Avoid mixing unique audiences, repeated exposure and changing interface labels in your content reports.
Read the metric description available in the current interface or official help before using it in a report. Reach generally concerns distinct accounts, while view or impression measures can count exposure differently depending on the format and reporting surface. Do not assume that a familiar name keeps the same definition after an interface update. Record the source screen, content format and period so that another person can reproduce the number you are discussing.
A higher exposure count does not automatically mean more distinct people encountered the content. The same person may contribute to repeated viewing where the metric permits it. Conversely, a wider reach does not establish that people watched the whole item. If you calculate exposure per reached account, treat it as a ratio under the source’s definitions, not a direct count of complete watches. Keep any missing or unavailable metric clearly marked instead of substituting another.
A Story, Reel and profile-level report can summarize different activity. Keep those views separate until you have checked their scope and time range. Also distinguish paid promotion, normal account activity and externally ordered service delivery in your own records. An order’s delivered quantity and a platform analytics total need not be identical measurements. When they differ, inspect the service conditions and reporting windows rather than declaring one number false without checking what it counts.
For each result, record the content ID, reporting period, metric name, raw value and a brief observation. Add a business action such as relevant enquiries or destination clicks where available. This lets you distinguish “more people were reached” from “more customers contacted us”. Preserve the previous definition when changing your report so that trends do not silently combine incompatible data. A useful report makes uncertainty visible instead of hiding it behind an impressive total.