Book Review: Storytelling With Data

Book Review - Storytelling With Data

Usually when talking about data one would refer to number crunching and staring at spreadsheets. In recent years however with the growing popularity of infographics and web analytics data goes far beyond just numbers. It is a visual process that require creativity, correct use of colors, correct use of graphs and most importantly storytelling.

I work a lot with data related to digital marketing and web analytics. I analyze a lot of websites and their marketing performance, user engagement performance and conversion performance and visualizing data is a big part of what I do.

A few months ago I stumbled upon Storytelling with Data blog, a blog which has actually been around for quite some time so my bad I only discovered it recently. It was clear from the very beginning that I had a lot to learn about data visualization from the awesome author Cole Nussbaumer Knaflic.

And it didn’t take me long to get convinced to buy her book Storytelling with Data.

The book is divided into ten chapter each one teaching a different concept in data design with real life examples and case studies. What I found particularly useful was Cole’s lessons in:

  • how to choose the most appropriate type of graph for your data (always a challenge when I’m working on my reports)
  • how to carefully choose colors
  • how to direct your audiences attention to where you want it

But what is most impressive about this book is its dedication to teach us how to leverage the power of storytelling to help get the message across and make stick with our audience. Chapter Seven called Lessons in Storytelling deals specifically with that looking at examples of storytelling in plays, screenwriting and creative writing and how the concepts can be applied to data visualization.

The book uses some really basic examples but as I was reading through all the tips and suggestions in the book I was amazed to realize how many silly mistakes I was making in visualizing my data in the reports.

Storytelling with Data specifically uses presentations and PowerPoint decks as an example. I wish the book had touched upon how to apply all these concepts and lessons when you deal with dashboards and large reports. In my line of work I sometimes need to put several visuals in one page. This is my biggest challenge every month when I prepare report for clients – how can I draw their attention to where it is most needed and how can I tell a story when there are so many graphs and tables in one place.

Still this book is a great starting point and an excellent resource to help you with any data visualization challaneges you are facing. You will get a basic introduction to data graphics, will learn the importance of color and how to apply storytelling concepts into data visuals.

How to Build Tables in Google Data Studio

Google Data Studio offers the abillity to build pretty tables with a lot of options to highlight your most insightful data.

In this video I will build a table with two dimensions and two metrics. I will show how to use bars or a heatmap in one of your metric columns to highlight the highest value. In this example we look at a table with landing pages per channel, sessions and eCommerce conversion rate for each landing page. The table will be sorted by highest number of sessions and I will implement a heatmap on the eCommerce conversion rate column to highlight the page with the highest conversion rate.

Also in this video you will see the new data panel which was very recently released by Google Data Studio. This new panel makes easier and faster to change or add new dimensions and metrics to your widgets.

If you are just starting with Google Data Studio check out my detailed Google Data Studio guide

Also check out my Google Data Studio examples of an Organic traffic dashboard and an eCommerce dashboard

Google Data Studio Examples

I recently published two videos with Google Data Studio examples of an organic traffic dashboard and an eCommerce dashboard. These videos show all the main features of the Google Data Studio to build any type of report.

Google Data Studio Examples

These Google Data Studio examples use Google Analytics as a data source and demonstrate how to add scorecards and backgrounds, apply styling, apply a heatmap to a column in your tables, and many other features of Google Data Studio to help you start building your own dashboard

Google Data Studio Examples – Organic Traffic Dashboard

This dashboard is an example of a monthly report on Organic search performance with a fixed date range of last month.

Google Data Studio Examples – eCommerce Dashboard

This dashboard is an example of an eCommerce performance dashboard with a data control option to let your client or manager select their own desired date range.

If you are just starting with Google Data Studio, I have written a detailed Google Data Studio Guide to get you started.

Line Chart Examples For Data Storytelling

These line chart examples show how to use annotations to not only show a trend whether in traffic, conversions or user behavior on your website, but to tell the full story about performance with your data.

I use line charts a lot in my reports since they are one of the best way to report and visualize trends, whether it is a trend in traffic, conversions or user behavior on your website.

