Did Your Email Campaign Work? Build a Report Beyond Opens

To judge an email campaign, start with the action it was meant to produce, verify that action in the system that records it, and use delivery and click data to investigate what happened along the way. An open rate alone cannot tell you whether people read the message, wanted the offer, or became customers.

For a business owner deciding whether to repeat a campaign, the useful report fits on one page: the audience, the intended action, the evidence, the limits, and the next decision. Here is how to build it without turning every send into an analytics project.

Give the campaign one job

Write the success condition before sending. “Promote our new service” is a topic. “Generate suitable requests for a consultation” is an outcome you can evaluate, provided someone reviews the requests.

Choose a primary outcome that matches the message. A workshop invitation might aim for confirmed registrations. A product offer might aim for paid orders. An educational newsletter might aim for visits to a useful guide, with no immediate sales claim. Don't grade all three against the same revenue expectation.

Define what counts, who checks it, and when you will review it. A seven-day response window could be a practical starting point for a particular campaign; it is not an industry benchmark. Choose a window that fits the offer and buying cycle, then record later outcomes separately.

Treat opens and clicks as signals to check

When enabled in Apple Mail, Mail Privacy Protection prevents senders from seeing whether a recipient opened a message. That makes an apparent open a poor foundation for declaring that someone read your message. Apple also notes that this protection does not extend to links or attachments.

Clicks have a different complication: automated security scans and link previews can interact with a message before a person does. Mailchimp documents how bot activity can inflate opens and clicks, and says its filtering identifies most bot activity. Filtering improves the evidence; it does not certify every remaining click as human.

Before comparing two campaigns, record whether bot filtering was applied to both. If the reporting setting changed, a lower number might reflect a cleaner measurement. In Mailchimp, the setting affects reporting across the account, so note it in the report rather than treating it as a setting unique to one campaign.

Our recommendation: keep opens as background context, use filtered unique clickers when available, and make the primary decision from the intended outcome. Avoid choosing a winning offer solely because its open rate is higher.

Compare three sources without merging their totals

The email platform answers what was sent, what was reported as delivered, and what interactions it detected. Check bounces, unsubscribes, and complaints alongside engagement. For this scorecard, treat “delivered” as the platform's delivery measure, not proof that the message reached the inbox or was read. If delivery is the concern, use the separate email authentication planning guide before changing the creative.

Website analytics answers which visits and actions it recorded. Campaign tags can help identify traffic, but a visit is not an inquiry. Google distinguishes a submitted information request from a completed purchase in its recommended GA4 events. The event name is only useful if the implementation records the intended action correctly.

Business records answer whether a request was suitable, a registration was confirmed, or an order was paid. Ask the person responsible for the inbox, booking system, or store to verify those outcomes. An accepted form submission and a qualified inquiry should remain separate counts.

Don't add email clickers to website visitors and call the result “people reached.” These systems count different things. Record each source and its counting unit. Explain any known gaps instead of forcing the totals to match.

Check the path before the send

Use consistent campaign tags on links that lead to your website. Google's URL tagging guidance explains how source, medium, and campaign values identify referred traffic; values are case-sensitive. Keep a shared naming convention so one campaign does not become several rows.

For a hypothetical workshop invitation, a simple naming sheet could use source “customer_newsletter,” medium “email,” campaign “workshop_october,” and content “main_button.” These are illustrative values, not settings to paste into every campaign. Check whether your email platform already adds tags before adding your own, and keep recipient names and email addresses out of the campaign labels.

Send a test to yourself, follow the actual email link on a phone, and complete the intended action. Confirm the destination, confirmation message, and resulting record. Verify the tracking records the successful completion once, and exclude the test from the campaign results. If a booking tool or checkout lives on another domain, include that step in the test.

For a reply-based campaign, check that replies reach a monitored inbox and assign someone to classify them. Link analytics will not replace that review. If your team needs a broader process for qualification, the inquiry-to-revenue tracking guide covers the handoff.

Copy this campaign scorecard

Use these fields in a document or spreadsheet. Leave an unavailable result marked “not measured”; don't substitute a guess or a zero.

• Campaign and audience: name, send date, segment, and reason this audience received it.
• Primary outcome: the action, exact acceptance criteria, review owner, and review deadline.
• Email evidence: sent, platform-reported delivered, unique clickers, bot-filter status, and delivery or unsubscribe concerns.
• Website evidence: recorded campaign visits and completed actions, with the counting unit and reporting dates.
• Business evidence: verified registrations, suitable inquiries, or paid orders, with the source record and duplicate-check method.
• Connection to the campaign: tagged visit, recorded reply, campaign code, or another documented method. Keep unknown sources separate.
• Limits: missing tracking, incomplete qualification, changed filters, test records, or an unfinished buying cycle.
• Decision: repeat, revise, investigate, or wait; name one next action and its owner.

For any percentage, show the numerator and denominator. “24 unique clickers out of 800 delivered emails” is easier to interpret than an unlabeled “3% engagement.” Keep the definitions and observation window consistent when comparing sends.

A worked example: interest did not equal bookings

Hypothetical example, not a Branding Bull client result: a service business sends an invitation to 800 recipients whose messages the platform reports as delivered. After bot filtering, the report shows 24 unique clickers. The website records eight completed request forms with that campaign label. The team checks the eight records, removes two duplicates, and finds that three of the six distinct inquiries fit the service. No booking has been confirmed by the agreed review date.

The report can say: 24 filtered unique clickers, eight recorded submissions, six distinct inquiries, three qualified inquiries, and zero confirmed bookings as of that date. It cannot say the campaign produced 24 prospects or eight customers.

The next step is to review the three unsuitable inquiries and the follow-up on the three suitable ones. Were the service requirements unclear? Was the next step too demanding? Is a decision still pending? Those are questions to investigate, not conclusions proved by the numbers.

Even a purchase connected to an email is not proof that the email caused a sale that would otherwise never have happened. For an ordinary campaign report, describe the documented connection and its limits. Don't present attributed activity as an experiment proving incremental sales.

Make the next decision small enough to test

If tracking is incomplete, fix and test that path before drawing a creative conclusion. If the evidence is sound but suitable inquiries are scarce, review the audience, offer, and page together. If suitable inquiries are waiting, check follow-up before sending more traffic. Small counts call for careful investigation, not sweeping claims about what always works.

If you want help turning this scorecard into a repeatable email process, bring one recent campaign and its available reports to a conversation about Branding Bull's email marketing service. Campaign planning, content, testing, and reporting can be scoped around the platform and evidence you already have.