Table of contents
  1. Key takeaways
  2. Deliverability starts with identity but succeeds through trust
  3. What mailbox providers really measure when deciding inbox placement
  4. What’s next for deliverability? Trust signals, DKIM2, and the role of AI
  5. Wrapping up
Experts’ opinions
yesterday

The future of deliverability with Laura Atkins: Why reputation will matter more than authentication

Author
Alina Samulska-Kholina
Alina Samulska-Kholina Copywriter and content writer at Stripo
The future of deliverability with Laura Atkins _ Why reputation will matter more than authentication
Table of contents
1.
Key takeaways

When the term “email deliverability” is mentioned, marketers expect a conversation about technical requirements, configuration errors, and how to fix everything in three easy steps. However, our conversation with Laura Atkins focused primarily on a recipient-first philosophy in email marketing, trust, and reputation. These are the things that determine deliverability now and will continue to do so in the future.

Read the interview with Laura Atkins, Word to the Wise co-founder, speaker, educator, writer, and email specialist, about why authentication establishes identity, how reputation is built through recipient behavior, and why the future of deliverability belongs to brands that create emails people genuinely want to receive.

Key takeaways

  1. Authentication establishes your identity. Technical compliance with SPF, DKIM, and DMARC is now the baseline, not the competitive advantage. Long-term inbox placement depends on whether subscribers consistently engage with your emails.
  2. The future of deliverability is increasingly personal. Mailbox providers are moving beyond sender-level reputation toward user-specific filtering, making recipient behavior a stronger signal than ever. 
  3. AI won’t replace a recipient-first strategy. While AI can support email production, it can’t compensate for irrelevant messaging or poor customer understanding.

Expert

Laura Atkins
Co-founder of Word to the Wise and Deliverability Consultant; Co-founder of Women of Email

Laura Atkins is a deliverability consultant, co-founder of Word to the Wise, and a pioneer of the email deliverability industry. For more than 25 years, she has helped both email senders and mailbox providers improve inbox placement, authentication, and sender reputation. She is also a co-founder of Women of Email, an active participant in M3AAWG, and has chaired the IETF DKIM Working Group, where she continues contributing to the development of DKIM2, the next generation of email authentication.

Originally trained as a molecular biologist, Laura entered the email industry in the late 1990s through the anti-spam community and later helped establish deliverability consulting as a profession. Today, she is widely recognized for her recipient-first philosophy: successful deliverability isn’t about passing technical checks but about sending emails that people genuinely want to receive.

Deliverability starts with identity but succeeds through trust

Stripo: You’ve been in the email industry since the late 1990s, before “deliverability” was even considered a separate discipline. Looking back at nearly three decades of change, what still makes email such a valuable communication channel, and how did your path from molecular biology and anti-spam communities lead you into this world?

Laura: For me, email is valuable because of its persistence. Over the years, we’ve lost a tremendous amount of valuable information. Companies get acquired, websites are redesigned, documentation disappears, and suddenly, years of knowledge are simply gone. Sometimes you can recover pieces of it through the Internet Archive, but often it’s lost.

The newsletters and emails I receive are different. I can save them on my computer, keep them in my mailbox, and come back to them years later. That persistence makes email incredibly valuable.

Laura Atkins

Laura Atkins,

Co-founder of Word to the Wise and Deliverability Consultant; Co-founder of Women of Email.

Another reason is that email is still largely an open channel. Yes, we’re seeing increasing dominance from a handful of large technology companies, but email remains fundamentally open. Anyone can still set up a mail server and send email. You can’t do that with platforms like Instagram. You can’t spin up your own Instagram server. You have to participate in someone else’s platform.

I think that openness will become even more important. Here in Europe, we’re seeing growing interest in reducing dependence on large technology companies that collect and monetize user data. Email gives people much more control because no single platform completely owns it.

I also think we’ll see a renaissance of smaller organizations running their own mail servers again.

