It only takes a couple of minutes to start streaming email event data from SendGrid to Keen IO and once you do you can start analyzing and visualizing your data in a whole bunch of different ways.
This guide covers:
- Sending Your SendGrid Data to Keen IO
- Testing Your Integration
- Using SendGrid Email Analytics
- Querying and Accessing Raw Data
- Embedding SendGrid Analytics Into Your App
- Additional Resources
Step 1: Create a Keen IO and SendGrid account
Step 2: Activate SendGrid’s Event Notification App
Log into your SendGrid account, click on Settings from the left menu → Select Mail Settings. Then click Event Notification.
Step 3: Set the HTTP POST URL (example below) as the endpoint where SendGrid will stream all of your email data.
You can snag a prebuilt URL in Keen’s UI or make your own.
In the settings, select the events you want to post to Keen IO (why not all of them?). To do this, under Select Actions, check All.
That’s it! Now, as your emails make their way through SendGrid, all of the event information will be posted to Keen IO. A new Event Collection for each type of email event will be created within your Keen Project.
Test that everything is working by clicking on the “Test Your Integration” button in SendGrid’s “Event Notification” Settings. This will send sample email data from SendGrid into your Keen IO Project. In the Keen UI, click on “Check for SendGrid Data” button.
Your SendGrid email data should now be populated in Keen IO.
Once you have data successfully flowing from SendGrid to Keen, a new application called SendGrid Email Analytics will appear on your projects page’s “Apps” tab.
To find it, visit your Keen home screen, select the project you’ve connected, in and click on the “Apps” tab. Find SendGrid and select “Launch App”. If you’ve successfully sent SendGrid data to Keen for storage, you will see the dashboard below.
A report for All of your Mail” will be created automatically. You can see Delivered Rates, Unique Opens, Unique Clicks, Click Rates, Unsubscribe Rates, Bounces, and Number of Emails Marked as Spam.
For each of these metrics, you can download a CSV of the detailed event data. Just click on any cell, and you’ll see a menu with the option to download a CSV. Learn who did or did not open an email for example, and with this list of email addresses you can build a more targeted user campaign.
A segment will allow you to track a specific campaign or filter on a particular field like a customer ID. To create a segment, select “Add Segment” from the top left corner. The following pop-out menu will allow you to define your segment name, the date you’d like to look at, and, most importantly, select a property by which you’d like to measure the effectiveness of a campaign.
This is explaining a different process from creating an customer email segment in SendGrid
Here’s an example of a segment created to look at delivered, open, and click rates for emails sent within the United States.
You can see Delivered Rates, Unique Opens, Unique Clicks, Click Rates, Unsubscribe Rates, Bounces, and Number of Emails Marked as Spam. For each of these metrics, you can download a CSV of the detailed event data. Just click on any cell, and you’ll see a menu with the option to download a CSV. Learn who did or did not open an email for example, and with this list of email addresses you can build a more targeted user campaign.
To edit a segment, click on the segment you newly created: you’ll find a menu that allows you to edit it, clone, or remove it.
SendGrid gives you the ability to include additional data about your contacts using custom fields. These fields are automatically fully available for segmentation and filtering in this app.
Once you’ve added an attribute you’d like to use for analysis (for example: customer_ID, company_name, email campaign ID or campaign names), you can use segments to track email performance for these custom attributes. Here’s an example:
You can track your campaign’s performance over time by selecting the timeframe you’d like to see analyses and metrics. You can change your timeframes to compare campaign performance from the first 24 hours to the first 14 days of the campaign.
The app’s dashboard defaults to 72 Hours.
Uniform timeframes that compare each campaign to the same period of performance gives you consistent data points that you can measure against.
Your data is stored in Keen and does not expire for as long as your account is active. The SendGrid Email Analytics integration results in a data warehouse or archive of all of your SendGrid email data. This means your user transaction and email behavior data is stored for as long as you need it.
The report has fixed timeframe options, but that doesn’t mean that we stop storing your data after 14 days. If you need to access older data, see the section below on Ad-hoc Querying.
Each column in the the Analytics App represents a different user action and each column is a unique step. From left to right, the columns display the number of people that successfully make it through each event. The fewest number of users have successful delivered events, open events, clicked events, and then unsubscribe events.
The value presented shows the number or percentage of users who completed the action successfully. With the Units option, you can toggle between percentage or raw numbers.
For each segment in each step, you can download the segment and retrieve a list of all unique email address who completed the user action or you can download those who didn’t complete the user action. You can download a CSV for every custom segment that you create. As an example below downloads a list of all customers who successfully opened and email and clicked on a link, or a list of all the users who opened the email but didn’t click.
In addition to the data views you see in the Keen SendGrid App, all of your SendGrid data flowing into Keen is also available for raw analysis, programmatic querying via API, and full resolution download.
Use the ‘Explorer’ tab to create ad-hoc queries and supplemental dashboards.
Step 1: Log into your Keen account, select your Project
Open the “Streams” tab, and select the
SendGrid Email Events to check out your data collection.
Step 2: Run some basic queries
Open the Explorer tab and enter
count as the Analysis Type, select
email_delivered as your Collection Name, and click Run Query. You ran a query!
For more information about the types of queries you can do with the Keen Data Explorer, see our Compute page. Some example query types are sums, averages, min or max, funnels, and select_unique.
For some extra-pretty graphs and custom analyses here are some tips:
- Adding an “Interval” (e.g. “daily”) will give you a line graph over time
- Include a “Group By” property (e.g. “campaign”) to view results categorically
- Use “Filters” to refine the scope of your query
- Use the chart type drop down on the upper right to choose how you view data: as an area graph, bar chart, column chart, table, or more
- Download a CSV of the data you found by clicking on the “Download CSV” button
Once you’ve found the query you need, you can either copy the URL right from your browser’s address bar and share it with a colleague, or you can save the query and make yourself a dashboard with Keen.
To create a dashboard:
Step 1: Save your query. Create your query in the Explorer and click Run. After running the query, click Save.
Saved queries are the most efficient way to revisit your favorite metrics. Rather than entering the same query parameters over and over again, queries can shared with your teammates.
Your saved query can be edited and cloned.
Step 2: Add your graph to a dashboard.
Select Create Dashboard, name your dashboard, and add the query we saved to the first tile. Then click Save.
With Keen, you have access to all of your raw data for as long as you need it. An archive of all of your email data is created, which means you have access to your historical data and can query for past user behavior.
Just like SendGrid, Keen IO is 100% powered by APIs. This means you can embed rich analytics anywhere you can write code.
Many customers find it useful to embed analytics into their products for their customers or customer success teams. For step-by-step instructions on how to embed your SendGrid analytics into your product, check out our Native Analytics Guide. Some customers also take advantage of Keen IO’s S3 Streaming capabilities to write a copy of all their data to AWS.
- Know someone who’d appreciate SendGrid Email Analytics? Share our product briefing with them.
- SendGrid’s complete Event Webhooks Documentation is a great resource on SendGrid email events. As you start to do more analysis, it’s worth a read. You can even skip a lot of the sections, because we’ve done much of the work for you.
- Build a campaign and email marketing dashboard. Create and display your email metrics using one of our dashboard templates.