I often hear marketers ask what email metric they should pay the most attention to, or which one best represents the condition of their email program.

That’s easy. The answer of course is your Delivery Rate, or the percentage of your messages that are getting the greenlight from ISPs and email administrators. Nobody reads what they don’t receive, so delivery rate is the lever with the greatest purchase. Note, however, that the number of delivered messages is not the number of people who actually receive your emails. Because of client-side spam filters and other rules that sequester some delivered messages outside of the inbox, about 1 in 5 messages that pass through to your subscriber never show up where they can be read. So maybe Delivery Rate doesn’t tell you everything you need.

In that case you need to focus on your Open Rate, definitely. Open rate is a proxy for read rate in that it lets you know how many times your message has been visibly displayed in someone’s reader or preview pane. Of course, being visible is not the same thing as getting read, so as proxies go this one is not able to boast six sigma accuracy. When your subscribers click on your subject line so that your message fills the preview pane or the reader briefly – even if their intention is to delete your message or filter it off to some other folder – that message may still be recorded as opened, even though it is not read. Open Rate then has some issues of its own.

What you really want to measure is engagement, right? Emails are supposed to drive clicks so monitoring the Click Rate of your messages is a great way to see how well you’re engaging your audience. There is one small problem with this metric though. Close to 40% of opens are now occurring on mobile devices. Opening a message on a phone can have profound implications on what actions someone takes. On the one hand, smaller screens and slower loading make the web experience on mobile devices less satisfying, prompting many people who read messages on their phones to hold off on clicking until they see the message again on a bigger screen. And a lot of people who do click end up fat-fingering the tiny links or buttons and go to someplace they didn’t mean to. So you may have more engagement than your clicks suggest, and your audience may be engaged in topics other than what they actually clicked on. Also, click rate is calculated by the number of clicks over the number of delivered emails. If you could boost your deliverability and your open rate, you’d have more clicks, right? Click Rate is useful, but no panacea.

If the focus is really about creating engaging content, the critical metric is Click-to-Open Rate, or the percentage of people who open your email and then click on it. This metric gives you much better insight into how well your email content creates engagement and drives action as it only measures responses from the people who read your email. Except that “opens” do not necessarily translate into people who read your email, and also because of that fudge factor about clicks I just talked about too.

OK, maybe we’re looking at this all wrong. We’ve been focusing on what we’re doing right with email, but what if we look instead at evidence that we’ve done something wrong? In that case, email’s ultimate metric may be the Unsubscribe Rate. The difficulty of earned media works both ways. Yes, it is harder to grow your list if you are earning permission from everyone on it. But you also have their permission until they say you don’t. Unsubscribe Rate then tells us how many people we’ve failed with, in a per-message percentage that is easy to track and trend. Even if we are not driving engagement and clicks, if we haven’t lost people to unsubscribe, we get another bite at the apple. But this treatment of unsubscribes is muddied by inactive subscribers – those people who technically haven’t unsubscribed but maybe have abandoned an email address or banished our emails to some filtered folder way down the list. We still have their permission, but not their attention. So even Unsubscribe Rate doesn’t give us a complete picture.

What’s left? ROI is flawed because the real expense in email is not what you pay to send, but the years you have put into earning your subscribers’ permission and attention. Lifetime Customer Value is a great theory, but with so many other contact points clouding attribution it is almost impossible to quantify what percentage of value comes from which channel. Metrics trended over time like clicks-per-subscriber-per-year or a homemade engagement score based on a number of metrics tighten the focus a bit, but even these leave off after the inbox, when the real measure is what subscribers do after the email has compelled a click.

Email is often considered a narrow marketing channel, definable by simple direct response metrics. That simply isn’t true. Understanding how well your program is working takes a big picture view with a conversational fluency in each of the metrics available to evaluate it – both in what they tell you and what they don’t.