Soapbox: Is my data telling me the truth?
How email security programs affected my perception of data integrity.
As marketers, we face the overwhelming challenge of demonstrating proof that our tactics are effective. But how can we convince management if we are not convinced of our own data?
Here’s the reality, which I recently learned for myself: If you’re running email marketing, it’s very likely that your performance reports are not disclosing the full truth… inflated CTRs (click-through rates) and open rates being the main culprits.
Email security programs – loved by recipients, hated by senders
Barracuda. SpamTitan. Mimecast. Email bots that serve a single purpose: to protect users from unsafe content. These programs scan inbound emails and attachments for possible threats, including viruses, malware, or spammy content by clicking on links to test for unsafe content.
For email marketers, this creates several challenges:
- Inflated CTRs and open rates due to artificial clicks and opens
- Disrupting the sales team’s lead followup process as a result of false signals
- Losing confidence in data quality (quantity ≠ quality)
Real or artificial clicks?
In reviewing recent email marketing performance reports, I noticed an unusual pattern: Some leads were clicking on every link in the email…header, main body, footer, even the subscription preferences link — yet they were not unsubscribing. Not only that, but this suspicious click activity was happening almost immediately after the email was deployed. I speculated that these clicks were not “human”, but rather “artificial” clicks generated from email filters.
Hidden pixels are your frenemy
To test my hypothesis, I implemented a hidden 1×1 pixel in the header, main body, and footer section in the next email. The pixels were linked and tagged with UTM tracking — and only visible to bots.
Sure enough, several email addresses were flagged as clicking on the hidden pixels.
All that brings me back to the question of whether or not marketing data can be trusted. It’s critical to “trust, but verify” all data points before jumping to conclusions. Scrutinizing performance reports and flagging unusual activity or patterns helps. Don’t do an injustice to yourself and your company by sharing results that they want (or think they want) to hear. Troubleshoot artificial activity and decide on a plan of action:
- Use common sense and always verify key data points
- Within your email programs, identify and exclude bots from future mailings
- Share results with management, sales, and other stakeholders
A word of caution…
Tread carefully before you start implementing hidden pixels across your email templates. Hiding links might appear to email security programs as an attempt to conceal bad links. You could be flagged as a bad sender, so be sure to run your email through deliverability tools to check that your sender score isn’t affected.
As the saying goes, “There are three kinds of lies: lies, damned lies, and statistics.” Sigh.
With different solutions circulating within the email marketing community, this is likely the “best solution out of the bad ones”. It all depends on what works best with your scenario and business model.
Opinions expressed in this article are those of the guest author and not necessarily Marketing Land. Staff authors are listed here.