Ok, you had a great time at TC19 and you just can’t stop raving about it with your co-workers and they watch to watch some of that goodness. Looks like, you also missed some sessions because there was too much goodness happening at the same time.
Back home, ready to watch, but not sure where to start?
If you are trying to find what zipcodes are the ones with highest volume of sales, would the following image provide you the info
OR, would this one?
Both options show the same amount of marks, which is about 40K data points
It turns out that creating polygons to draw the zipcode boundaries is a bit more intensive than creating circles. I am not exactly sure if this will matter for all of you, but its a point worth noting.
My advise for this one would be to limit “filled maps” for county/zipcode level maps in the cases where you want to highlight density but use “density maps” instead.
However, when you are looking at maps at the State/Region/Country level, this might be a moot point
Ok, this one seems obvious. However, since I am writing about it, you can guess, that it’s not.
The guidance I would like to give you is to utilize the Tableau provided connectors in cases where it exists. For some of you data-savvy folks, you might have a very good understanding of ODBC connectors and in many cases, you might already have ODBC setup on your machines to connect.
We highly recommend utilizing the option that Tableau provides when you click on the name of the vendor
So, in the example of Oracle, we would want you to use the “Oracle” option and NOT THE “ODBC” option
The number of filters in a dashboard can have a direct impact on your dashboard performance.
Think of each filter as another query for your dashboard. Let’s take the example of our dashboard
In the above example, every user will easily understand that each sheet may potentially cause a separate query to the data source. However, what’s not so easy to understand is that each filter will also cause an (potentially) additional query as well. So, instead of running 6 queries, we are running 14 in the above example.
At the very least, we should try to reduce the number of queries by reducing the number of filters needed.
Remove unneeded filters
Remove filters that are already in another sheet and are easily clickable (in the example above, I may not remove country as a filter since its hard to find small countries on the map). However, I might remove Ship Mode and Product Sub categories as filters
You can also reduce queries by creating different types of filter. (Read more in the linked post)
There are so many things you can do by setting up the right set of filter conditions on a dashboard. Use Visual Filters Use filter conditions that make the smallest SQL (think include or exclude) Use hierarchical filters instead of relevant filters when possible Use Wildcard filters vs. enumerated lists Watch the video clip starting … Continue reading Tableau Dashboard Performance Series: Tip#5: Filter Conditions
I would minimize the number of sheets in each dashboard and ideally keep it to less than 5
Not sure how much more I can add, other than the fact that most of my dashboards are with sheets between 5 and 10. Having had many engagements providing the insipiration for these dashboards, I think my rule of thumb works pretty well
In today’s modern tech world, everyone expects more. Every one wants to be able to feel and touch their data. That is exactly what “Visual filtering” allows you to do.
you click on a data element and all the other reports are showing only contextual data.
Once your audience starts using visual filtering, the entire perspective on data changes.
Here’s an impact story:
One of my customers implemented a Snr leadership facing dashboard and they recently conducted a meeting where not only did they use the Tableau dashboard to provide info, they decided to actively participate in some slicing and dicing of the data on the dashboard using “visual filtering”
This is the dream
One of the other execs in that meeting upon seeing the power of visual analytics