When the trend is consistently showing an increase or decrease or staying stale it might be a little easier to tell a story with your data with a simple line chart and couple of comments. But when a trend is fluctuating as a result of marketing activity a chart line might look confusing and the data story difficult to understand.

Below is a line chart showing the trend for social media traffic for 2017 for a certain website:

Line Chart Data Storytelling

Traffic from social media is heavily dependent on social media campaigns activity and trend has been fluctuating throughout the year.
In the line chart example I have simply visualized the numbers for traffic from Social media each month in 2017.

Upon exploring I understand that these fluctuation are resulting from advertising activity as well as success of the specific campaign.

In order to fully tell a story with the data and explain traffic increases and sharp drops an analyst might want to include annotation to explain each significant fluctuation in the trend.

Line Chart Examples With Annotations

Here is a line chart example of how I improved the visualization and insert explanation of why we are seeing big increases and decreases.

Annotated Line Chart Examples Data Storytelling new

It might make more sense than the first example however after looking at it I thought that the background color was not a good idea. I also thought there are options to improve the annotations and make the main takeaways more visible at first glance.

So I made a few changes:

Annotated Line Chart Examples Data Storytelling FINAL

I changed the background to white and did some styling to the annotations text. I also decided to remove the data points and the data labels. All this is to try and draw the attention to the traffic fluctuations and the main reasons for these fluctuations which is what I am trying tell with this data.

Create a Combo Chart in Google Data Studio

*Disclaimer: For this video I have used data from the Google Merchandise Store Google Analytics which is publicly available.

A combination chart visualizes data for several metrics to help you compare. In this video I show you how to create a combo chart in Google Data Studio. I am creating a trend chart for sessions, unique purchases, product adds to cart and revenue for a period of 13 months. In this video I also show how to style a chart and choose different colors for each metric.

If you are just starting with Google Data Studio, I have written a detailed Google Data Studio Guide to get you started.

How to Create Prettier Reports in Excel

While it is the data in any Excel reporting that matters, good styling, colorful charts and graphs will turn a report from raw data to a compelling story and be easier to digest and understand.

Here are a few basic tips to help you create prettier reports in Excel and tell a story with your data

How to make pretty excel reports - Finished report

Styling Excel Tables

1. Start your data table in B2

Tables in Excel are not very easy to digest. Data is best understood when visualized with graphs and chart. However there are data numbers that need to be presented in a table.
Moving the table to start from B2 gives a nice spacing between the top and left margins of your report. This spacing simply makes it easier to look at a table of data and easier to read and understand the data

2. Branding and Report Title
It might also make sense to start even further down in column B and insert logos of your agency, client or brand at the top of each sheet on your report. Underneath this branding fill one of the rows with a suitable branding color by selecting the entire row and then fill it with an appropriate color. You can begin your table of data one row below this filled row.
You can also insert report title and date.

3. Always use table formatting
Tables with data are easier to read if formatted properly. Excel has a great table formatting feature with some really good built in table formatting templates. You can also build your own table formatting style to use or you can manually format your tables. When formatting tables try to use a header row, lines between each row and rotate colors between rows.

4. Remove background noise by deleting the gridlines from the sheet. You can do this by going to the View tab and untick the Gridlines option. This makes the sheet background white without any distraction making it easy to focus on the data.

Styling Excel Charts

5. Align axis titles on the chart horizontally
One of the things I hate the most about Excel charts is when I have to tilt my head to read the axis titles.
There are many ways to avoid this. You can change the chart type or stretch the chart to fit in the titles.

6. Delete gridlines and legends with one data series
In order to tell a story with your data the reports need to be clean of any background noise or anything not useful on the charts. I personally love charts and graphs that have no gridlines. Gridlines might be useful with a line chart but any other chart is better without them. It is very easy to delete gridlines, simply click on one of the lines to select and hit delete.
Also if your chart has only one data series it is no use to keep a legend of it. Delete it to focus on the important story the chart has to tell.

