• by David Baker , Op-Ed Contributor, February 13, 2012

There’s been a lot of debate about whether to target those who never open, read or respond to your email.  Conventional wisdom would conclude that consumers who don’t have a measured interaction through a channel in a given period of time would naturally be suppressed from future mailings.  But the email marketing community doesn’t give up that easily.  While we aspire to develop and deliver the best consumer experience we can through the email channel, it is still a numbers game and it’s hard to continually suppress audiences based on non-measured responses.  A few things to consider when/if you decide to isolate these audience segments:

-      What constitutes a non-responder?   Are they not opening your email, are they not clicking on the email? Over what period of time?  Some say 12-18 months is the appropriate window of time to measure this.  To me, to do this right, you really have to go back and apply non-responder rules to the context of the “source” where you received permissions.  Was it through promotions from the past, site promotions, third-party promotions? There are a lot of reasons why people don’t respond to email – as marketers, we can only guess. Some rules should apply to flaggin non-performing segments, but should you suppress from mailings?  That’s the million dollar question!

-      Non-responder vs. non-buyer?  One and the same?  Not necessarily.   There are those who will engage with your brand through email versus those who won’t and may be considered “lurkers.”  Many in the space will not suppress a non-responder, even if they are an active customer.  Suppressing a non-buyer/non responder segment is an easier pill to swallow, but suppressing email from an active buyer (through another channel), who simply chooses to not engage through email, is a tough decision to make. 

-      Non-responder impact on deliverability?  The purists in the space will say, don’t send customers/prospects email they don’t want.  By continually suppressing non-responders, over time, that deliverability and inbox place would improve and protect the brand.  That is true in some cases, but in many, many instances where volume and cadence is critical, keeping non-responder segments in the mailing audience actually helps deliverability and inbox placement by keeping volume high.  I’d not recommend suppressing non-responders unless you have a financial justification for mailing less.  The economics just don’t justify the extra effort.

-      Smart Filtering by the ISPs.    While I believe in natural and forced attrition/hygiene of mailing lists, I also believe that consumers are fickle and don’t always know what they want.  More doesn’t mean better, yet timing and context are the real factors we have a hard time measuring through response models.  As the ISPs get smarter about filtering relevant content, I’ve yet to see a material impact on performance.

-      Audience Management vs. Response Management.   These are two completely separate things.  Response management is optimizing response to a given behavior, while audience management is designed to maximize reach and definition.  This is important for assessing non-responder segments, since ”audience” has value even if they don’t click/convert on the email you send, for publishing and send ratio mix.

Non-responder segments, again, mean something differently to different type of mailers.  A cataloger, publisher, retailer and commerce-driven businesses won’t see a lot of value.  If life is about trade-offs, my belief is there are more important areas of segmentation, trigger, automation that need to be focused on, other than non-responder segments for these businesses.  In other industries, it may be more important, as it’s not about pure response models, it’s about reach, experience and connecting experiences -- some that can’t be measured well.  CPG, Automotive, Entertainment, Pharma are great examples of industries where audience matters over performance. 

In the many years I’ve been in the space, I’ve seen very few great response modeling exercises for email that were truly actionable -- very few that actually “paid” for themselves.   It’s definitely something to look at, to even use as a showcase on how you are improving the email or cross-channel experience. But when it comes down to performance improvement, take a conservative view -- it likely won’t give you the pop you may expect.