Visualizing Customer Journey with Sankey Diagram

Struggling to gain useful insights from your data? Perhaps you just need to see at it in a different way.

Many traditional tools for path analysis can be misleading or downright confusing, and so analyzing your customer data can be a frustrating process.

With a Sankey diagram, PPC advertisers and data analysts can quickly find insights and will find it easier to present their findings to business leadership.

Once you know how to use these diagrams, you will be much better at visualizing the customer journey in your ad campaigns.

A Sankey diagram is a powerful visualization that provides an overview of the flows in a system, such as energy or materials, or in advertising, the customer journey.

As each flow has a varying height depending on its quantity, analysts can easily determine the most prominent contributions. This type of segmentation analysis makes it easy to draw conclusions from data.

This system is similar to an alluvial diagram, although the latter typically visualizes change over time, rather than depicting changes in several different variables.

While the Sankey diagram is attributed to Irish Captain, Matthew Sankey, an earlier version has arguably achieved more acclaim. In 1869, Charles Minard created a flow map to analyze Napoleon’s Russian Campaign of 1812. This was almost 30 years before Sankey’s first diagram came to light in 1898, which suggests the possibility that Minard’s map may have been an inspiration for Matthew Sankey.

Regardless of who did it first, modern marketers have discovered that this form of visualization is a useful tool in advertising and data analysis.

In the current digital landscape, data analytics is an integral component in marketing success. But collecting the data on your customers’ behavior isn’t enough on its own – you need to be able to analyze it to produce actionable takeaways.

Consider the data below:

There are a lot of better ways to visualize tabular data. By using a Sankey flow diagram, marketers can perform a much easier analysis, forming a strong basis for smarter decisions.

Here are some great benefits of this system for PPC advertisers:

The minimum requirements to create a Sankey diagram aren’t much at all. You just need:

• Two Dimensions
• One Metric

This type of path analysis clearly shows how different values transition from one category set to another. In advertising, knowing where your traffic is coming from is extremely useful information.

The ability of a Sankey chart to visualize user flows from one dimension to the next is very powerful. CEOs and CMOs will be able to get a good overview of how their ads are performing without having to spend a long time trawling through data.

Many users of Google Analytics are already familiar with the concept of a Sankey diagram. However, before now, using this type of segmentation analysis in PPC advertising simply wasn’t possible.

PPCexpo is the first company to make it possible, allowing users to visualize the customer journey and quickly find insights through their visualization library, ChartExpo.

Now that you know the benefits of using a Sankey diagram in advertising, let’s explain how to do use it.

There are two main aspects to a Sankey flow diagram.

The nodes in this type of flow diagram are essentially like a bar in a classic bar chart. The height of the bar represents the value.

The link is the connection between two nodes. This indicates the flow. In the figure below, you can see how the link color matches the node color.

Using a Sankey diagram for PPC advertising is one of the better ways to visualize the customer journey, as the segmentation analysis allows you to quickly find insights on which keywords are driving traffic, and also where that traffic is coming from.

In this visualization we’re showing three dimensions and one metric as follows:

• Left – Match Type
• Middle – Impressions
• Right – Device Type

 

Going from left to right, the data is laid out in sequence. First, a match happens between a search-query and a keyword, followed by an impression (or not), which happens on a specific device. The visualization is bidirectional, which means you can “read” the chart from left to right, or right to left.

The metric is impressions.

Google Ads reports Impressions as a percentage. The number in the brackets is a calculated value, which is the actual number of impressions.

In the middle is the impressions dimension. The impression share is 73.7%, which means this campaign missed out on 26.3% of the “opportunity”. The campaign manager can take steps to capture the remaining 26.3%.

Impressions percentages are not additive, and if you try adding them, you will end up with values greater than 100%.

Let’s look at the match types, where we have 85.4% exact match and 47.0% broad match. Adding them will exceed 100%, and that’s before you even include the other match types.

The reason for this is because each percentage is independent of the other values in that category.

What that means is that out of all the exact match keywords in your ad campaign, 85.4% matched the user’s search queries. Similarly, out of all the broad match keywords in your campaign, 47% of them matched the search queries.

On the right-hand side, the percentages for the devices are also independent of each other.

Having ‘63.7% on Desktop’ means that out of all the possible impressions you could have got on ‘Desktop’, your keywords matched a search query and earned an impression 63.7% of the time.

The Sankey diagram is a great tool for visualizing the customer journey in PPC advertising, but it is important to limit your analysis to single sessions. If you overload the chart with too much data, it can quickly become unreadable.

By focusing on a single ad, you can perform analysis to discover where your traffic is coming from, and which keywords and match types are working best for your campaign.

Customer journey is the next big thing. If you can measure it good, you can optimize it much better. Omnichannel customer journey would help brands to seamlessly approach audience on different channels with personalized brand messaging and to create a unified customer experience across channels.