Compare dashboards and exports

Two reports can show different numbers because they count different activity, populations, or periods. Start by matching the scope before requesting a correction.

Compare the same question

1. Record both page or report names, the campaign or employee involved, and the exact metric labels. Keep the export generation time as well.
2. Match the audience: one campaign, a recurring campaign’s runs, selected groups, an individual employee, or the whole organization.
3. Match the date coverage. The formatted campaign report overview can cover the campaign’s history while its chart covers a shorter period. Employee performance also restricts simulations by delivery date. Click Timeline uses its own six-month window.
4. Match the event definition. Clicked, Entered Info, and Failed are not interchangeable. See campaign results.
5. Match the counting unit: events, simulation recipients, employees, or campaigns. Repeated events, multiple campaigns per employee, and optional summary rows can make CSV row counts differ from dashboard totals.
6. Match the percentage denominator. A percentage based on delivered simulations is not comparable to one based on scheduled recipients, failures, or training assignments.
7. Let pending report processing finish, then refresh or generate a new export for the same scope. An older file will not update when new activity arrives.

Common explanations

Difference	What to check
Clicked is higher than Entered Info	Some recipients opened the page without submitting information. An information-entry campaign does not treat every click as a failure.
Raw report events exceed Top Reporters credit	Successful-report rankings and the campaign overview’s Reported count exclude reports with a preceding counted failure. Raw reports can still appear in recipient details and chart data. If the overview count itself differs from the leaderboard, compare the audience, period, and counting unit.
Employee totals differ from campaign totals	Compare delivery dates, event dates, test-run inclusion, and the selected population. Removed records and historical reporting scope can also affect which recipients appear.
CSV has more rows than a count tile	Determine what each row represents, and exclude headers or summary rows from manual row counts. Do not equate event rows with unique employees.
A click spike looks implausible	Security scanners or proxies can interact with simulation content. Filtering reduces noise but does not guarantee every recorded interaction was human. Review your allowlisting setup and ask Support to investigate.
Training completion differs	Distinguish campaign follow-up training from standalone assignments, and check the selected period and completion-processing state.

Contact Support

If the difference persists, email support@jerichosecurity.com with:

• The campaign/report names and exact metric labels being compared.
• Selected dates, audience filters, and the time you generated the export.
• The two counts you expected to match and why.
• Redacted screenshots or a minimal relevant export sample, using your organization’s approved sharing process.

Do not send passwords, integration tokens, or submitted simulation form contents. Ask Support to confirm the interpretation before using disputed figures in an executive report.