The Target Expression builder provides an estimate of the number of subscribers your selection will target. This feature is available across all sections within the platform that use target expressions, including activities, integrations, audiences, segments, extracts, and assets.

Real-time estimations

As you begin to build your expression, the system will start estimating the number of subscribers in real time. The calculation duration depends on the complexity of the expression—it can take anywhere from a few seconds to a couple of minutes (e.g., when querying more than six years of transaction history).

Once calculated, an approximate count is displayed. An error range is shown in brackets beside this number, indicating that the final targeted number will fall within this range. This estimate should not be considered as the final targeted number.

Channel-specific estimates

In email and SMS activity targeting, estimates will account for the contactable portion of the database relevant to the selected channel. As a result, the estimate may differ from the same target expression when viewed in other contexts, such as the extract dialog.

Estimated vs Targeted counts

After saving the target expression, the estimated count will be shown in list views on the dashboard.

Once an Activity Check has been performed, the displayed count will update from Estimated to Targeted. This reflects a more accurate number of recipients that will actually be targeted.

This serves as a helpful visual indicator for users, showing which activities have had an Activity Check completed.

Why estimates may differ from Activity Check numbers

When confirming deployment, you may notice two different recipient counts displayed at the same time—an estimate shown in the confirmation dialog (e.g. "Approx 213 (153 – 273) recipients") and an Activity Check count shown in the pre-deployment checklist (e.g. "Number of recipients: 234"). It is normal for these numbers to differ, and this section explains why.

How each number is calculated

Estimates are calculated using statistical sampling. Rather than querying every subscriber record in full, the system analyses a representative sample of the database and extrapolates a result. This makes estimates fast, but introduces a margin of error—shown as the range in brackets (e.g. 153 – 273). The true count is expected to fall somewhere within this range.

Activity Check counts are calculated by querying the full subscriber database at the moment the check is run. This produces a more precise count, but because it is a point-in-time result, it may still differ slightly from the final deployment count if the database changes between the check and the actual send.

Common reasons for the difference

• Sampling vs full query — The estimate is based on a statistical sample, so it will rarely match an exact count. A result that falls within the estimate's error range (e.g. 234 falling between 153 and 273) is expected and correct.
• Database changes over time — Subscribers join, unsubscribe, or become uncontactable between when the estimate was calculated and when the Activity Check is run. Each query reflects the state of the database at that particular moment.
• Estimate caching — Target expression estimates may be cached for a period to improve performance. The Activity Check always reflects a fresh query at the time it is performed.
• Channel contactability filtering — Activity Check counts apply channel-specific filtering (e.g. only subscribers with a valid email address for an email activity), which may also be applied to estimates but can produce slightly different results depending on timing.

What to expect

If the Activity Check count falls within the estimate's error range, no action is needed—the numbers are consistent. If the Activity Check count falls significantly outside the error range, consider re-running the estimate or reviewing the target expression to ensure it is correct before deploying.

---