Use Einstein Marketing Insights
After completing this unit, you’ll be able to:
- Describe Einstein Marketing Insights.
- Create and use Einstein Marketing Insights.
- Identify the parts of an Insights Bot.
What Is Einstein Marketing Insights?
Einstein Marketing Insights (EMI) is an insights discovery tool that analyzes your data to uncover key factors affecting your marketing performance—whether good or bad. With a tool like EMI, you can be proactive with your data and find areas to improve or reinforce. To do this, you create an Insights Bot within EMI.
The Insights Bot creates two types of insights:
Over and Under Performers: Segments of your data with performance and volume in a selected time period that stand out.
- For example, last month a campaign’s click-through rate was 60% above average, and its volume was substantially high—11% of overall impressions during that month.
Change Drivers: Segments of your data that caused a change in overall performance between two date ranges.
- For example, if your click-through rate dropped significantly between last week and this week, it’s possible that your site’s click-through rate decreased by 70%, and its impressions increased from 2 million to 5 million, which caused the drop.
Let’s create an Insights Bot to learn more.
Create an Insights Bot in Einstein Marketing Insights
In this module, we assume you are a Datorama user with the proper permissions to create an Insights Bot. If you’re not, that’s OK. Read along to learn how you would take the steps in a production org. Don’t try to follow these steps in your Trailhead Playground. Datorama isn’t available in the Trailhead Playground.
In Datorama, go to Analyze & Act, then Einstein Marketing Insights and click the Add New (+) icon to create a new Insights Bot.
To create your new Bot, follow these steps.
- Select the KPI you want EMI to optimize, like Visitor Bounce or CTR (click-through rate).
- Add a Filter to narrow down the data set, if needed.
- Select whether you want to optimize your KPI by increasing or decreasing its value.
- Once you finish setting up your EMI Bot, click Save & Run to run the Bot with its default settings.
After clicking Save & Run, default Analysis Settings are applied to your Bot. You can change these defaults by editing a bot. Here’s how.
Edit and Customize Your Bot
To change your settings, go to the Insights Bots main page, click and then click Edit Bot. Next, click Analysis Settings in the sidebar.
Here, you are able to:
- Change the types of analysis and date ranges that are analyzed.
- Change the bot scheduling.
- Change the dimensions that the bot examines.
Customize your bot settings further in the Advanced tab.
These options are available:
- Volume Measurement: Use this to change a bot's automatic selection of Volume Measurement to better reflect the logic behind the measurement.
- Mail Notification: Send an email notification to the bot creator or other users when new insights are available.
Now that you’ve completed your bot set up and have run the bot, EMI starts analyzing your data based on the settings. After analyzing, you can open your Insights Bot and begin optimizing your data!
Inside an Insights Bot
From the Insights Bots main screen, you can see all the Insights Bots you created. You can search for a specific bot (1) or group them by KPI (2) and view a summary of your KPI performance (3).
Click your newly created Bot to open it.
At the top, you see a summary of what changed (1). Next, you see a visual display of what changed over time and a quick way to view the insights (2). There are three views available: Week Over Week (WoW), Month over Month (MoM), and Overall change.
Next, there’s the Filters bar (3) to narrow the insights you are viewing (by a specific dimension, for example).
Lastly, you see a list of all insights that were found in the Insights Bot, ordered according to EMI’s evaluation of the most valuable insights (4).
Each insight within the list includes detailed information. Here’s an example showing that the campaign Seasonal Sweepstakes and creative type Social Video Link caused an increase in the overall CTR month over month.
Click an insight to better understand why and how the KPI was affected. You can also see the selected dimension and its value, as well as its impact on the KPI. Here’s an example of a change driver insight.
For this type of insight, you’ll find:
- The change in the KPI’s numerator between the two date ranges. In this case, the numerator is clicks.
- The change in the KPI’s denominator between the two date ranges. In this case, the denominator is impressions.
- The KPI’s change between the two date ranges, compared to the overall KPI change.
When looking into a Performance Insight, the detailed view displays a value comparison of the dimension the insight is based on. You can see how other dimension values performed.
Color coding is used so you can distinguish between the overall data and the Dimension data.
Here’s an example of a performance insight.
For this type of insight you’ll find:
- The highest or lowest performing media buy according to the KPI you set. This shows an overall view of all the media buys’ performances.
- A breakdown of the highest or lowest performing media buy so you can identify the cause of high or low performance.
As you can see, there are many ways to use EMI to make data-driven decisions—without having to manually review and analyze data. So, while you focus on improving your marketing performance, allow Einstein Marketing Insights to surface the information you need.