Reporting and experiment questions

These answers help you choose the right measurement and make sense of the numbers you see. Start with the action you want visitors to take, then check the report's dates, audience, and counting rules.

How do I measure the conversion I care about?

Use form or offer reporting for a RightMessage submission or CTA click. For a page view, click, or form submission on your website, use a conversion goal. Attach that goal to a website campaign when you want to measure the campaign against it. Campaign results require Scale or Pro.

A page view, a recorded flow ending, and an email-platform update measure different actions. None automatically confirms a purchase or booked appointment in another system. See how your flow is doing before adding more experiments.

Is showing different content to each audience an A/B test?

Not on its own. An audience's personalized content is chosen from a visitor's conditions or answers. Those audiences can have different conversion rates before you change anything.

An A/B test randomly assigns visitors to versions for comparison. A campaign's holdback compares personalized content with your page as it is, and an A/B test on an audience or the Original page compares Control with up to three other versions. See resource split-test reporting and campaign experiment decisions for their different measurement rules.

How do I compare personalization with my page as it is?

Use the campaign's Testing tab to set its holdback ("How many visitors should be held back?") and attach the goals you want to measure. Held-back visitors see your page as it is, and Campaign vs holdback compares the two groups. Read the campaign's evidence status rather than choosing whichever version shows the largest lift. To compare versions of a page or an audience's changes, run an A/B test.

A baseline of similar visitors isn't the same as a randomized holdback. If a report uses a prior unchanged experiment or a descriptive reference, read the explanation beside the comparison.

Can confidence change after reaching 95%?

Yes. A running resource split test's confidence can change as new observations arrive. Reaching 95% once doesn't guarantee how the test will end or how that version will perform in the future. Question, form, offer, and flow split tests are available on every plan.

Campaign A/B tests and the holdback use a separate decision system. Results are checked every Friday, the earliest verdict comes at the week 2 check, and each version shows a likely range against Control. Checking every week doesn't raise the chance of a false winner. Don't judge them by looking for the old 95% confidence panel. Read A/B test results covers campaign tests, and the split-test reference covers resource reports.

Where can I find completed tests?

Open A/B Testing and set Status to Completed. The page lists page and audience tests alongside question, form, offer, and flow tests, with Page tests and Audience tests filters. Completed tests keep a reporting snapshot. If you manually kept a version, that records your choice rather than declaring a statistical winner.

For campaign tests, open the campaign's Testing tab. Each test card shows its status, such as Winner found, Can't reach a verdict, or Finished. Campaign verdicts use different rules from question-test confidence. See the Testing tab reference.

Why do two dashboard sections show different numbers?

They may use different dates, include different activity, or count different things. The dashboard picker changes Activity, Results, and Working, while saved summaries and recommendations keep their own evidence dates. All-time Contacts with answers differs from selected-period Contacts synced.

Daily visitor counts, channel visitors, and page visitors can overlap. Adding them won't necessarily give you the period's unique visitor total. See the dashboard reporting windows and website stats counting rules.

Why doesn't revenue by segment add up to my Stripe total?

Because RightMessage can only credit a buyer it recognized by email, or, for a question, a buyer whose email platform contact holds a synced answer. Buyers it can't place in a segment group are listed under Not placed, and buyers in none of the group's segments appear under Not in this group (or No answer for a question). People can also be in more than one segment, so segment shares can add up to more than 100%.

A segment shows no revenue share until it has at least 5 buyers. How revenue by segment is counted explains how buyers are matched and which segments they're credited to.

Why can't I see visitor data or an AI summary?

Your access, the activity RightMessage captured, and the report's loading state all affect what you can see. A visitor can remain anonymous while their answers are recorded. A summary can still be generating or unavailable while individual measurements are ready.

An unavailable metric isn't zero. Older snippets can lack events a newer report needs. Follow the report's coverage notice and check tracking after you publish. A missing summary alone doesn't tell you why data is missing.

Does website attribution update my email platform too?

Contact attribution depends on visitor identification and the supported attribution fields or actions you've configured for your connected platform. Website reports describe the visits and conversions RightMessage captured.

To confirm a field, tag, or list update, check the contact in your email platform. Lead attribution explains the setup, and checking an integration change walks through that test.

Is an answer the same as a segment or synced field?

No. A response records what someone submitted. A segment describes membership based on an answer or rule. A synced field is data sent to an identified contact in your email platform.

Typed answers don't automatically create a segment for each value. A visible response also doesn't confirm that RightMessage identified the visitor or that your platform accepted an update. See visitors, responses, segments, and contact fields.

Can I export responses, and how long are they kept?

You can use response or flow-table exports when your project has export access. Filtering, exporting, and deleting responses explains the options.

Reports don't necessarily keep unlimited history. Use their covered dates and your account's data controls. If you need a specific retention guarantee, confirm it with support before relying on RightMessage as your only archive.