Tableau Embedded Analytics: 5 Keys to Implementation Success

Tableau embedded analytics dashboards user experience

A poor visual design will almost certainly sink any embedded analytics solution, even if everything else is perfect. There are numerous ways to muddle dashboards to the point where your tableau embedded analytics put users off.

Solution:

Complexity and clutter are two of the most powerful adversaries of effective data visualization design. Everyone wants to cram as much information as possible into their dashboard in order to avoid leaving anything out:

• put the most important parts the largest, Highlight them, put them front and center

• remove everything that isn’t the most important insight

Establish common language in your Tableau embedded analytics

One of the first steps toward implementing successful Tableau embedded analytics is to understand how the business describes its work, its objectives, and the metrics it uses to measure it. Bear in mind that even within the same company, the terms used to describe the same thing can vary.

Solution:

Conduct interviews with business executives, managers, and front-line salespeople to establish a naming convention for the terms that will be used to describe specific metrics.

When labeling and titling your Tableau embedded analytics dashboards and charts, use their language, not yours. Then, as you demonstrate the Tableau Embedded Analytics dashboard to them, elicit their interpretations of the words and titles. Always look for better words to convert charts meaning better.

What do users want to see on your tableau embedded analytics?

Your users can not be overwhelmed with data. They need to understand how data can assist them in performing their jobs more effectively. They want to know how they can make a tangible difference. You need to understand the context in which our customers will be viewing the data. With Tableau embedded analytics, the level of sophistication is critical.

Solution:

Maintain a low level of sophistication when demonstrating Tableau embedded analytics to end-users. Provide them with only the information they require to take action or make a strategic decision.

Allow a higher level of sophistication only for analytics for more advanced users, who will be performing data exploration work. Those users want to sift through the data, identify the insights, and verify their validity. Provide them with filters, parameters, and a large amount of data.

Match the current processes and mental models your users

Always keep in mind your users are accustomed to seeing data within the various systems they use, like a marketing automation system or CRM. They're so accustomed to looking at those that they can become reluctant to adopt Tableau embedded analytics just because it looks different.

Solution:

Deploy Tableau embedded analytics consistent with how people are accustomed to consuming data. Mimicking how they are accustomed to viewing and processing data reduces their mental friction.

Plan authentication integration in advance so that the user is not required to log in to yet another system in order to access additional data. This significantly reduces the technical friction associated with your customers' experience.

Understand the behavior you want to change with the embedded analytics

If you want to convince skeptics who are accustomed to making decisions based on gut instinct, you must quantify the success of analytics. Do not be afraid of financial targets.

Solution:

Define the behavior you wish to modify or promote. Connect positive behavior to a positive outcome for the business to gain a better understanding of the business's overall value.

When possible, connect that altered behavior to the amount of revenue or other critical metrics that it will eventually generate. Insights integrated into people's workflows are acted on quickly and easily, resulting in the desired results.

Select the key data visualizations you will be showing to the users

The traditional IT project management approach, where IT and business leaders meet to define requirements is not efficient for highly visual reporting. This way you can spend weeks trying to transform the business needs into a usable report.

Instead, launch prototypes to drive a fast-paced test-and-learn lean process. Focus on the most important business value that business users want to convey in the data. Let them show you how they want to use the Tableau dashboard.

Only when business users get a chance to use and understand how the analysis works you will uncover all their needs. Indeed, people often can't articulate what they want until they see it.

Identify the key data sources you will be accessing

Data is everywhere. An analyst’s intuition and a subject matter expert’s input will guide you to the right data from the right sources to address and solve your business problem.

Meet with the business users and IT departments to clarify the key source data and plan how to connect to data sources. The most common examples of these are databases, flat files (i.e. spreadsheets), databases (i.e. PostgreSQL, SQL Server), and cloud systems (i.e. AWS Redshift, Google BigQuery). This will allow you to better understand which users own data and plan how to access it in a secure way.

Once the business has defined its data requirements, you can focus on delivering high-quality data while maintaining the highest quality of data security.

Decide where users will be accessing the dashboards

Understand who your target users are as well as their roles and responsibilities. Discuss embedded analytical solutions, accessing the dashboards and usability requirements like toggle sub-navigation.

This will allow you to better plan the embedded solution. We define the security level for each user (or group) and the trusted authentication method, i.e. enable single sign-on.

Applications contain analytics in one or more of the ways shown below. An embedded analytics application can start at any stage in the diagram and progress over time.

Prepare interactive visualizations using Tableau editor

Interactive data visualization is the process of creating a visual representation of data that can be explored and analyzed directly within the Tableau editor.

Take advantage of the fact that Tableau analytics is designed to deliver interactive visualizations to your users and, if possible, distribute the data across multiple visualizations. Consider your design carefully—the fewer sheets and data sources you use, the faster your visualization will perform.