Building great dashboards in Einstein Analytics is a combination of art and science. You can build pretty visualizations. If a user can’t take action from the dashboard, it won’t be used. If you throw tons of data at users and they can’t draw insight from it, the dashboard will be ignored.
Dashboards need to be a visual representation of the user’s business goals, with insights that are actionable. Both the art and the science are required to create a meaningful dashboard that engage users. Here are 5 steps to building killer Einstein Analytics dashboards that users will love.
The first step is all about your users. There is no development involved. It vital to understand the business goals of the people using the dashboard. It’s often helpful to build a persona for the user - who they are, what they do, their goals and frustrations.
When working towards business goals, users will often have questions they need to answer. For example, an Executive needs to meet revenue growth and customer satisfaction targets. Questions she needs to answer is how much revenue has closed, what is the forward looking pipeline and customer satisfaction scores.
From the questions, Key Performance Indicators (KPIs) are defined. The KPIs allow users to understand performance at a glance. For example, our customer satisfaction is 8.2 out of 10 and trending upwards.
An often overlooked step in KPI development is KPI prioritization. If you have 20 KPIs, you essentially have no KPIs. Human nature is to focus on just two or three of the metrics when presented with too many. Each user will pick a different set of KPIs to focus upon, which means the business is not moving in unison. It’s better to have 2-3 KPIs that everyone focuses on.
With KPIs defined, it’s time to start slapping a few charts on the dashboard, right? Not exactly, as a charts purpose is to support the story behind the KPI.
If the KPI is about Year to Date Revenue, a supporting chart could include the top accounts by revenue. A key takeaway from this chart is how much revenue is concentrated in a handful of accounts.
While it’s important charts support the story, picking the right visualization also matters. Consider the two examples below that show how much revenue comes from each customer. Which one is easier to see the largest accounts and their relative size to each other?
When working to compare data a Bar or Column chart works well. To display the composition of a metric, donuts and stacked bars/columns are appropriate. Trending is best illustrated with line or column charts.
As the story begins to unfold, don’t limit your dashboard to just the data at hand. Identify data that complements or enriches the analysis.
When considering the growth in revenue for an account, seeing the support case history and customer satisfaction unlocks another level of insight. The case data will not be stored with the opportunity records, but the information can be combined.
Within a dashboard, it is easy to link two datasets together. Use the Connect Data Sources item in the More menu. Then specify the two (or more) data sources and the fields used to connect them. The fields don’t even have to be named the same. Just be sure to connect the datasets on each attribute you use to facet or in a filter/selector.
What if your data does not reside in an Einstein Analytics dataset? There are options available to pull data into dashboard. Data in Salesforce can be queried in SOQL (Salesforce Object Query Language) and displayed in a visualization.
Data access can even be extending outside of Salesforce with an Apex step. In the case of our top accounts, code can be written to pull in quarterly earnings data to understand if the company is trending up and down.
When the story a dashboard tells drives a user to an insight, what should be done next? This is an area where Einstein Analytics shines. Rather than jumping from a external reporting tool back into a system of engagement like Salesforce, Einstein Analytics dashboards can link into the existing business processes.
It only takes a few clicks on the dataset to set up actions. Actions can perform operations such as Opening the Record in Salesforce, Creating a Task or Starting a new Opportunity. For data from an external source, it can be opened with the action framework as well by specifying a url.
For advanced business process integration, it’s possible to create custom actions that utilize code based actions. For example, create an action plan for key accounts with poor customer satisfaction.
As designers, we spend most of our time inside of Analytics Studio, but not our users. Users spend their time inside of Salesforce, either online or through the Mobile App. Dashboards need to be embedded as pages within the user’s apps, as part of the normal workflow.
There is a Wave Dashboard component that can be put on any Lightning Component. If use Visualforce or classic, dashboards can be embedded there as well. Using Lightning Out, the dashboard can be available outside of Salesforce entirely.
Beyond providing a view of the dashboard in Salesforce, give users contextual insights. When an Account is opened, show the dashboard filtered to just that Account. Everything a user needs to know, right when the page opens. This is done by adding a filter to link the values on the record to the analytics dataset.
Finally, don’t forget about your mobile users. Dashboard support multiple layouts. Build a long, narrow layout suitable for viewing on a mobile phone. Users will have what they need, no matter where they are.
