Email Personalization Stats That Actually Matter (Not the $36 ROI Story)

Most email personalization stats you see online are telling a flattering story about a channel that is quietly losing ground. The headline number, $36 return for every $1 spent, comes from Litmus’s State of Email research. It is true. It is also measuring the wrong thing. That $1 denominator is your ESP platform cost, not subscriber attention, not creative effort, and not the actual cost of acquiring and retaining each person on your list. When you flip the lens to revenue per subscriber, the picture changes fast.

This post collects the email personalization stats that matter to ecommerce marketers who send real volume every day. Not the stats that make your board deck look good. The ones that explain why your per-subscriber revenue is going the wrong direction, and where the actual opportunity sits. We update this piece regularly. If you are looking for the 2024 version, it is still available for reference.

Email Personalization Stats That Challenge the $36 Headline

1. The $36 ROI stat measures platform cost, not subscriber value. Litmus calculates email ROI by dividing total email-attributed revenue by ESP subscription fees. That math rewards you for sending more, even when each additional send produces less. If you double your send volume and revenue stays flat, your ROI per dollar of platform spend stays roughly the same. But each subscriber on your list just became half as productive. That distinction matters.

2. Real per-subscriber revenue in retail email has fallen roughly 35% since 2018, from about $51 to $33 in inflation-adjusted dollars. In nominal terms the drop is around 20% (from $41 to $33). But Bureau of Labor Statistics CPI-U data shows roughly 22% cumulative inflation over that period. Once you account for what a dollar actually buys, a subscriber today is worth about a third less than one in 2018. No industry benchmark report adjusts for this, which systematically understates the decline.

Frequency and Measurement Are Broken

3. Retailers sent 63% more emails per subscriber in 2024 than in 2016 (roughly 155 sends per year vs. 95). Volume went up. Revenue per subscriber went down. The industry’s default response to declining engagement has been to send more, which accelerates the fatigue cycle. More emails at lower per-send yield is the definition of a diminishing returns problem.

4. Apple Mail Privacy Protection pre-loads tracking pixels on the vast majority of iPhones, inflating open rates regardless of whether anyone reads the email. Litmus data showed that within six months of MPP launching, aggregate open rates jumped 18 percentage points (from 22.6% to 40.5%) across roughly 2 billion messages. Click rates did not move at all. Open rates and actual engagement are now measuring completely different things. If your email personalization stats still lean on open rate as a performance signal, you are working with corrupted data.

5. Batch email Revenue Per Mille (RPM) fell approximately 43% from its 2018 peak, from around $227 per thousand sends to roughly $130 per thousand by 2024. RPM is one of the better per-send performance metrics because it ties directly to revenue. A 43% decline over six years is not a blip. It is a structural trend that more sends cannot fix. (For more on measuring email block performance, see our guide on email block analytics.)

The Automation Gap and the 95% Problem

6. Automated emails convert at roughly 25x the rate of batch campaign emails. Klaviyo’s benchmark data shows automated flows producing click rates of 5.58% versus 1.69% for campaigns, with placed-order rates running 13x higher. Abandoned cart flows average 3.33% conversion, with top performers hitting 7.69%. The gap between automation and batch is not subtle.

7. That small slice of automated email volume, roughly 2% of total sends, drives about 41% of all email-driven orders. Omnisend’s ecommerce benchmark data makes this concentration impossible to ignore. Two percent of your sends doing 41% of the work tells you something about the other 98%.

8. Roughly 95% of retail email volume, the daily broadcast sends, ships with no content-level personalization. Most programs segment into two or three audience groups. That is not the same as personalizing the content each person actually sees. The email personalization statistics on automation performance are impressive. But they apply to a tiny sliver of your total volume. The broadcast send is where the untapped opportunity lives. (We wrote a full breakdown of this in our piece on broadcast email personalization.)

