Reporting FAQs

Here are answers to some frequently asked questions and some tips to help you troubleshoot campaign reports in Blueshift.

General

Q. Can I get a detailed user activity report per campaign?

Yes, Blueshift provides a summary, detailed, and user activity report for each campaign.

Q. How are push notifications tracked?

For push notifications, opens = sends/users, click = app opened and an event was generated within 15 minutes of opening the notification.

Q. How are bounces tracked?

Hard bounces are displayed on the dashboard. For soft bounces, Sendgrid (or your vendor) will retry up to 72 hours.

Q. Do you track deliverability of push messages?

Once we send a message to GCM/APNS, it is upto their service to ensure deliverability.

Q. How can I find out the number of invalid users or users who were not messaged due to some errors?

The Campaign execution summary gives the number of skipped users and the reason for the message not being sent.

Q. Why don't the numbers in various reports match?

One of the reasons the numbers in reports might not match is because they might be run for different time periods. A particular time window might not capture all the data. Hence ti is important to select the correct time period for the report.

Consider the following example:

There are 2 one-time campaigns, A and B.

Campaign A is sent to 100 users on 12/1/21.
- 20 users opened the message on 12/1.
- 20 users opened the message on 12/2.
- Another 20 opened the message on 12/3.
Campaign B is sent to 10 users on 12/2/21.
- 2 users opened the message on 12/2.
- 2 users opened the message on 12/3.
- Another 2 opened the message on 12/4.

Open rate for 12/2: The total sends on 12/2 = 10 and total opens on 12/2 = 22. So open rate = 220%.
Open rate for 12/3: The total sends 12/3 = 0 and total opens on 12/3 = 22. So open rate = ∞.
Open rate for the the period 12/2 through 12/4: The total sends = 110 and total opens = 66. So open rate = 60%

Only by measuring the open rate for the period 12/2 through 12/4, do we get the correct open rate for these 2 campaigns.

Campaign performance report

Q. Why is the Unique open rate for a particular trigger in the campaign is greater than 100%?

The Unique open rate is calculated as unique impressions/delivered in percentage format.

Consider that you have a campaign that has sent “X” messages in week 1 and “Y” messages in week 2.
If you look at the metrics for week 2, you see that the delivered messages for week 2 are Y-n and the unique impressions for that week are Y-n+Z which is greater than the number of messages delivered. (“n” is messages that were sent but not delivered).

This is possible if some of the unique impressions are for messages that were delivered in week 1 but were opened by customers in week 2. If you look at a broader time period or for the time period of the entire campaign, you will see a more accurate unique open rate.

Q. Why doesn't the campaign performance report show any data for the campaign but the campaign index page shows metrics for the campaign?

Verify that you have selected the correct date range for the campaign performance report.
If you have multiple channels in the campaign, verify that you are selecting the right channel.
Check if the triggers that were used to send the messages are active or are archived and select the filter options accordingly.

Campaign Engagement report

Q. Why is there a difference found in the unique clicks in the Campaign Engagement report and the external tool we are using?

Blueshift uses the HyperLogLog algorithm for estimating unique impressions and unique clicks. This algorithm has an accuracy of 98%, which means that results will be within a 2% error margin. The external tool that you are using might also have a certain margin of error. The observed difference in the unique clicks could be due to this.

Insights reports

Q. How do I analyze user versus prefetch opens for emails?

You can create a Campaign performance insights report to analyze user versus prefetch opens for emails.

Select the following criteria for the report:

Metric = Impressions
Time = Based on your requirements.
Group by = Opened by. By selecting Opened by for grouping, you can see the open count source – whether it’s by mail client using prefetch, by a proxy server or a real human (user).
If you are looking at opens for a particular campaign, select the time period as the duration of the campaign. Add a Filter for Campaigns and select the required campaign.

Q. How do I analyze clicks for legitimacy (user-generated vs. bot-generated) in my campaigns?

You can create a Campaign performance insights report to analyze user versus bot clicks for emails.