Checking a broadcast's results from your phone, without spiralling

You sent it at eight in the evening. By ten past eight you have checked twice. By nine you have a system: dashboard, back, dashboard, back, with a detour through the app switcher to see whether anybody has replied. Nothing you look at in those two hours will change what you do next, and you look anyway, because a send is the one thing in this job that produces suspense. The point of this piece is not to tell you to stop. It is to give the checking a shape, so that ten minutes of it does the work that three hours of it currently does badly.

Why the first two hours tell you so little

Email does not arrive at once. It lands over hours, and it gets read over days, in gaps between other things. Somebody who buys from your Tuesday email on Thursday morning is a completely normal buyer, and they are invisible on Tuesday night. Judging a send at the two-hour mark is like judging a dinner party by who has arrived by seven.

Then there is the open rate, which deserves its own warning. Mail privacy features load images on behalf of the reader, which registers as an open that no human performed. Some of your opens are real, some are machines, and the ratio differs by which mail apps your readers use, which you do not know. It is not a useless number — a collapse across several sends means something — but as a live measure on the night, it is close to noise.

The genuinely hard part is emotional rather than technical. A quiet evening after a send you worked on for two days feels like a verdict on the work. It usually is not; it is a verdict on a Tuesday. But knowing that does not stop the feeling, which is exactly why a routine helps more than resolve does.

Refreshing, or checking

Here is the difference between the loop most people fall into and a routine that takes ten minutes in total.

The two checks

Both of these are things the iOS app does well, because both are about orders and messages rather than about charts. The first check is a safety check. The second is the actual reading.

An evening that looks like nothing and is not

Take a send to 500 people at eight in the evening, promoting a £35 workshop recording. By ten, three orders. That reads as a disappointment, and on the night it is hard to read as anything else. There are also two messages: one asking whether the recording includes the templates, and one saying the link opened a blank page on an in-app browser.

By Friday the picture is different in the way it usually is, and per-product analytics show it plainly. More orders arrived over Wednesday and Thursday, from people who read the email in a queue and came back later. The templates question turns out to be the same question two other buyers asked last month, which means it belongs in the product description rather than in your inbox. The blank-page report leads to a check that the link behaves inside social in-app browsers. None of that was visible at ten o'clock, and none of it required the evening you spent refreshing. The three orders were the least informative thing that happened.

The mistake: acting on two hours of data

The classic sequence is a quiet evening, a conclusion that the offer is wrong, and an unplanned discount email at eleven the next morning. The people who were going to buy on Thursday now buy at 20% off, the people who bought on Tuesday feel silly, and you have taught the list to wait. If you are on Pro and reaching for a discount code, reach for it as part of a plan you made before the send, not as a reaction to an evening. The same applies to countdowns; a deadline invented in a panic is a deadline your readers will learn to disbelieve.

The second version of the mistake is editing the product while the traffic is live — changing the price, swapping the cover, rewriting the description at nine in the evening. You lose the ability to tell what the send did, because you changed the thing it pointed at halfway through. Make notes, act on Friday.

Two checks, ten minutes, and the rest of the evening back. The version of you that refreshes for three hours is not gathering information; it is managing anxiety with a tool that makes anxiety worse. Look for a broken link at fifteen minutes, read the replies in the morning, and let the numbers settle before you decide anything. They will still be there on Friday, and by then they will mean something.