SparkPost offers a real-time analytics metrics dashboard with over 35 different metrics that we actively track for our customers. You can customize your SparkPost reporting interface to reflect any metrics that are important to you at a particular time. This is a powerful set of tools for evaluating how your email performs.
Other providers give you the basics of how many messages you sent, and then how many bounced, opened and clicked. But no other provider gives you as much data to troubleshoot issues. For example, if you sent 100 messages and 10 bounced, you might assume that 90 got delivered. But that’s rarely the case because of how the message traverses the providers’ systems. Some messages get rejected because of malformed email addresses. Others get backed up in long queues because another sender has a large mailing ahead of yours. But how would you know – until it’s too late to do anything about it – if your ESP doesn’t provide you with the data.
If I put on my operations hat on, the metrics I might care about are:
- Targeted – how many did I intend to send? This is the basis for the rest of my comparison.
- How many policy rejections did I have? In other words, how many messages did SparkPost not send out because of account limits, or because they hit the suppression list?
- How many generation failures? If you are using the API to inject lists and templates, how many messages failed to generate because of missing or malformed template data?
I won’t go through every metric, but you can go through this exercise for your messages to get an accurate picture of which messages are actually getting out to the Internet and troubleshoot those that aren’t. Hint: you can scroll over each metric to see what it means or go here for a full list: https://support.sparkpost.com/customer/portal/articles/1929969-sparkpost-deliverability-metrics
Now, if I put on my deliverability hat on, I can start to diagnose which messages are making it to the recipient vs. which aren’t and why. Some metrics to look at are:
- How many were accepted? This means how many were accepted by the receiving ISP.
- How many were hard bounces? These are permanent failures – bad or nonexistent email addresses. If this is high, you might need to investigate how you collected those addresses.
- How many were spam complaints? This happens when recipients click on “this is spam” in their inboxes. A large number here means you’re not getting permission or sending content that’s significantly different from what the recipient expected. A high number of bounces or spam complaints will likely land you in the spam folder next time.
You can get additional detail by going to the Bounces and Rejections reports.
Finally, if I put my Marketing hat on, I’m likely looking for engagement. In the Summary Report, I’m looking at Accepted, Rendered, Clicks, and Unique Clicks. Further detail is available in the Engagement Report, which provides you with a funnel chart that clearly indicates how many unique emails were attempted in the campaign, how many were accepted by the email providers, the number of unique emails that were opened, and the number of unique emails that were clicked.
In addition to looking at all this data in aggregate, SparkPost provides you with a number of ways of looking at this information. For example, you can filter by recipient domain to see where you had a high render rate or where, perhaps, you might need to make adjustments because of a high bounce rate with a particular provider. You can also filter by Campaign and Templates to hone in on what content is working and what isn’t.
SparkPost also gives you access to the raw data, letting you peer deeper into the numbers to see in precise detail how your recipients are engaging (you can download it as a CSV file or access it via webhooks).
I encourage you to dig into the numbers and see the wealth of information available to you in SparkPost, then think about tags that are specific to your business to learn even more about your email engagement.
To learn more:
SparkPost Deliverability Metrics https://support.sparkpost.com/customer/portal/articles/1929969-sparkpost-deliverability-metrics
Consuming SparkPost Webhook Events with Loggly – Part 1 of 3 https://www.sparkpost.com/blog/consuming-sparkpost-webhook-events-loggly-part-1-3