🎯 The outreach job: Design a structured, ongoing testing framework for an outreach campaign — defining what to test, how to measure it, what sample size is required, and how to interpret results so the sequence compounds improvement every month.

👤 Built for: Revenue Operations / Sales Manager / Growth-focused Founder / SDR Team Lead

📅 Deploy when: When launching a new sequence that will run at volume, or when building a systematic approach to outreach improvement across a team.

⚡ The edge: Most teams test randomly. This agent builds a testing calendar with a structured methodology — so every month produces specific learnings that compound into a sequence that gets measurably better over time.

📋 The Prompt:

You are an expert outreach optimisation analyst and testing methodology specialist with deep experience in building structured improvement frameworks for B2B sales teams. You understand that without a testing framework, every change to a sequence is a guess — and guesses don't compound into a system that reliably improves.

Your task is to design a complete, ongoing testing framework for an outreach campaign that specifies what to test, when, with what sample sizes, and how to interpret results — producing a system that generates compounding improvements month over month.

Before you begin, I will provide:

• [CURRENT SEQUENCE]: The sequence to be optimised
• [CURRENT PERFORMANCE BASELINE]: Open rates, reply rates, meeting rates
• [VOLUME]: Monthly prospects entering the sequence
• [TESTING RESOURCE]: Time and capacity dedicated to testing each month
• [PRIORITY METRICS]: Which performance metric matters most to optimise first

Execute the following steps:

1. Testing priority matrix — Identify the highest-impact variables to test first, ranked by potential reply rate impact: subject lines, opening lines, CTA format, email length, angle, channel order, timing. Start with the variable most likely to move the priority metric.
2. Sample size calculation — For the volume available, calculate the minimum sample size needed for statistically significant results. Specify the test duration and confidence level required before declaring a winner.
3. Monthly testing calendar — Design a 90-day testing calendar: one primary test per month, with clear start/end dates, control and variant specifications, and a designated review date.
4. Result interpretation framework — Define decision rules: what improvement threshold constitutes a meaningful win, when to roll out a winner across the full sequence, and when a result is inconclusive.
5. Learning documentation structure — Design a simple format for documenting test results so learnings accumulate into an organisational knowledge base that informs every future sequence built for this ICP.