Pretty Excel Reports - Don't tilt

vs this:

Pretty Excel Reports - don't tilt 1

7. Sort your data before charting
Make it easier to understand what data is saying by sorting largest to smallest numbers before charting your table.
It is way easier to visualize how numbers in your data relate if they are sorted beforehand.

Pretty Excel Reports - Sort before charting

vs this:

Pretty Excel Reports - Sort before charting 1

8. Color your tabs
If you have a report with data in several tabs you have the option to give colors to your tabs. You can choose a color by right clicking on the tab and choose “tab color” option and select the color of you choice.

Pretty Excel Reports - Color tabs

Recreate Moz Stacked Bar Chart with Google Data Studio and Google Sheets

In this video I show you how to use Google Sheets as a data source in Google Data Studio. Using data that has been saved in a Google Sheet I build a stacked bar chart in Google Data Studio to show a month by month trend of organic rankings and visibility. I have taken the data from Moz from one of my campaigns that I am running. The stacked bar chart that I am building in Google Data Studio is the same as the stacked bar chart that Moz uses to show ranking and visibility trend.
Continue reading “Recreate Moz Stacked Bar Chart with Google Data Studio and Google Sheets”

Create Dynamic Charts with Combo Box in Excel

*Disclaimer: For this video I have used data from the Google Merchandise Store Google Analytics which is publicly available.

Combo Boxes in Excel create drop down menu options and dynamically populate tables and charts.

In this video I will show how to create a dynamic mini dashboard in Excel with the Combo Box. The drop down menu will give us the option to select All Devices, Desktop, Mobile and Tablet and the table and charts will dynamically change to reflect the number of sessions and the revenue made for each device or all devices. The date range is January 2017 to June 2017 and the data is broken down by month.

The most important part of creating a combo box is to properly configure the input range and cell link (demonstrated in the video). The other important part if to understand the Index function in Excel. For the combo box the Index function contains an array, which includes the column or certain rows in a column, and a row_num which would be the cell we have selected as a Cell Link. All of these are demonstrated in the video.

Styling and designing is just to give ideas and show you options but you can style it in any other way that suits your reporting.

How to Create a Heatmap in Excel

*Disclaimer: For this excel heatmap video I have used data from the Google Merchandise Store Google Analytics which is publicly available.

In this video I show how you can create a heatmap chart in Excel. The excel heatmap will highlight the most and least popular times of day and days of week when users make purchases on the online store. You can use it in any way to suit your data and your reporting needs. You can use it to look at marketing channel and location or conversion and browser technology, just to name a few ideas.

Continue reading “How to Create a Heatmap in Excel”

Build an H1 (DOM Element) Variable in Google Tag Manager

There are many examples where having this type of data, the text of the H1, available will make auto events and pageview tracking easier and cleaner to report. By implementing an H1 variable I was able to simplify reporting and present a much cleaner report using the H1 test on a page instead of filtering through a large number of URLs.

H1 Variable in Google Tag Manager

Building a variable with the H1 test is actually pretty easy to do. You need to create a DOM Element variable with the following settings:
Variable name: H1
Selection method: CSS Selector
Element selector: H1

H1 Variable in Google Tag Manager

Ideally only one H1 should be present on a page. If you have multiple H1s Google Tag Manager will pick up the H1 header that appears first in the code of the page.

How to Use H1 Variable for Auto Events and Reporting

Here is how I set up an event for a client to report on the most popular service page using the H1 on the page. There are a number of ways to report on the most popular product or solution or blog post but except you are using the enhanced eCommerce that will report using the actual product name, you will use the page URL in your reports and not the name, H1, of the page.

After building the variable, build a new trigger with the following settings:
Trigger name: {Page Name} Pageview
Trigger type: Pageview
Fire on: Page URL contains {URL identifier}

H1 Variable in Google Tag Manager

Then build your auto event tag with the following settings:
Tag name: Most popular product
Tag type: Universal Analytics
Track type: Event
Event category: Most Popular Product
Event action: {{H1}}

H1 Variable in Google Tag Manager

Then select the trigger and you are all set!

You should be able to see your report under Events. Drilling down to the event action will report on the page H1 and the number of pageviews it has received for a selected period of time.