Interested in seeing these five steps in action? You’ll have a chance to attend a live Webinar on Thursday, October 18 at 2:00 PM EDT. If you are not able to attend in person, register and watch the recording. Register at - http://sforce.co/2A21x21
Dashboards need to be a visual representation of the user’s business goals, with insights that are actionable. Both the art and the science are required to create a meaningful dashboard that engage users. Here are 5 steps to building killer Einstein Analytics dashboards that users will love.
1. Know Your Audience
The first step is all about your users. There is no development involved. It vital to understand the business goals of the people using the dashboard. It’s often helpful to build a persona for the user - who they are, what they do, their goals and frustrations.When working towards business goals, users will often have questions they need to answer. For example, an Executive needs to meet revenue growth and customer satisfaction targets. Questions she needs to answer is how much revenue has closed, what is the forward looking pipeline and customer satisfaction scores.
From the questions, Key Performance Indicators (KPIs) are defined. The KPIs allow users to understand performance at a glance. For example, our customer satisfaction is 8.2 out of 10 and trending upwards.
An often overlooked step in KPI development is KPI prioritization. If you have 20 KPIs, you essentially have no KPIs. Human nature is to focus on just two or three of the metrics when presented with too many. Each user will pick a different set of KPIs to focus upon, which means the business is not moving in unison. It’s better to have 2-3 KPIs that everyone focuses on.
2. Pick the Right Visualization
With KPIs defined, it’s time to start slapping a few charts on the dashboard, right? Not exactly, as a charts purpose is to support the story behind the KPI.
If the KPI is about Year to Date Revenue, a supporting chart could include the top accounts by revenue. A key takeaway from this chart is how much revenue is concentrated in a handful of accounts.
While it’s important charts support the story, picking the right visualization also matters. Consider the two examples below that show how much revenue comes from each customer. Which one is easier to see the largest accounts and their relative size to each other?
When working to compare data a Bar or Column chart works well. To display the composition of a metric, donuts and stacked bars/columns are appropriate. Trending is best illustrated with line or column charts.
3. Data Mashup
As the story begins to unfold, don’t limit your dashboard to just the data at hand. Identify data that complements or enriches the analysis.
When considering the growth in revenue for an account, seeing the support case history and customer satisfaction unlocks another level of insight. The case data will not be stored with the opportunity records, but the information can be combined.
Within a dashboard, it is easy to link two datasets together. Use the Connect Data Sources item in the More menu. Then specify the two (or more) data sources and the fields used to connect them. The fields don’t even have to be named the same. Just be sure to connect the datasets on each attribute you use to facet or in a filter/selector.
What if your data does not reside in an Einstein Analytics dataset? There are options available to pull data into dashboard. Data in Salesforce can be queried in SOQL (Salesforce Object Query Language) and displayed in a visualization.
Data access can even be extending outside of Salesforce with an Apex step. In the case of our top accounts, code can be written to pull in quarterly earnings data to understand if the company is trending up and down.
4. Actionable Insights
When the story a dashboard tells drives a user to an insight, what should be done next? This is an area where Einstein Analytics shines. Rather than jumping from a external reporting tool back into a system of engagement like Salesforce, Einstein Analytics dashboards can link into the existing business processes.It only takes a few clicks on the dataset to set up actions. Actions can perform operations such as Opening the Record in Salesforce, Creating a Task or Starting a new Opportunity. For data from an external source, it can be opened with the action framework as well by specifying a url.
For advanced business process integration, it’s possible to create custom actions that utilize code based actions. For example, create an action plan for key accounts with poor customer satisfaction.
5. Analytics Everywhere
As designers, we spend most of our time inside of Analytics Studio, but not our users. Users spend their time inside of Salesforce, either online or through the Mobile App. Dashboards need to be embedded as pages within the user’s apps, as part of the normal workflow.There is a Wave Dashboard component that can be put on any Lightning Component. If use Visualforce or classic, dashboards can be embedded there as well. Using Lightning Out, the dashboard can be available outside of Salesforce entirely.
Beyond providing a view of the dashboard in Salesforce, give users contextual insights. When an Account is opened, show the dashboard filtered to just that Account. Everything a user needs to know, right when the page opens. This is done by adding a filter to link the values on the record to the analytics dataset.
Finally, don’t forget about your mobile users. Dashboard support multiple layouts. Build a long, narrow layout suitable for viewing on a mobile phone. Users will have what they need, no matter where they are.
Nice article. I realize its a few years old, but there are broken image links so its hard to get the full effect of your message. Wasn't sure if you were aware
ReplyDelete