The persistence of email matters in practical ways, too. I’m part of several Slack communities with other deliverability experts, and we’ll often have long discussions about changes at mailbox providers or new filtering behavior. Those are valuable conversations. But a few months later, they’re simply gone unless someone pays significant sums to preserve that history.

If those discussions had happened over email, we’d still have those conversations available 10 or 15 years later. That’s one of the things that continues to make email such a valuable communication channel.

Stripo: We already went through the fundamentals of authentication, spam filtering, and Gmail/Yahoo requirements with you in our previous conversation. From your perspective, are we now entering a “post-authentication era” in which the role of authentication and technical compliance in deliverability is evolving, and what do you see as the real differentiator for senders going forward?

Laura: In some ways, I think authentication is almost a red herring. Don’t get me wrong: authentication is absolutely required today. It’s no longer optional. But authentication doesn’t create deliverability. Authentication establishes identity. It tells mailbox providers, “This is who I am.”

Everything that actually determines deliverability is built on top of that identity. Deliverability doesn’t start with authentication; it starts with people.

Laura Atkins

Laura Atkins,

Co-founder of Word to the Wise and Deliverability Consultant; Co-founder of Women of Email.

From my perspective, deliverability has always been about the end user. What does the recipient want? Reputation only exists once you have a verified identity. Authentication provides that identity, and then reputation develops based on how recipients respond to your mail.

I don’t think this is a new way of thinking. It’s always been about meeting the recipient’s needs. The email belongs to the subscriber. It’s their mailbox, their inbox. They decide what happens there. Authentication simply gives mailbox providers and recipients confidence that they know who the sender is. Once that identity is established, they can determine whether it’s a sender they trust.

Over time, we’ve moved from IP-based reputation, which was all we had in the early days, to domain-based identity through authentication. Before SPF and DKIM existed, the only identifier we could trust was the sending IP address, so filters relied heavily on IP reputation.

Today, we have authenticated domains that provide a much stronger identity. Authentication establishes who you are. Reputation determines whether people actually want your email. That’s the real differentiator, and it’s been that way for a long time.

Stripo: You often say that deliverability ultimately comes down to “send mail people actually want.” That sounds deceptively simple, yet the industry continues searching for technical shortcuts. Why do you think marketers still struggle to accept that recipient behavior is the real center of deliverability?

Laura: I think marketers don’t like that idea because they’re not fully in control of it, and it’s genuinely difficult. Good marketing to an individual person is hard. Traditional marketing is about convincing a target audience that they have a problem your product can solve. But in email, that audience isn’t an abstract market. They’re real people.

When you send an email, you’re entering someone’s personal space. Years ago, I wrote a blog post comparing the inbox to someone’s home. Whether they’re sitting at home on their phone or working at their desk, you’re entering a space that belongs to them. Your email has to feel welcome there.

That’s exactly how mailbox providers build their filters. If you talk to the people at Google or Yahoo who develop these systems, they’ll consistently tell you their goal is to make their users happy.

Of course, there are business reasons for that. The longer users stay engaged with their inboxes, the more valuable those platforms become. But regardless of the business model, the filters are designed to prioritize what recipients actually want to receive.

As marketers, we don’t own that space. Unlike television advertising or other media, where advertisers pay for placement, you don’t pay Google to deliver your email. The inbox belongs to the recipient, and the mailbox provider’s job is to protect that experience.

You might also like

Gmail and Yahoo’s sender requirements: The after-effectsGmail and Yahoo’s sender requirements: The after-effects

What mailbox providers really measure when deciding inbox placement

Stripo: You’ve spoken about the idea of “two filters” influencing inbox placement: technical filtering and user filtering. Can you explain how mailbox providers increasingly rely on recipient behavior signals, and why many marketers still underestimate this shift?

Laura: It all starts with the mailbox provider trying to make its users happy.

When I explain filtering, I often say that mailbox providers ask two questions:

1. Is this email safe?

That’s the technical filtering layer. There is an enormous amount of malicious email on the internet, including phishing attacks and malware. Before mailbox providers think about marketing, they have to protect their users.

