How Well Can I Trust the Open Rates in Marketo? | Community
Skip to main content
Greg_Jackman
Level 2
June 22, 2016
Question

How Well Can I Trust the Open Rates in Marketo?

  • June 22, 2016
  • 4 replies
  • 8834 views

When an email recipient opens an email, is Marketo only able to count it as an Open if the images load and show up for that recipient? The article below says that Marketo only registers an Open if images are loaded. It uses Gmail's "Display Images" (does this option still exist in Gmail?) as an example : Can I trust my Marketo email open rates?

Also, I've heard that if a recipient sees an email loads in the preview pane in Outlook, Marketo registers that action as an Open, even though the recipient might just delete it then and there.

Can anyone provide some insight on these points? Thanks.

4 replies

Grégoire_Miche2
Level 10
June 22, 2016

Hi Greg,

You are right,  but this is not specific to Marketo. Almost all emailing systems have the same flaws as they all rely on the download of a tracker pixel image.

Open rates are not reliable.

Yet, they can be useful as a tendency or in an A/B test: their evolution over time with comparable content on similar targets is meaningful, since the image settings and email client statistic distribution remains the same.

-Greg

June 22, 2016

Hi Greg,

Rather than say "this is the number of people who opened" I frame it as "this is the MINIMUM number of people". We know at least that many people opened it but more might not have images turned on or they might have viewed through a preview pane.

Grégoire_Miche2
Level 10
June 22, 2016

Hi Christina,

Unfortunately, you cannot even be sure of the fact that this is a minimum . As Greg pointed out, Outlook can also send back false positives.

-Greg

SanfordWhiteman
Level 10
June 22, 2016

Not just a question of images enabled/disabled but HTML/text client as well.

And yes, downloading images in the quarter-second while a delete is processing doesn't mean the lead actually read the email. Anti-spam software can also download tracking pixels without human intervention, so some "Opens" weren't user-facing at all.

June 22, 2016

Hi Greg J. -

For the reasons stated by Greg M. and Sandford, it's a pet peeve of mine that Marketo (and most of the rest of the world - alas!) insists upon placing so much emphasis on Opens - particularly the Clicks to Opens ratio (as opposed to the Clicks to Delivered ratio, which is useful). Since Opens cannot be accurately measured, evaluating an email based on Clicks as a percentage of Opens is simply meaningless - because it's not accurate. What Opens can be useful for is to give an indication of deliverability issues. For example, if you have a fairly consistent Open rate over time and suddenly it plummets, you might have a deliverability problem. I have a detailed write-up on this on my web site here - TargetRight Marketing - see the middle of the page where it says "A Cautionary Word About Opens."

Denise

Grégoire_Miche2
Level 10
June 22, 2016

Hi Denise,

when you are talking about deliverability here, I gather you are talking about inbox deliverability, and not the server deliverability that is measured otherwise by Marketo together with the hard & soft bounces.

For future readers of this post, Inbox deliverability is about the capability of you emails to sneak through the anti-spam and other inbox filters / rules such as gmail categories, for instance.

-Greg

June 22, 2016

Hi Greg,

Yes - but it's not limited to inbox deliverability. If the email is trapped by an ISP or recipient server spam filter this fact won't necessarily show up as a bounce statistic. Spam filters often simply trap the email silently - i.e., without communicating back any indication of the fact.

Denise