March 13, 2019

Four potential problems with Return Path’s SmartSeeds, a bot-driven email data solution.


    • Google announced plans to halt any use of Gmail inbox data from third parties by March 31, and other mailbox providers are expected to follow.
    • Return Path’s answer to the threat of losing email panel data is SmartSeeds, computer-generated (fake) behavior to mimic human interaction (real).
    • Postmasters are sounding the alarm over its attempt to game spam filters, and we’re concerned about negative impact on sender reputation to not just those paying for SmartSeeds, but senders SmartSeeds interact with (without their knowledge).
    • Four big problems with SmartSeeds: 1) It won’t provide true insights; 2) It’ll skew other data points; 3) It doesn’t depict real subscriber behavior similar to panel data, and; 4) It could negatively impact engagement data for non-subscribing senders.

After hearing email panel data’s death knell, some email data vendors realized they needed to come up with a new way to provide the same type of recipient and engagement data they’d supplied their customers for years (against better judgment) once their biggest source of data dried up.

In theory, marketers use email panel data to get a better understanding of email engagement, activity, and success by using real people’s behavior, leaving privacy-minded folks on edge. The fact remains, they use this information to make decisions about their email programs to better optimize for success. Without a viable, non-sketchy method of collecting this email data from Gmail users, marketers who rely on Return Path’s panel data (and the value that comes with it) are in jeopardy of seeing that value disappear once Gmail pulls the plug.

Return Path’s answer to the impending loss of panel data is SmartSeeds…and it’s got issues. Essentially, SmartSeeds are bots mimicking human behavior, acting as an engaged seedlist. At its core, it’s computer-generated (fake) activity meant to appear as real to mailbox providers.

That’s troubling, because any pattern of behavior linked to “trying to evade existing spam filtering tech” tends to be treated with extreme prejudice by spam filters once it’s detected. Will Google, Verizon, and Microsoft adjust their algorithms to allow this type of fake activity? Unlikely.

Even more alarming for marketers, if data networks like SmartSeeds get flagged, the negative reputation effect would be felt directly by the sender.

Here’s one more troubling tidbit: Per Return Path, “SmartSeeds will also allow marketers to track specific campaigns or segments, while also providing enhanced monitoring of their competitors’ campaigns.” How will they do that? If SmartSeed addresses sign up for non-customer email, other senders not using SmartSeed data to optimize email could still be unknowingly implicated. So not only could this technology put the paying recipient of data at risk, but it may also put other innocent senders in the crosshairs, without giving them the benefit of an email intelligence payoff.


You might be wondering, what frame of reference do we have to make such claims when SmartSeeds are brand new? Plot twist: Because it’s not necessarily their innovation. In fact, Return Path’s competitor eDataSource is using a “new” technology called “IntelliX AI Network™.”  eDataSource explains how this works as, “a large cohort of enhanced, intelligent seeds, programmed to simulate a representative range and distribution of global user locations and attributes.” That sounds a lot like the behavior powering SmartSeeds to us. In fact, we know IntelliX AI Network™ signs up for non-customer email, making our suspicions above feel more than reasonable.

And yet, that’s not the plot twist. eDataSource wasn’t the first to do this either.

Email bots mimicking human behavior is not a new idea.

In 2010, Boris Mizhen, a U.S.-based spammer, created fake accounts and moved hundreds of thousands of his spam messages every day from Hotmail spam folders into their inboxes. Microsoft ended up suing him for using bots to exploit spam filters.

We asked former AOL postmaster and current 250ok Expert Sridhar Chandran his thoughts about this behavior from a mailbox provider perspective. He laid out his concerns in no uncertain terms.

“The premise of SmartSeeds is what spammers have been doing for years, so mailbox providers have measures in place to catch these accounts. These ‘spammy honeypots’ are actively monitored by postmasters, and they will not be thrilled to learn about a commercial product designed to mimic behavior.”

Sri continues, “Seed testing has been around for so long because, at the end of the day, it doesn’t affect the deliverability filters at mailbox providers. Faking activity, however, can have negative consequences. From the postmaster’s perspective, faking activity falls into the same category as spammers trying to break reputation filters. The behavior is very similar.”

Another postmaster, who asked for anonymity, at one of the Big 3 email providers (Gmail, Verizon, Microsoft) further commented, “Receivers have had a very difficult time trying to get email service providers to come to terms with how their systems can be abused. I cannot believe a bot network will be any different. Paving the way for inauthentic IP and domain reputations will simply invalidate reputations as a tool receivers can use to assist in the spam determination.”

Let’s set aside all ethical and operational concerns with SmartSeeds. Will this technology provide a comparable alternative to email panel data for marketers?


Sketchy or not, there are four reasons why the “solution” proposed by Return Path is potentially a bad idea.

    1. It’s not actually going to provide much insight. Gmail, who comprises 50% or more of most modern lists, absolutely leverages recipient level filtering. So once one of these SmartSeeds opens or clicks, any subsequent sends will be incredibly likely to inbox due to this personal history of engagement.
    2. It could mess with marketers’ own metrics, because savvy senders track engagement data, too. SmartSeed accounts could skew metrics like open rates.
    3. Even if it can mimic human behavior, Return Path always claimed panel’s value is derived from it being real recipients. SmartSeeds aren’t actual recipients from a marketer’s list. Beyond that, bots cannot make emotional decisions or react to content, leaving the more valuable information (like what’s resonating or what’s not) completely uncovered. This is why analytics products are important, because they do give you that information.
    4. If SmartSeeds subscribe to newsletters of not only paying customers, but innocent sender-bystanders, who’s to say SmartSeeds will appropriately interact with that mail? And if SmartSeeds marks mail as spam, as humans do, people who aren’t even benefitting from Return Path data could be penalized.

We don’t keep it a secret. We think optimized seedlist testing is a safer, more accurate, and ethically sound way to get the same kind of data, and we recommend you combine it with advanced engagement analytics. Our approach works, just ask four of the top six US internet retailers (by sales share) in 2018.

Will mailbox providers crush these spammy accounts from Return Path? Time will tell. In the meantime, you’ll want a more reliable way to get email data than relying on the next panel data experiment.

Let’s discuss how to get it.

Author: LoriBeth Blair

LoriBeth "LB" Blair is an email geek with experience working for two ESPs and her own private deliverability consulting practice.

You may also like...

The Year in Email 2019

It’s hard to believe we are nearing the end of yet another exciting year in email, and 2019 proved to be one of the most momentous and active years to date. Over the past year, the number of new technologies, mergers and acquisitions, mailbox provider (MBP) announcements, news, and highlights is evidence of the versatility […]

[Infographic] Global Privacy Relationship Status: It’s Complicated

I recently gave a presentation on global privacy regulations to a post-graduate marketing class and one of the things I noticed while preparing was that even within a single country, privacy is complicated. On a global scale, it is really complicated. For example, Canada has one federal private sector privacy law, three similar provincial laws, […]

The Year in Email 2018

*Update: This article was featured on email influencer Jordie van Rijn’s emailmonday blog! To see it in action, plus a great round-up of other articles and thought leadership looking forward to the future of email, click here.* The Black Friday emails are deleted, marketers’ email lists are checked twice, we pretty much know which senders […]

Ready to get started?