We’re seeing more and more consistent issues with open rate metrics. We’ve been on a major project to improve deliverability and open rates thinking that mailbox placement was the main culprit, which likely does play a big role, however we’re now finding less and less validity in open rate metrics.
We are B2B with clients often using enterprise email security solutions so difficulties are slightly nuanced. All DNS settings are configured as well as they can be considering return path cannot be customised on the professional tier, so an SPF misalignment is forced.
We are in testing and sending to verified client accounts yet still seeing well below benchmark open rates. Upon talking to some, we can confirm they have received opened and read our email yet the metrics are not reflecting this. in the absence of a CTA click it is impossible to quantify the performance of the email in any way.
In light off this I am also wondering if bot clicks are being incorrectly categorised, but do not want to skew metrics by switching off.
is anyone else facing a similar situation and if so, how are you navigating it?
Hi @RSingleton1,
My take is that it’s not possible to get reliable open information anymore - and that marketers have an easier life accepting that they shouldn’t use it as a hard metric, more a general indicator of performance.
See also here: Understand bot filtering in marketing email analytics
The article mentions that HubSpot “suspects” certain activity to be bots. It’s an attempt and an ongoing arms race between bulk email tools and email service providers, clients and IT departments. Between these attempts, activity will be incorrectly categorized.
Best regards!
Would you even categorise open rates as a general indication of performance at that point or would you ignore it altogether?
Assuming we’re taking open rates with a huge grain of salt, and the metrics to be focused on are click rates, how do you determine whether low metrics are a result of poor placement or content related, besides explicitly asking each recipient?
@RSingleton1 I look at it with a bad/average/good lens. The current mechanisms should be good enough still to flag whether an email is performing far below normal performance, for example.
In terms of optimization, I think this new reality forces marketers to even more critically look at their recipients, the content they’re sending and to ask whether what is being sent is actually relevant or just trying to get someone to download something for getting the numbers up.
Surveys, short email conversations with subscribers or actual interviews can be a great way to optimize content.
Another way to optimize is looking at downstream metrics, like page views, form submissions etc. It’s definitely a more manual process, for sure.