Most legitimate marketers don’t think much about this because they aren’t sending phishing emails or malware. But every email still has to pass through those safety checks, and that won’t change.

2. Is this email wanted?

That’s where reputation comes in. Mailbox providers combine data from authenticated identities, IP addresses, blocklists, their own internal systems, and, most importantly, user behavior. They ask whether people have had a positive experience with mail from this sender before. If the answer is yes, the message is much more likely to reach the inbox.

That’s the technical filtering side. The second layer is user-specific filtering. Sometimes a sender is popular overall, but an individual recipient doesn’t want those emails. Other times, one recipient loves a sender that most people ignore.

Mailbox providers increasingly personalize inbox placement based on those individual preferences. If they have enough behavioral data for a particular user, they may deliver the same campaign to one person’s inbox while placing it in another person’s spam folder.

For example, if someone uses Gmail through Google’s web interface or the Gmail mobile app, Google can observe a great deal about how that person interacts with email. If that same person reads Gmail through Apple Mail on a Mac, Google has far less visibility into those interactions. In those cases, inbox placement relies more on the sender’s overall technical reputation than on individual behavioral signals.

I think many people in deliverability, myself included, have simplified this explanation over the years because it’s conceptually more difficult. But recognizing that mailbox providers combine technical and user-specific filtering is essential to understanding how modern deliverability actually works.

Stripo: You’re closely involved with the future of DKIM2 through the IETF. Without getting too deep into specifications, what problems is DKIM2 trying to solve that current authentication standards cannot adequately address?

Laura: DKIM is nearly 20 years old. SPF is even older. These protocols were designed for an internet and an email ecosystem that looks very different from what we have today. Part of DKIM2 is simply modernizing authentication to better reflect how email actually works today.

There are really two major problems that DKIM2 is trying to solve:

1. Forwarding: When you send an email, you’re handing it to another mail server. Once that server accepts the message, it owns that message and becomes responsible for delivering it. Along the way, that server may legitimately modify the email.

For example, many companies add “[External]” to the subject line of incoming messages or rewrite links so security appliances can scan them. Mailing lists often add footers. The problem is that these perfectly legitimate changes can break DKIM validation.

DKIM2 introduces a way for each system handling the message to record exactly what changes it made. The next server can verify those changes, reconstruct the original message if necessary, and confirm that the authentication is still valid. Instead of relying on trust, it provides verifiable proof of what happened to the message.

2. Backscatter: When a spammer spoofs your email address as the sender, receiving mail servers mistakenly send automated bounce messages to you when the spam is rejected.

Historically, if a server accepted an email but later couldn’t deliver it, it generated a bounce message. Attackers exploited that behavior by forging sender addresses, causing thousands of bounce messages to flood innocent recipients.

To prevent that abuse, many systems today either accept or reject mail during the SMTP transaction, often within milliseconds. That leaves very little time to make accurate delivery decisions.

DKIM2 creates an authenticated chain showing every system that handled the message. If a server later determines that a message shouldn’t be delivered, it can return it along that verified path rather than generating unauthenticated backscatter.

That authenticated delivery path helps solve one of email’s long-standing problems while giving mail systems greater flexibility to make accurate delivery decisions after accepting a message.

Stripo: How do you see the balance today between technical deliverability requirements and the actual subscriber experience? Do you think the industry sometimes over-focuses on infrastructure while underestimating the human side of email?

Laura: Absolutely. I think the industry over-focuses on infrastructure because it’s the easy part. Sending a technically correct email isn’t difficult. It can be time-consuming to make sure everything is configured properly, but anyone with a bit of training can identify infrastructure issues. There are countless online tools that will tell you if your SPF record is broken or whether you need to rotate a DKIM key.

From a deliverability perspective, those are straightforward problems. If someone comes to me with an infrastructure issue, I’ll usually tell them exactly what needs fixing. Sometimes I fix it myself, and sometimes I send them to someone else who specializes in that work. Those aren’t particularly interesting problems.

The human side is where things become difficult. The real challenge is understanding what your audience wants, how they’re interacting with your emails, and how to send messages they’re genuinely happy to receive.

