.

Reporting is different than documenting. Reporting is less about capturing and more about communicating. Communicating is about connecting with real life people.

When you spend a lot of time with your work you start to believe that it needs an advocate. For example, if you have dedicated your last few years to collecting data, you will likely feel a responsibility to advocate for that work.

But our data work is a means, not an end. Data is only valuable in so much that’s it’s useful. And useful requires a user.

We do need to advocate for the people.

Thinking about your audience as you are documenting your work is not enough. You need to be an advocate for the people who could use your work.

Unlike trees, people have tongues. But most potential data users do not have a seat at the table as you collect and analyze your data. So they don’t have a voice in the decisions you make when sharing your work.

Being an advocate for your audience means seeing your data through their eyes, not your own.

This is why a report can be 1 pager or an infographic. Because the goal for these designs is rarely to document a project but communicate something of value for a specific audience.

If you can, join me tomorrow.

It’s a launch party, it’s going to be informal and fun. It’s also free with door prizes!

The proof copy is on its way and if everything goes right, I will have The Reporting Revolution up on Amazon next week. So what now? It’s time for a party!

First, Thank You!

All told over 450 of you downloaded a free draft version of the book. I decided to share the book before it was finished because I knew it would help drive me to write, heighten my audience focus, and receive some good feedback.

I was not wrong.

The support and interest definitely helped me bring this book to market lightning fast.

Virtual Book Tour

This fall I want to have a modern virtual book tour and I could use your help.

I would like to virtually visit with:

• Local Association Groups
• Non-profits
• Evaluation Classrooms
• Evaluation Agencies
• Podcasts (as a guest)
• Blogs (as a guest poster)
• and anywhere else where I can talk to at least a handful of interested evaluators and researchers.

(I bring my own Zoom room but can also use your own meeting platforms)

If you’re interested please send me an email (chris @ freshspectrum.com), message me on LinkedIn, or DM me on Twitter (I have opened up access so you can DM me even if we don’t follow each other).

Launch Party

Next Wednesday I’m having a book launch party. We’ll have fun with a Q&A, games, cartoons, and course/consulting giveaways worth hundreds of dollars.

If you think you can make it, Please RSVP. You have to show up to win prizes.