The fruit seller's cart
A fruit seller fills his cart from any orchard that will sell to him. He checks every apple the same way: is it red, is it round? Then it goes on the cart. He never cuts one open. Never asks which orchard it came from. Never remembers the doors that told him not to knock again.
One town complains. He moves to the next. Then the next. By the fourth town, people know his cart before he reaches the gate. He never changed the fruit. He only changed the town.
Most outbound runs the same way.
Red and round is not a validation strategy
You buy a list. You run it through a verifier. The verifier says most addresses are valid, so you send. That's the equivalent of checking color and shape. A validation pass tells you an address is syntactically correct and the domain exists. It does not tell you whether that address belongs to a real person who wants your email, whether it's a role inbox like sales@, whether it sits behind a spam trap, or whether the person behind it already unsubscribed from another campaign you ran.
Here's the leak: you have no record of where the contact came from. You don't know if it was scraped, bought, or pulled from a conference list that was three years old. You don't know who already said no. And when your domain reputation dips, you buy a new domain and start over. That's moving to the next town.
Mailbox providers recognise the cart. They see the pattern: same sending behaviour, new domain, no engagement, high complaint rate. The new domain gets burned faster than the old one.
What cutting the apple open looks like
MailHeight is built for the part that happens before you hit send. The original LinkedIn post uses the fruit seller to make the point, and the product implication is simple: provenance should be a first-class field, not an afterthought.
Every contact in MailHeight shows where it came from. Every import lists what was dropped and why. If you upload a file and part of it gets rejected, you see the rejection reasons: bad syntax, dead domain, role inbox, known spam trap, previously unsubscribed. That audit trail matters because "I don't know" is the default answer in most outbound stacks.
Spam traps, role inboxes, and dead domains are blocked before anything is sent. Not flagged, not scored lower, blocked. That prevents the slow reputation bleed that happens when you keep mailing addresses that will never engage.
An unsubscribe or hard bounce blocks that address across every campaign. Not just the campaign that triggered it. That single rule stops the most common form of self-inflicted damage: mailing someone who already told you no, from a different list, three months later.
The town is not the problem
If you find yourself rotating domains, cutting sending volume, or blaming the mailbox provider, stop. The cart is the problem. The fruit is the problem. The addresses are the problem.
A clean list, with provenance and a suppression layer that persists across every send, lets you stay in one town. You can warm the domain gradually, keep engagement high, and let reputation build instead of resetting it.
That's the whole argument. Not a clever sending trick. Just the boring discipline of knowing what you're sending, to whom, and why they're allowed to receive it.
If you want to know where your reputation is leaking, send me a message and tell me how you run your outbound today. I'll tell you honestly. Just don't ask me to add your 100 free accounts from Brevo.