Laura Atkins

Laura Atkins,

Co-founder of Word to the Wise and Deliverability Consultant; Co-founder of Women of Email.

That takes time. Most of my consulting engagements last anywhere from six weeks to six months because we have to analyze customer data, understand the audience, and evaluate the overall performance of the email program. There’s no quick fix.

That’s one reason so many people focus on technical deliverability. Infrastructure problems are easy to scale. You can move clients through that process quickly. Understanding people doesn’t work that way.

My scientific background also influences how I think about this. Even something as common as A/B testing is often done poorly. People run tests without thinking about proper randomization or statistical significance. They’ll tell me a website calculated a p-value for them, but they don’t know how.

As a scientist, that drives me a little crazy. Good testing requires good methodology.

Stripo: One interesting paradox today is that many legitimate marketers are fully authenticated and technically compliant, yet still struggle with inbox placement. What are the most common reasons technically “correct” senders continue losing inbox trust?

Laura: The biggest reason is that they don’t pay attention to feedback from their recipients. The inbox belongs to the recipient, and mailbox providers make filtering decisions on behalf of that recipient. Trust is built between the sender and the individual recipient.

Sometimes marketers simply ignore where customers are in their journey.

Imagine you’re selling hardware with a three-year lifespan, but you’re sending weekly promotional emails encouraging customers to replace something they just bought. Why would someone want to receive 150 emails trying to sell them a product they won’t need for years?

That’s the difference between mass marketing and individual marketing.

A television ad might eventually reach someone who’s ready to buy. Email is different because you’re speaking to people you already know something about. If the message isn’t relevant to their current situation, it doesn’t provide value, and recipients will eventually stop engaging with it.

What’s next for deliverability? Trust signals, DKIM2, and the role of AI

Stripo: From your perspective, how will deliverability evolve over the next five years if authentication becomes fully standardized across the industry? What new trust signals do you think mailbox providers will prioritize next?

Laura: I think the important areas of development will be the following:

1. Increasingly individualized inbox experiences.

We’ve already reached a point where traditional inbox placement tools are becoming much less reliable because mailbox providers personalize delivery for individual recipients. Test email addresses don’t behave like real people. They rarely send emails, don’t build normal engagement histories, and don’t generate the same behavioral signals as actual users. I’ve seen clients whose inbox testing tools reported either 100% inbox placement or 100% spam placement, while their campaigns were generating 30% open rates. Clearly, the testing wasn’t reflecting reality.

2. New ways to measure deliverability. 

The tools and processes we’ve relied on for the past 15 or 20 years simply aren’t keeping up with how email has evolved. We need new ways to measure deliverability. I don’t know exactly what those solutions will look like, but it’s an area where the industry needs real innovation. The good news is that marketers probably won’t need to make major changes when DKIM2 is introduced. Most of the work will happen behind the scenes through ESPs and infrastructure providers. In many cases, marketers may only need to publish a new DKIM key in DNS.

3. Senders may begin receiving more meaningful feedback when messages are rejected as a consequence of DKIM2. 

Instead of silently dropping messages or placing them in spam folders, mailbox providers may increasingly return authenticated bounce responses. That could provide much better insight into which recipients or mailbox providers don’t want particular messages, giving marketers more useful feedback than they have today.

4. Mailbox providers will keep placing a strong emphasis on domain reputation. 

We may also see reputation signals expand to include practices like affiliate sending.

5. Cold email, particularly in business inboxes, will receive much more attention. 

Over the past few years, cold outreach has become an entire industry, and many professionals now receive dozens of unsolicited sales emails every day. I wouldn’t be surprised if mailbox providers eventually begin treating the use of certain cold email infrastructure or automated warm-up tools as negative trust signals. If a brand relies on those systems, mailbox providers may increasingly conclude that it’s engaging in unwanted outreach and filter those messages more aggressively.

Stripo: AI-generated marketing content has become impossible to ignore. Do you see parallels between the spam problems of the early internet and what’s happening now with AI-generated email?

