Florin Armasu, CEO and founder of Data Innovation and creator of the Sendability platform, has spent fifteen years building email analytics for clients including Nestle and Brown-Forman. In a conversation with Sophie Steffen for The Open Rate Club, he dismantled the metrics hierarchy that most email marketers operate under and replaced it with something considerably more useful – provided you are willing to build the dashboard.
The Death of the Open Rate
Open rates, as a reliable metric, are finished. This is not a prediction or a hot take. It is a description of what Apple Mail Privacy Protection did to email analytics starting in October 2022. Every email opened on an Apple device is pre-cached by Apple’s servers, triggering a pixel fire that registers as an “open” regardless of whether the recipient looked at the message, thought about the message, or was even awake when the message arrived. Companies are reporting average open rates of 60% and smiling about it. They should not be smiling. A meaningful percentage of those opens were generated by a server in Cupertino performing the digital equivalent of stamping every letter “delivered and read” before the postman has knocked.
The distortion is not limited to iCloud. Any Gmail address read on an iPhone generates an Apple-cached open. Any Hotmail address read on an iPad generates an Apple-cached open. The infection is device-level, not provider-level, which means it corrupts your data across every mailbox provider proportional to Apple device market share in your audience – and in most Western markets, that share is not small.
Click-through rate – the metric many marketers retreated to as a proxy for engagement – has declined accordingly, not because clicks went down but because the denominator ballooned. When opens are inflated by 40%, your click-through rate drops by 40% even if actual engagement is unchanged. You are now measuring a ratio of real clicks to fictional opens, which is approximately as useful as dividing your revenue by the number of unicorns in your office.
Revenue Per Email Sent
The first metric Armasu actually cares about is revenue per email sent. The concept is simple. The implementation is not, because the attribution problem – connecting a specific email send to a specific revenue event – has never been properly solved, and email sits in the awkward middle of a multi-channel attribution landscape where every channel claims credit for the same conversion.
But the fact that attribution is imperfect is not an argument for ignoring it entirely. Most companies, Armasu notes, track the cost side of email meticulously. They know what they pay per contact stored or per thousand emails sent. They do not track what each email generates in revenue. The result is that the email channel looks like a cost center on every spreadsheet, which makes it perpetually vulnerable to budget cuts and strategic deprioritization – even when it is, in fact, the highest-ROI channel in the marketing mix.
The fix is not a perfect attribution model. It is any attribution model. Even a simple last-touch approach – crediting the email if the recipient clicked through and converted within a defined window – provides a floor estimate of email revenue that is more useful than the current standard, which is no estimate at all.
The Golden Metric: Active Database Growth Rate
When pressed for a single metric – the one number that tells you more about the health of your email program than any other – Armasu did not hesitate. Active database growth rate: the percentage of your total subscriber base that is actively engaging with your content, measured over time.
This is not the same as list growth, which measures how many new subscribers you are adding. You can grow your list by 10,000 contacts a month and still have a shrinking active database if your existing subscribers are disengaging faster than your new ones are engaging. List growth tells you whether your acquisition is working. Active database growth rate tells you whether your entire program is working – acquisition, content, deliverability, and retention combined into a single directional signal.
“If your database is decreasing, that means that you are overly conservative. You don’t try to reactivate so many users, and you are very precise. You deliver perfectly a very small quantity of emails, but eventually your database will gonna go to zero.”
The metric has a brutally clarifying effect on strategic decisions. A shrinking active database means one of two things: either your content is not reaching people (a deliverability problem) or your content is reaching people and they do not care (a relevance problem). Either way, the trend line tells you that your current approach is not sustainable. You are running a program that is consuming its own audience, and the logical endpoint – as Armasu puts it with the sort of understatement that suggests he has watched it happen – is zero.
Complaint Rate: The Provider-Level Problem
Complaint rate matters, but only if you measure it correctly – which means measuring it by mailbox provider, not in aggregate. The reason is architectural: different providers expose complaint data differently, and treating them as interchangeable produces averages that obscure the actual problems.
Gmail reports complaint rate at the campaign level through Google Postmaster Tools. You can see that a campaign triggered complaints, but you cannot identify which individual recipients complained. You cannot suppress them. You cannot remove them. You can only adjust your targeting and content strategy at the campaign level and hope the next send performs better. Gmail’s threshold for “dangerous” is 0.3%, which sounds restrictive until you learn that some ESPs flag 0.15% as “very high” – a complaint rate that Gmail itself does not consider alarming. The ESP, in Armasu’s view, is being more Catholic than the Pope.
Yahoo and Hotmail, by contrast, report complaints at the individual level via feedback loops. When a recipient hits the spam button, you get a notification with their address. You can suppress them immediately. You can remove them from your list. The feedback is actionable in a way that Gmail’s aggregated data is not, which means your list hygiene practices for Yahoo and Hotmail should be structurally different from your practices for Gmail – a distinction that almost no one makes because almost no one builds provider-level suppression logic.
The Health Score: One Number to Rule Them All
For organizations that find per-metric analysis overwhelming – or, more charitably, for organizations that want a single dashboard widget that tells them whether to panic – Armasu has built what he calls a health score. The concept exists in some commercial platforms (SparkPost offers a version), but Armasu’s implementation aggregates click rate, open rate, bounce rate, complaint rate, and engagement signals into a single composite number that represents overall sender reputation.
The health score is not a metric in itself. It is a weighted index – a way of compressing multiple signals into a single directional indicator so that someone who does not have time to analyze six provider-level dashboards can glance at one number and know whether things are getting better or worse. The weights are the key. Open rate, given its current unreliability, should be weighted lower than click rate. Complaint rate, given its direct impact on reputation, should be weighted higher than bounce rate. The specific weights depend on the sender’s audience composition, geographic distribution, and Apple device penetration.
The value is not in the number itself but in the trend. A health score that has declined for three consecutive months tells you that something structural is changing, even if no individual metric has crossed an alarm threshold. It is the deliverability equivalent of a check-engine light – it does not tell you what is wrong, but it tells you that something is wrong before you find yourself stranded on the side of the road wondering why Hotmail stopped accepting your emails.
For the full interview breakdown, see our complete Expert Insight with Florin Armasu.
Tools Mentioned in the Interview
The following tools and platforms were referenced during this conversation.
Related Content
Email Marketing Execution Behind the Scenes: Expert Insights with Florin Armasu
Florin Armasu, CEO of Data Innovation and creator of Sendability, unpacks the open source stack, deliverability disasters, and metrics that actually matter after 15 years of sending emails for brands like Nestle and Brown-Forman.
Written by Sophie SteffenThe Open Source Email Stack That Powers Real Campaigns
Florin Armasu runs Mautic, KumoMTA, and Qmail -- a 1998 mail transfer agent -- in production for enterprise clients, and his stack outperforms most commercial ESPs.
Written by Sophie SteffenSilent Deliverability Failures That Cost Millions
A major FMCG brand got blocklisted at Hotmail, risking 5 million euros in lost revenue. The worst part: they did not notice until someone checked the per-provider metrics.
Written by Sophie Steffen