The issue report

Every sent issue has a report, and it opens where the composer used to be. A draft or a scheduled issue has no report, because there is nothing to report on yet: the report replaces the composer the moment the issue is out. You can reach it from the issues list, or from any row in recent issues on your dashboard.

The funnel, and what each share is of

The funnel runs sent, delivered, opened, clicked, and each percentage is of the stage above it rather than of the top. “This share of the people it reached opened it” is the sentence a curator reasons with, and measuring every stage against the number sent hides a delivery problem inside an engagement figure.

Underneath the four stages are the three figures the issue cost you: unsubscribes, with their share of the messages delivered, bounces and complaints. What to make of those, and when a number is worth acting on, is in reading delivery numbers.

Why some did not arrive

When something failed, the report says so in your sending provider’s own words, grouped, with the commonest reason first. When nothing failed, the section is not there at all, because a permanent empty heading invites worry where there is none.

The provider’s wording is the useful part. “Email address is not verified” means a sending account still in Amazon’s sandbox, and no amount of general advice about list quality gets anybody to that answer. A failure is also not a bounce: a failure means the message was never accepted in the first place, while a bounce comes back afterwards and is counted separately.

The click curve

The curve plots clicks per hour across the first 48 hours after the send. Every hour is drawn even when it saw nothing, so a quiet night reads as a flat stretch rather than as a gap in the data.

Most of a newsletter’s life happens in the first day, which is why the window is that short. Clicks that arrive later are not lost: they count in every total on the page and in the link table below, and they are off the end of this chart, no more than that. Scanner traffic is excluded from the curve, as it is from every figure here.

Devices and clients

The device and client breakdowns are built from the clicks, not the opens, so they describe the people who actually followed a link. Every device and client with a share is listed, with its count.

Both are best effort, and they say so. Issue Mint answers Unknown rather than guessing, because a plausible wrong answer on a stats page is worse than an honest gap: several mail clients report the platform’s webview instead of themselves, and a privacy relay reports a generic desktop browser for a phone. Apple Mail, Outlook, Gmail, Yahoo Mail and Thunderbird are named confidently. So are Edge, Chrome, Firefox and Safari, which turn up when somebody clicks from your archive or from a copy that was forwarded to them.

Every link

The link table lists every tracked link in the issue with three columns: unique clickers, total clicks, and share of clickers. Sort it by clicking any column heading.

Share of clickers is this link’s unique clickers as a share of everyone who clicked anything in the issue, so it answers “of the people who engaged, how many went here”. The column can add up to well over a hundred per cent, because one reader may follow five links. Curated picks show their own title, and anything from your intro or your footer is labelled as such rather than dressed up as a pick.

Archive clicks appear in the total column and never in the unique column, because a web click has nobody attached to it. How click tracking works explains that split. An issue sent before click tracking existed keeps its direct links, so its funnel is intact and its table is empty.

The CSV export

Export CSV gives you the same table as a file: title, host, full destination URL, whether the link was a curated pick, unique clicks, total clicks and share of clickers, named after the newsletter and the issue number.

This is the file to send a sponsor. It contains the destination URL in full, so the row for their link is unambiguous, and it is the evidence a sponsor asks for when they are deciding whether to book again. There is more on what to send them in selling sponsorship.