I’ve been an avid user of Tableau for years. I pretty much can’t solve a quantitative problem these days without using Tableau to help me visually explore my data and iterate through ideas and hypotheses. But, some problems require more heavy lifting in Tableau than a viz can handle simply. Today, we will discuss Tableau Clustering and why it is useful in creating better analysis of data.
What is Tableau Clustering?
Tableau has recently begun adding more statistical tools that provide powerful ways of visualizing and exploring data. Tableau clustering is one of the newest features in Tableau 10. It puts advanced statistics into your hands with just a few clicks.
Tableau Clustering allows you to easily identify statistically similar groups. In plain English, based on attributes you tell Tableau, it will go through and determine similarities and create look-a-like groups. You can then drill into those for more detail or compare how each group behaves relative to each other.
Like we discussed above, the ability to segment data into useful groups or bins is as important as ranking and identifying your top and bottom values. It’s a must for any data analyst. Tableau Clustering takes that ability to a whole new level. You don’t need code or need to be a trained statistician to access it.
Tableau Clustering excels at visually seeing the relationships between data. For instance, we might wonder, “How do these 6 things interact together and what results do they produce?” What if we wanted to add Measures instead of Dimensions? For example, purchase patterns (Sales) and amount we actually make (Profit) and return or discount patterns (Discount, Returns).
Tableau clustering allows us to add this additional information. This helps us move beyond simple segments to advanced, incorporating data on behavior patterns and actions (Measures), as well as attribute information like Region or Marketing Channel (Dimensions).
Why is Tableau Clustering Useful?
Getting better insights faster enables us to take more action. Being able to take action that makes an impact makes you a hero; it makes you the person with all the answers. That’s an awesome feeling and it’s what Tableau clustering enables us to achieve. The ability to find hidden insights with Tableau’s easy drag and drop functionality is a major step in getting to action faster.
Tableau Clustering Examples
Here are some example vizzes of how people have used Tableau clustering to create segments and find insights they couldn’t get to easily:
Marketing pro Chris Penn used the Tableau clustering tool to find insights about his own blog that were obscured with traditional methods of visualization. Namely, drilling into what topics of social media posts drove new users, large number of reshares, or were stagnant:
Chris Wood gives an insightful interactive analysis of at risk youth in the Washington, D.C. school district, also explaining how he used Tableau clustering to do so.
Tableau Clustering Uses
Check out more Tableau clustering applications at work below.
1. Customer segmentation
Say you have a group of customers that logs in very infrequently, never calls support, started with low monthly recurring revenue, but spent tons on upgrades over time. That’s an odd group with tremendous organic growth and low costs, even though initial revenues were low. Tableau clustering can find groups like this.
2. Market research
How do we determine different groups in the market and create products and marketing messages that resonate with those people? For example, a bank found a group of entrepreneurs that was using equity from their homes via a 2nd mortgage to fund their startups. Knowing that led to a whole new line of products for the bank that resonated much stronger with that group.
3. Customer surveys
What Tableau clusterings crop up among satisfied customers, what clusters crop up among unsatisfied customers? Are the unsatisfied customers also utilizing your excellent support services?
4. Matching or recommendation algorithms
Netflix: For example, based on movies that have a Strong Female Protagonist, Witty Humor, and British Actors, we recommend all movies based on every Jane Austen book ever.
5. Telecom
Position the cell towers so that all customers receive optimal signal strength based on addresses, usage patterns, roaming, subscriptions, peak times, traffic patterns and roads, etc.
6. Scheduling
Say you’re a police chief trying to maximize your officer time with limited budget. You need to schedule patrols at peak times in the most crime-likely areas, again based on any number of factors, like time of day, weather, income and education levels, past crime events, types of crime, known gang locations, etc.
I personally use Tableau clustering all the time in my daily analytics work and I find that it has unrivaled abilities in telling a story about groups of data. Stay tuned for part 2 where I cover how to create your own Tableau clustering charts.
How to Create a Tableau Clustering
Let’s jump in and create a Tableau clustering chart from the Superstore Dataset that shows the relationship between sales and profits with highlighting of other fields such as marketing channel or product category. We start with a view with these fields pulled out:
A. Tableau clustering creation
- First, click ‘Show Me’ at the top right and choose the ‘Scatter Plot’ option to get this into more of a useful format. Then, you’ll see that Marketing Channel and Region are on the Shapes and Color shelves, respectively.
- After that, set it to ‘Entire View’ from the drop down menu at the top.
- Then, let’s add several more ‘Dimensions’. Add Product Category, Customer Segment, and Product Subcategory to the Detail shelf.
- Subsequently, click on the ‘Shapes’ card to set each of the marks to Filled from the drop down menu labeled ‘Select Shape Palette.’ Then, choose Assign Palette and click Ok
- Next, click on the Analytics tab at the top left, above your Dimensions.
- The next step in Tableau clustering is clicking Cluster and dragging it out. Be sure to place it on top of the Cluster box that appears.
- Also, notice that 2 clusters are generated automatically from the data.
B. Play with potential Tableau clusterings
- Let’s play with the number of potential Tableau clusterings. Change the number from Automatic to 5. Then, you should see the different colors.
- Go over to top right where the data highlighter shows the different Tableau clusterings. Click each one in succession to highlight that segment on the scatter plot. Are you seeing some interesting groups, like a group of high sales, low profit?
- Click on the down arrow on each pill that you put on the Detail shelf and select “Show Highlighter.”
- These should appear on the right-hand side. Click through these to see if there are any interesting insights that emerge. For example, under the Marketing Channel highlighter, choosing “SEO” or “Social Media” reveals some interesting insights. Or choosing “Google Adwords” reveals an interesting outlier.
