Improve With Wisdom, Not Hype · Email, CRM & Follow-Through
The basic truth
The open rate was never a good measure, and it’s now barely a measure at all.
Privacy features across major mail clients pre-load images on the recipient’s behalf, which registers as an open whether or not a human ever looked. So the number most email programs are still judged on is partly a record of software behaviour. It moves, it can be charted, and it means considerably less than it appears to.
But the deeper problem isn’t that opens became unreliable. It’s that they were always a measure of the send rather than of the relationship — and in long-cycle B2B, the relationship is the entire point. Email’s job is rarely to produce a purchase this month. It’s to still be there, and still be welcome, when the need finally arrives.
Almost every standard email metric measures the activity. Almost none measure the asset.
What to do
Measure movement, not motion. The useful questions are about what changed:
- Did anyone reply? A reply is worth more than a hundred opens and is almost never reported.
- Did anyone move? Progression to a next stage — a quote requested, a call taken, a spec discussed — is the real signal.
- Did sales find it useful? Ask them. If a rep says a lead “had already read everything,” the program is working, whatever the dashboard says.
- Did people stay? Retention, repeat purchase, and referral are marketing outcomes, not just service ones.
Separate new customers from existing ones. Spend divided by total customers acquired flatters itself when a share of those “acquisitions” were people you already had. Splitting out the cost of genuinely new customers usually changes the picture — and it’s the number that tells you whether paid effort is growing the business or re-touching demand you’d already earned. (Position 10.)
Judge a contact by lifetime value, not by first order. In equipment and industrial markets, the first purchase is frequently the smallest thing a customer ever does — parts, consumables, service, the second machine, and the referral all come later. Optimizing against first-order value systematically undervalues exactly the customers worth having.
Report on the list’s health, not its size. How much of it has done anything in a year, how much has never done anything, how many people came in through a route that produced real customers. A growing list is not an improving one, and count-based reporting can only ever say it got bigger.
The common trap
Measuring the channel instead of the relationship.
Every email tool reports sends, opens, clicks, and list growth — beautifully, automatically, and immediately. Almost none report replies, progression, retention, or whether sales thought any of it helped, because those live in other systems and require someone to go connect them.
So the available numbers become the goal, and they all share one property: they can be improved without anything real improving. Send more often and totals rise. Write a sharper subject line and opens rise. Buy or merge a list and growth looks excellent. None of it touches whether a single additional person is closer to buying, and none of it would detect the slow erosion that eventually empties a list.
The related trap: judging email on last-click attribution. Email’s actual work in a long consideration cycle is maintenance — keeping a relationship warm across months where nothing visibly happens. That work is nearly invisible to attribution models, so a program doing it well can look like it’s underperforming, get cut for something more measurable, and take the pipeline with it a year later when nobody connects the two events.
The tell: your list is growing, your metrics are stable, and sales cannot name a single deal it helped.
The deeper judgment
A list is an asset, and no standard email report is a balance sheet.
Everything the tools show you is a flow — what was sent, what was clicked, this period versus last. None of it describes the stock: how many people would genuinely welcome hearing from you, how many tolerate it, how many have quietly gone. That stock is the thing with value, it changes slowly, and it’s the one number nobody is looking at.
Which is why permission erosion is so easy to miss. Every individual report says fine. The count goes up. The damage only shows in a metric nobody produces — and by the time it’s undeniable, it took years to happen and will take years to repair.
The honest measure of an email program is whether the relationships are in better condition than last year. That’s harder to produce than an open rate and it’s the only number that would have told you anything. It’s also the one that resists the temptation the whole channel creates: because sending is nearly free, the marginal cost of one more email always looks like zero — and it never is. It’s just paid by someone who won’t tell you.
And underneath: do not remove expertise from the moments where trust is being formed. Deciding what counts as success here is a judgment about what the relationship is for — and if that judgment defaults to whatever the platform reports, the platform has quietly defined your marketing strategy as “send more.”
Where this thread goes
You can only measure a relationship you actually maintained.
- They shouldn’t have to explain it twice — carry the context — Connect the Experience · Email, CRM & Follow-Through
- The list is a history, not an audience — Understand the Human · Email, CRM & Follow-Through
- The same instruction problem, in a channel that acts on it faster — Improve With Wisdom, Not Hype · Search & Advertising