Creating an Excel Dashboard Widget

Creating a dashboard in Excel can be quite simple with Analytics Edge. It’s really just a matter of downloading the right data and some easy formatting techniques.

creating-a-dashboard-widget

Before you begin, make sure to plan out your whole report. There is nothing worse than spending a lot of time preparing a dashboard only to discover that the data you need to show at the bottom won’t fit in the layout you’ve used at the top of the page. Remember that Excel uses rows and columns, and the row height and column width are constant across the entire page (see Building a Marketing Dashboard for other considerations).

creating-a-dashboard-widget-layout

The Widget Details

The classic dashboard widget includes a main metric, sometimes referred to as a key performance indicator (KPI). Since context is critical for any report, the number is usually presented with a comparison to the previous period; in this case, we show both the previous value as well as a percentage change from the previous period to the current period. To provide even more context, a small annual trend curve (sparkline) is shown with the percentage change from the same period in the previous year.

creating-a-dashboard-widget-details

The Data Behind the Widget

Reviewing each element in the widget, we see that we need the following data:

  • value for the current period (top number displayed)
  • value for the previous period (displayed as a number and used to calculate the % change)
  • values for each of the periods over the past year (shown as an annual trend line)
  • value for the same period in the previous year (used to calculate the % change)

While we could grab each piece of data in a separate query, you may realize that a single query can get everything you need — an annual query + 1 reporting period (as shown below). For example, if you wanted a monthly report, pulling monthly values for the last 13 months (1 year + 1 month) would get you everything you need:

  • the current period (most recent period)
  • the previous period (next one up the list)
  • an annual trend (the last 12 months of numbers)
  • last year’s number (the oldest value in the list)

creating-a-dashboard-widget-data

Building the Query in Analytics Edge

First, I strongly recommend that you place all query results on a Data worksheet and use Excel cell references to your dashboard worksheet. This provides some flexibility later when things change (which they always do), and will save you from tons of work caused by doing something as simple as inserting a row or column.

When building queries with time dimensions, like date, week or month, be careful which dimensions you chose. Using month as an example, you can use the ‘Month of the year’, which is just a 2-digit month number. The problem with this dimension is that it will not appear in the proper time sequence when sorted ( Jan 2016 [01] would sort before Dec 2015 [12]). To avoid this problem, you could:

  • include the Year dimension in the query, sorted by Year then by Month
  • use the combination dimension ‘Month of Year’
  • use the normalized ‘Month Index’ dimension

My preference is the combination dimension ‘Month of Year’, because it is easier to sort and provides instant validation about what time period the data represents.

Note: Although not needed in this widget example, with the Analytics Edge Core Add-in, you can use the Convert function to convert this text column into Excel dates using the date format ‘yyyyMM’.

creating-a-dashboard-widget-date-conversion

Formulas and Charts

All that is left is to tie the numbers back to the widget using some simple Excel formulas. The current and previous period measures are simple Excel cell references. The percentage change calculations follow the formula of =(current value-previous value)/(previous value). If you have Excel 2010 or higher, you can use a simple sparkline chart for the trend curve.

With Excel 2007, you will need to shoehorn a standard line chart into the tiny space by turning off all the axes, legend and title and playing with the plot area to make it the same as the chart area.

creating-a-dashboard-widget-formulas1

Presentation Tricks

Of course, you can’t just present plain old black numbers on a white grid. This is where a little bit of Excel formatting ‘magic’ comes in.  Here are the things I typically do:

creating-a-dashboard-widget-formatting

  • first, turn off the gridlines to give the dashboard a cleaner look
  • use simple Excel Cell Styles to change title background and font colors, or choose your own to match your corporate color palette
  • make the current period number stand out (larger, bold) and de-emphasize the other numbers (smaller, light gray)
  • put a border around your widget to visually group all the elements as related items
  • add conditional formatting on the percentage change numbers for visual reinforcement of the direction of change (note: change the thresholds to above/near/below zero)
  • add high and low points to the sparkline charts to make it easier to determine where they might be on those tiny lines

Formatting your dashboard can be extremely time consuming if you let it.  It is easier/faster to do all of the formatting at once, so get all your numbers in place first, then format the whole worksheet. I have found over the years that it helps to review your draft report unformatted if possible (or at least all in black and white) to avoid wasting time discussing colors and font sizes, and keep the discussion on the metrics themselves.

Leave a Reply

Your email address will not be published. Required fields are marked *