Email Metrics That Matter: From Opens to Orders
Opens flatter, clicks inform, and orders decide. This guide ranks the email metrics for a small business - what each can and cannot tell you, how privacy changes broke the open rate, and the short dashboard that ties mail to revenue.
The short answer
Judge email by the chain that ends in money: delivered, clicked, converted, revenue - with unsubscribes and complaints counted as costs on the same page. Opens sit at the chain's weak end and have grown weaker: privacy features on major mail platforms now pre-fetch messages, registering opens that no human made, so the open rate is weather, not climate. Click rate and click-to-order rate are the working gauges of whether mail is wanted and whether it lands well, revenue per campaign and per flow is the verdict, and revenue per subscriber across a season is the programme's honest grade. Everything else is diagnostics, useful in its place and dangerous as a headline.
Opens: what broke, what remains usable
When a platform opens mail in advance to scan it, the tracking pixel fires without a reader, inflating opens unevenly across audiences and making historical comparisons meaningless - an open-rate rise may be a software rollout, not a better subject line. What remains usable: opens as a rough deliverability alarm (a collapse still signals trouble), and relative opens within one send across devices or segments, read cautiously. What is not usable: open rate as a campaign score, a flow target, or evidence in a subject-line test. The subject line craft of subject lines and preview text is judged downstream instead, on clicks the pre-fetch cannot fake.
Clicks and click-to-order: the working pair
Click rate shows the email persuaded its readers; click-to-order (or click-to-conversion for an enquiry business) shows the landing kept the promise, the split the continuity guide, email-to-page continuity, builds its audit on. Together they diagnose nearly everything: strong clicks and weak conversion indict the page or the offer's terms; weak clicks with a healthy history indict the email's relevance to its segment, which points back to segmentation. Track the pair per campaign and per flow email, in the same log row, every time - the discipline is the dashboard; decoration is optional.
Revenue attribution without self-deception
Attribute revenue with a stated, steady rule - the platform's click-based window is acceptable - and never change the rule to flatter a quarter. Report two revenue views: per campaign, which ranks sends against each other, and per flow across a season, which shows the automated programme's real contribution (welcome, cart, post-purchase) against the campaigns that get the attention. Add revenue per thousand sent for comparing across list growth, and keep the costs in the same view: discounts given in the mail, and the platform's price. The foundations bench's warning applies unchanged: a flow can appear to create revenue it merely collected early, notably the discount cart sequence, so read flow revenue beside the store's overall rate, not alone.
Unsubscribes and complaints are data, not shame
Every campaign's unsubscribe count prices its relevance: a steady low rate is the cost of mailing at all, a spike names the send that broke the promise made at capture, and the campaign that caused it should be identifiable in the log and discussed without flinching. Spam complaints are the expensive version - few in number, heavy in deliverability consequence - and any complaint cluster traces to a capture promise or a segment error worth fixing at source, in the spirit of list quality over quantity. A programme that hides these numbers from itself will repeat their causes; one that reads them monthly rarely has to read large ones.
Score flows on their own jobs
Do not rank flows against campaigns on one table. The welcome flow is scored on second opens, second visits and first purchases across its cohort's season. The cart flow on recovered orders per hundred carts, beside the store's checkout completion, per the cart guide. Post-purchase on repeat purchase and support outcomes. Win-back on retained subscribers per hundred mailed, with the sunset's list-health effect counted as its success rather than its failure. Each flow's row in the dashboard states its job in words next to its numbers, so a future reader cannot misgrade a service flow for a quiet sales month.
The one-page dashboard, reviewed monthly
The whole discipline fits on one page, reviewed in one sitting each month: sends and delivered; clicks and click-to-order per campaign; orders and revenue per campaign and per flow; unsubscribes and complaints, named by send; list movement - new quality signups, sunsets, engaged share. Trends across the page, quarter over quarter, are the programme's strategy input: which capture points earn, which flows pay, which segments respond. The parent guide, email capture and flows, closes its bench here deliberately: a programme that can see itself clearly needs few heroics, because its next improvement is usually legible in its own numbers.
Sources and evidence on the record
- Google, Google Analytics support - reports, events, key events and funnel documentation.
- CXL, CXL blog - practitioner writing on testing method, research process and optimisation programme discipline.
- Microsoft, Microsoft Clarity documentation - heatmaps, session recordings and frustration signals such as rage clicks.
Evidence basis: Method guide. Any figure quoted is attributed to a named published source listed below; worked examples are labelled illustrations with round numbers, not measurements. Applies when: Monthly email programme review, and every campaign post-mortem