Email Personalization Stats That Show What Actually Works

9. The average email click-to-conversion rate is approximately 2.5% across the industry. Click-to-conversion (CTC) measures the percentage of people who click on a specific content block and then purchase within a defined attribution window. It is a harder metric than click rate because it connects the click to actual revenue. The 2.5% baseline is what the whole email typically produces. (More on why this metric matters in our post on how to measure ROI email performance.)

10. Personalized Smart Banner and Smart Kicker content averages 18.3% click-to-conversion, a 7x improvement over that 2.5% baseline. This is Zembula platform data across our customer base. The gap is large because the content is conditional: it renders at the moment of open based on that subscriber’s behavior, preferences, and real-time signals. A banner showing your abandoned cart item with a price drop performs differently than a static hero image that is the same for everyone.

11. Abandoned cart Smart Banner combinations typically reach 15 to 25% CTC. Cart abandonment is the highest-performing use case because the intent signal is strong and the content is immediately relevant. But it is worth noting that loyalty point reminders, back-in-stock alerts, and browse abandonment also consistently outperform static content. The pattern is not limited to one use case.

Why Static Content Fails and Signals Matter

12. Nielsen Norman Group eyetracking research confirms that users learn to ignore static content in predictable screen positions. Banner blindness is one of the most replicated findings in UX research. It applies directly to email: the same promotional banner in the same position, send after send, becomes invisible. Varying the content per recipient per send is the evidence-backed counter to this pattern.

13. Image-based personalization rendered with brand fonts and photography converts 5 to 7% higher than the same logic rendered as generic HTML text. This is a Zembula platform observation. Quality of presentation matters. When personalized content looks like the brand, subscribers engage with it differently than when it looks like a system-generated fallback. Brand consistency is not just an aesthetic preference. It moves conversion numbers.

14. The highest-converting personalization stacks behavior, profile data, and urgency into a single content block. Single-signal personalization (just a first name, just a product recommendation) consistently underperforms multi-signal combinations. An example: “Sarah, the jacket you browsed is now 20% off, and your 500 loyalty points expire Friday.” That message combines browse behavior, a price trigger, and a time constraint. Each signal alone is fine. Together they compound. McKinsey research supports this: companies that excel at personalization generate 40% more revenue from those activities than average performers.

The Math on the Broadcast Personalization Opportunity

15. A 10 to 17% revenue lift on the 95% of broadcast email that currently has no personalization adds $3 to $6 per subscriber per year. That range is enough to reverse the 35% real decline since 2018. The math is simple. If your list has 500,000 subscribers, we are talking about $1.5 million to $3 million in recovered and new revenue annually. This is not hypothetical. It is arithmetic applied to the gap between what automated email proves is possible per send and what batch sends currently deliver.

The email personalization stats on automation prove that relevant, timely, behavioral content converts at dramatically higher rates. The problem is not that the approach does not work. The problem is that it has been trapped in 2% of volume. Bringing that same personalization logic to broadcast, through tools like Smart Banners and Smart Kickers that render at the moment of open, is how you close the gap without rebuilding your email program from scratch. (Zembula’s approach to customer return on spend is built around making this math work for every account.)

Key Takeaways

• The $36 ROI stat is real but misleading. It measures platform cost, not subscriber productivity. Per-subscriber revenue is down 35% in real terms since 2018.
• Open rates are broken. Apple Mail Privacy Protection has inflated open rates by 18+ points without any change in actual engagement. Use click-to-conversion instead.
• Automation proves the model, but it only covers 2% of volume. Automated emails drive 41% of orders. The other 98% of sends are where the real upside sits.
• 95% of broadcast email has no content-level personalization. Segmentation alone is not personalization. Each subscriber should see different content based on their behavior and profile.
• Personalized content in broadcast email averages 18.3% CTC, versus 2.5% for the email baseline. That is a 7x gap, measured at the block level.
• Multi-signal personalization outperforms single-signal. Stack behavior, profile data, and urgency for the highest conversion rates.
• The broadcast personalization opportunity is worth $3 to $6 per subscriber per year. For a 500K list, that is $1.5M to $3M in revenue most programs are currently leaving behind.