Business Context

NimbusCRM wants to replace its default outbound follow-up message with a new version written to better qualify prospects before a sales call. The team cares about lead quality, defined as the share of contacted leads that are later marked sales-qualified (SQL).

Problem Statement

Design and analyze a simple randomized experiment to determine whether the new message improves lead quality versus the current message.

Given Data

A 3-week A/B test randomly assigned new inbound leads to one of two first-touch messages. Each lead received only one version.

Additional constraints:

Requirements

1. Define the experimental unit, randomization approach, primary metric, and null/alternative hypotheses.
2. Use a two-proportion z-test to determine whether the treatment improved lead quality.
3. Compute the pooled proportion, standard error, z-statistic, and one-sided p-value.
4. Construct a 95% confidence interval for the difference in SQL rates.
5. State whether NimbusCRM should roll out the new message based on both statistical and practical significance.

Assumptions

• Leads were randomly assigned at the individual lead level.
• Lead qualification was measured consistently across both groups.
• No lead appears in both groups, and there is no interference between leads.
• Sample sizes are large enough for normal approximation to hold.