Compare outreach audience segments with consistent denominators

Define the decision

A segment can represent role, company size, industry or a verified business condition. Those dimensions overlap, and list changes can move people between them. Preserve the segment definition used at enrollment so a later enrichment update does not silently rewrite the historical experiment.

Work through the procedure

1. Save the segment rules and membership snapshot for the cohort being evaluated.
2. Use either mutually exclusive reporting groups or clearly label overlapping views.
3. Compare the same message treatment and reply observation duration where possible.
4. Inspect reply reasons and qualification quality alongside the conversion rate.

Worked example

The following is a synthetic example for this procedure, not a customer result or performance benchmark.

Segment A: 6 positive respondents / 150 contacted = 4%
Segment B: 8 positive respondents / 400 contacted = 2%
Thirty people appear in both descriptive segments.
Do not report 550 unique people by adding the denominators.
For a controlled test, assign each eligible person to one analysis group.

Read the result

A higher percentage from a smaller segment may be commercially useful, but it can also be unstable or reflect a different offer. The overlap note prevents an inflated reach claim. Read the actual conversations to distinguish useful interest from requests that the business cannot serve.

Check before moving on

1. Keep unknown values separate from confirmed negative qualifications.
2. Document membership changes after the reporting cutoff.
3. Avoid selecting only the best-looking segment after seeing every result.

Limits and next action

Segmentation does not by itself make a comparison causal. These examples are analytical workflows, not a claim that Zintara automatically stores every historical membership snapshot or performs statistical segment testing.

Source: Zintara product context; analytical methods and examples are defined in the guide