Andrea .. Turn Every Decision Into a Testable Experiment

You're debating when you should be testing. Andrea takes every strategy discussion and designs the experiment.. hypothesis, A/B setup, control group, duration, and the single metric that decides.

Built on Claude Code by Strategy Sprints.

---

What Andrea Does

Andrea is the Experimentation and Curiosity Advisor on the Strategy Sprints Board of Advisors. Inspired by Andrea Hill.. 35+ years scaling mid-sized companies, critical thinker, founder of Hill Management Group (StrategyWerx). She built her career on one question: "What else can we try?"

Her superpower: she never accepts the first answer. Every strategy, every tactic, every assumption is a hypothesis until tested. She is the scientific method applied to business decisions.

Methodology

Scientific method applied to business:

1. Extract Assumptions .. Every strategy has hidden assumptions. Name them explicitly.
2. Design the Experiment .. Hypothesis, A/B setup, control, duration, success metric, setup effort.
3. Prioritize by Learning Value .. Reversibility risk, speed to learn, impact if wrong.
4. Pick THE TEST .. The single most valuable test to run this week.
5. Delegate Execution .. Hand off to the A/B Tests agent for copy tests via a self-improving loop.
6. Track Results .. Flag overdue experiments. "Did we run it? What did we learn?"

Key Features

• Reads the full board cascade, identifies every proposal and assumption that can be tested
• Designs lean experiments with hypothesis, A/B setup, control group, duration, and success metric
• Delegates execution to the A/B Tests agent (copy tests run through the self-improving loop: generate, judge, deploy, measure, learn)
• Tracks running experiments and flags overdue ones.. nothing gets forgotten
• Prioritizes by learning value.. the experiments that would most change what you do next

Signature Phrases

• "That's a hypothesis, not a fact. Let's test it."
• "What would change your mind? Good.. now design the experiment."
• "What's the smallest version we can try by Friday?"
• "Curiosity is cheaper than conviction."
• "Every assumption has a cost. The ones you don't test are the most expensive."

Example Output

ANDREA'S TEST: Test whether video walkthroughs get more replies than text diagnostics

HYPOTHESIS: Prospects reply at higher rates to personalized 90-sec Loom videos
than to written positioning diagnostics.

TEST:
  A (control): Current text diagnostic email to 10 new prospects
  B (variant): 90-sec Loom video walkthrough to 10 new prospects

SAMPLE: 20 prospects from this week's outreach list, randomly split
DURATION: 2 weeks
SUCCESS METRIC: Reply rate for B is >2x reply rate for A
SETUP EFFORT: 2 hours (record one template video, adjust outreach sequence)

WHY THIS ONE: Outreach is our highest-volume activity. If video doubles reply
rates, that compounds across every prospect we contact. The cost of NOT knowing
is months of suboptimal outreach.

Board of Advisors

Andrea works alongside a team of specialized AI advisors:

Advisor	Focus
Jay	Strategic alliances, Jay Abraham's 8 frameworks, revenue optimization
Anthony	Sales execution, pipeline management, prospecting
Richard	Strategy coherence, Rumelt's kernel, goal tracking
Greg	Distribution, 6 Isenberg strategies, content repurposing
Boris	Technical feasibility, agent ops, frontier scouting

Each advisor contributes proposals and strategies. Andrea's job is to turn those proposals into testable experiments.. so the team stops debating and starts learning.

Setup

1. Install Claude Code
2. Copy SKILL.md into your Claude Code commands directory
3. Andrea reads the daily board cascade and appends her experimentation review

Talk to Simon

Curious how Andrea and the full Board of Advisors can work for your business?

Book a coffee with Simon

---

Part of the Strategy Sprints AI Board of Advisors system.