Using some recent data from Comscore, we have created a dashboard examining the year-over-year change in traffic at the world's largest websites. The data for Average Daily Visitors tells an interesting story: although Yahoo clearly trumps Microsoft for daily visits in the US, worldwide Microsoft is ahead of Yahoo but still far behind Google, the undisputed champion. Explore the different metrics and find your own angle on this internet street fight.


Although one may assume that Google trumps both Yahoo and Microsoft in all metrics, there are some surprising upsets in this data. If you select Average Minutes per Visit in the filter on the left, the worldwide data shows Microsoft in the lead and Google bringing up the rear. Perhaps Microsoft is losing the search engine battle, but winning the content battle overseas? Examine this viz and make your own conclusions!


That's cool. How are you setting a filter on METRICS? I've been wanting to do that for ages, but can't see a way to do it in Tableau Pro 5...

Thanks! Whatever Dimension or Measure you want to filter in Tableau, bring it onto the filter shelf on the left hand side of the worksheet. Then, click on the drop down menu on the right hand side of the field you dropped onto the shelf and select Quick Filter. Feel free to check out Turning on Quick Filters for more color.

The original dataset had only one measure value, 'Value', cut by the various combination of dimensions (Month, Property, Metric, etc.)

For the dropdown menu you need to do three things:

1. Create a separate measure value for every metric via a CASE statement. Below is the statement to create the Total Unique Visitors metric:

CASE TRIM([Metric])
WHEN 'Total Unique Visitors (000)' THEN [Value]

2. Create a new measure value which uses a CASE statement on the Metric dimension to pull in the calculated measure value for each separate metric:

CASE [Metric]
WHEN 'Percent Reach' THEN [Percent Reach]
WHEN 'Average Daily Visitors' THEN [Average Daily Visitors]
WHEN 'Average Minutes per Visit' THEN [Average Minutes per Visit]
WHEN 'Average Visits per Visitor' THEN [Average Visits per Visitor]
WHEN 'Total Minutes' THEN [Total Minutes]
WHEN 'Total Pages Viewed' THEN [Total Pages Viewed]
WHEN 'Total Unique Visitors' THEN [Total Unique Visitors]
WHEN 'Total Visits' THEN [Total Visits]

3. Use the measure value from step 2 in your view, and place the Metric dimension on your filters shelf. In the dashboard view this filter is set to Compact List to get the dropdown functionality.

I've attached an example that hopefully is clearer than my explanation.

Thanks all for your input. Especially thanks to you Mark - I misunderstood the question!


In this case, the filter you have labeled, "Select a metric" appears to be actually a calculated dimension based on the Measure Names field, or else a calculated dimension other dimensions. Is either one of these the case? Then, it looks like you used the Compact List style for the quick filter?

Peace and All Good!
Michael W Cristiani
Market Intelligence Group, LLC

Great responses - thanks Mark. This workaround had been mentioned in the forums. The downside is that it multiplies the size of your dataset by the number of dimensions you want to filter on. However, it makes for a really nice trick, and maybe I'll use it for some of our smaller datasets.