Impact of Apple Mail Privacy Protection (MPP) on Adobe open tracking

[ninalunde]

Hi all,

I was wondering if Adobe has made any effort to adjust open rates and/or provide insight into how open rates have changed after Apple rolled out Mail Privacy Protection (MPP) for their mail app?

In our case, we suspect that open rates in Adobe Campaign are too good to be true. Part of our brand portfolio is managed through Hubspot where we experience much lower open rates than in Adobe Campaign. Hubspot does offer the possibility to filter out machine opens and calculate an adjusted open rate. Does Adobe have anything similar? If not, how do we deal with the false positive open tracking?

[Marcel_Szimonisz · Accepted solution]

Hello @ninalunde,

there is no OOB machine open filter in adobe campaign classic. You will have to implement such detection on your own.

Marcel Szimonisz

[ninalunde]

Do you know how much, on average, open rates have increased? And, any tips on how to approach or interpret open rates going forward? 

[Marcel_Szimonisz]

I would create process that would either remeove or hide logs that will be considered opened by robot

Here have some ways how to figure it out (asked GPT), I know we did once implement the first two:

1. User-Agent Analysis: Bots often use different User-Agent strings than regular email clients or web browsers. You can log and analyze the User-Agent header of the HTTP request that loads tracking pixels or resources in your emails. Look for patterns that match known bot User-Agents and flag those interactions.

2. Abnormal Behavior Patterns: Create algorithms that identify unusual patterns of email opens, such as rapid-fire opens within a short time frame. Bots may exhibit consistent and unnatural behaviors, like opening emails every few seconds.

3. IP Address Analysis: Track and analyze the IP addresses of email open events. Look for IP addresses that belong to known data centers, cloud providers, or hosting services, as these are often used by bots. You can use IP reputation services to identify suspicious IPs.

4. Geolocation Data: Check the geolocation data associated with the IP addresses. If the IP resolves to an unexpected location or a location where you don't expect legitimate recipients, it may be a sign of bot activity.

5. Inconsistencies in Email Rendering: Bots may not render emails exactly like human recipients. Monitor for inconsistencies in email rendering, such as missing images or CSS styles, which could indicate bot activity.

6. Email Client Detection: Track which email clients are used to open your emails. Some bots may use custom or uncommon email clients. Analyzing the client data can help identify suspicious behavior.

7. Click Tracking: If your emails include links, monitor for clicks from the same email open event. Bots may not follow the expected click-through behavior of human recipients.

8. Rate Limiting: Implement rate limiting on your tracking pixels or resources to prevent excessive opens from the same IP or User-Agent in a short time.

9. Machine Learning: Train machine learning models using historical data to detect patterns of bot behavior. Machine learning can help identify anomalies that may indicate bot activity.

10. Behavior Analysis: Consider tracking user behavior beyond just email opens, such as page visits on your website after clicking links in the email. Bots may not follow typical browsing behavior.
    
    
    Marcel

[AmberlyGw]

No and i dont have a device of my own im a consumer and developer i need to get prouducts asap i have a open source license 

Apple hub wiki 4.0 license but they have im a non profited organzation i need help please help me please i dont want to loose everything ive worked so hard for i also have a MIT LICENSE