Laura: I’m very skeptical of AI. What strikes me most isn’t the technology itself but the messaging around it. Companies and individuals are constantly told that AI is inevitable, that they’ll be left behind if they don’t adopt it.

That mindset reminds me of the early spam era, when people argued that if you weren’t buying massive email lists or scraping addresses from Usenet, you’d lose to competitors who were.

Laura Atkins

Laura Atkins,

Co-founder of Word to the Wise and Deliverability Consultant; Co-founder of Women of Email.

The email industry proved that it wasn’t true. Mailbox providers built better filters, and the ecosystem adapted. I think AI is appealing to marketers who still approach email as mass marketing. But I don’t believe it will be as transformative for email marketing as many people expect.

We’re also reaching the point where AI is becoming expensive. Some companies have even rehired employees after discovering that AI costs more than having people do the work.

Laura Atkins

Laura Atkins,

Co-founder of Word to the Wise and Deliverability Consultant; Co-founder of Women of Email.

So I’m waiting to see where things settle. I don’t think AI will be as effective in marketing as many people hope.

Stripo: Do you think mailbox providers will begin treating AI-driven email differently in the future?

Laura: That’s a fascinating question, and I don’t think we know the answer yet. Companies like Google are investing heavily in AI and integrating it into almost everything they build, including email experiences and filtering systems.

At the same time, the teams responsible for email also pay close attention to what recipients actually want. 

So we’re in an interesting moment where two forces are competing. On one side are companies investing enormous amounts of money in AI. On the other side are subscribers whose preferences ultimately determine whether those systems succeed.

Right now, it’s too early to know which force will have the stronger influence. I’m beginning to see signs that recipients may ultimately win, but there’s still a tremendous amount of investment pushing AI forward.

Stripo: You’ve spent your career balancing technical expertise with a strong recipient-first philosophy. As deliverability becomes more automated and AI becomes more embedded in marketing workflows, what do you think the industry risks losing if it becomes too focused on scale and efficiency over human relevance?

Laura: I'm resisting that future with everything I have because I believe email is fundamentally about people. I’ve built my career through email. I’ve gotten jobs through email. I’ve met some of my closest friends through email. I even started my company with my business partner after we met through email.

To me, email has always been about human connection. I think marketers have always struggled with one simple fact: an email address isn’t just an address. It’s a person.

The more we focus on scale and efficiency, the easier it becomes to forget that. If a message doesn’t resonate with the recipient, they’ll delete it or mark it as spam. Scaling irrelevant communication doesn’t solve the problem: it simply creates more irrelevant communication.

I also think marketers sometimes become too focused on campaign metrics. We celebrate open rates, click rates, and other performance numbers, but the more important question is what actually happened afterward. Did someone visit your website? Did they buy something? Did the email create value?

It’s also worth remembering that marketing is only one part of the email ecosystem. Newsletters, publications, and creator platforms like Substack or Beehiiv serve different purposes. Those emails aren’t always trying to generate clicks or conversions. They’re simply people communicating with other people.

I think those kinds of emails will continue to grow because they feel authentic. They reflect genuine human communication. Email isn’t going away. But marketers need to remember that behind every email address is a real person, not just another metric or another step in an automated journey.

Wrapping up

As authentication standards mature and mailbox providers continue refining how they evaluate email, deliverability is becoming less about technical compliance and more about earning recipient trust. Throughout our conversation, Laura reminded us that behind every email address is a real person, and that’s the perspective both marketers and mailbox providers need to keep in mind.

Thank you, Laura, for sharing your expertise and giving us a glimpse into where deliverability is headed. Your insights are a valuable reminder that, while technologies like DKIM2 and AI will continue to shape the email ecosystem, the most successful email programs will always be those that create genuine value for the people receiving them.

Was this article helpful?
Tell us your thoughts
Thanks for your feedback!
0 comments

Stripo editor performs its best on desktop

How about we send you a reminder to test Stripo later on your computer?

I still want to test on mobile