With this advanced Tableau clustering chart created, stay tuned for part 3 where I cover how to interpret, explain, and visually fine-tune Tableau clustering charts.
C. Tableau clustering visual fine tuning
We can easily build an awesome Tableau clustering chart, but there is some visual fine-tuning to do before we can hit a home-run with our boss. Therefore, we have created an easy guide to Tableau clustering so data interpretation and explanation becomes easier to do. Before we jump in, let’s take a look at some of the numbers under the hood.
D. Tableau clustering interpretation and analysis
How do I understand each of my Tableau clusterings beyond just eyeballing it?
- First, click the down arrow on the Clusters pill which should be on your Color shelf.
- Then, choose Describe Clusters.
- Lastly, a window will appear with a lot of information about how this was created. You want to pay attention to the following:
E. Tableau clustering variables
- Variables of Tableau clustering – these are the measures that you are crunching to find look-a-likes (i.e. group similar customers by sales and profit)
- Level of detail – these are the dimensions that you’re incorporating into the Tableau clustering (i.e. show me look-a-like customers by sales and profit, by analyzing customer segment, marketing channel, product category, etc. and finding commonalities across all of those).
- Number of clusters – these are the distinct groups or segments that the algorithm found
- Clusters – you need to scroll down to find these.
- Number of Items – shows how many data points are in each cluster (these could be your bars or the circles on a scatter plot)
- Centers – this is the average value within each cluster. You’ll see the obvious differences.
- It’s OK to have Tableau Clustering of different sizes as data may group more strongly at one end then another, but you want each cluster to have enough data points to be meaningful.
- If it only has one or two, consider excluding those from the view as they might be outliers skewing your results, or consider changing the number of clusters.
- Note: Most of the cluster centers will appear in scientific notation, which is frustrating. If you click the Copy to Clipboard button and paste it into Excel, you can format the numbers so you know correctly what they represent.
F. Cleaning up Tableau clusterings
Now, Let’s clean the clusters up with a trick to rename them with the added bonus of being able to use them in other charts and analyses. (Note that once you complete this step you cannot view the previous underlying numbers, so make sure you have copied the numbers or taken a screenshot.) This is the final product:
-
- Hold down the Ctrl key and then click on the Clusters pill on the color shelf, and then drag this over into Dimensions.
- Now, double click the Clusters pill you just dragged into Dimensions and rename it to “Sales & Profit Clusters.” This is now a field that we can reuse again later, which will be very helpful in analyzing certain segments of customers.
- Click the down arrow on the renamed pill and choose Edit Group.
- Right click on Cluster 1 and choose Rename. Type “Low Sales, Low Profit.”
- Follow the same procedure for Cluster 2 (note that they may not be in numeric order!). Rename is to “High Sales, Low Profit.”
- Rename Cluster 3 to “Top Performers.”
- Then, rename Cluster 4 to “Mid-tier Sales, Low Profit.”
- Also, rename Cluster 5 to “Medium Sales, Medium Profit.”
- Now drag the updated “Sales and Profit Clusters” pill and replace the existing Clusters field on the color shelf. You can do this by placing this pill directly on top of the other one. Or, by dragging the current field on Color off and replacing it with the new. Follow along with the GIF below to see it completed up to this point (Click to see it full screen):
G. Changing colors in Tableau Clustering
Now, let’s change the color scheme in Tableau clusterings, so that our colors convey a little more meaning.
- On the legend, click the drop down arrow at the top right, and choose Edit Colors.
- Set the color palette to Superfishel Stone in the drop down menu.
- Now choose the “Top Performers” segment and the click on the dark green pill.
- Repeat this procedure and change “Low Sales, Low Profit” to the orange color. Change “High Sales, Low Profit” to red. Change “Mid-tier Sales…” to the light olive color. Change “Medium Sales” to the aqua color.
- Choose OK.
We now have some statistically valid segments that we can reuse and that are highlighted with meaningful titles that indicate the next step. For example, “High Sales, Low Profits” leads us to the very obvious “why” question. We can then drill down deeper to see what else surfaces from these data points that indicate actions we need to take.
How do I explain Tableau clustering to other people…?
…and get the “thumbs up” from your boss?
Use the following tips:
Explain Tableau Clustering In English
Find members of a potential group (could be customers, could be cities, could be anything you’re trying to group on) that are as similar to each other as possible, and as dis-similar as possible to the next group. We want each group to be as unique and distinct as possible, while we want each member of a particular group to be as similar as possible.
Explain Tableau Clustering Quantitatively
For a given number of clusters or look-a-like groups (denoted by the letter “K”), the algorithm partitions the data into that many clusters or groups. The algorithm will determine what it thinks is the optimal number of clusters for you, based on your data. But you can easily change that to see if new patterns emerge. Each Tableau Cluster has a center (centroid) that is the average value of all the points in that cluster. Each cluster is a valid statistical grouping that will update dynamically as data values change or as new data is added.
Share an Example of Tableau Clustering
Let’s say you have information about four Domino’s pizza chains, and a list of customer addresses. But those customer addresses aren’t tied to any particular Domino’s location. You’d have to manually sort through the addresses and compare them on Google Maps to determine which location they should order from. Tableau clustering does this automatically. It would crunch through the data and then determine which neighborhoods are around each Domino’s location. You’d have four clusters. This is essentially what Google does when you search for “pizza near me,” by the way.
What have you used Tableau clustering for and how did it allow you to find more insight? Let us know in the comments! To learn more about Tableau, check out our course offerings.