Who This Helps
This is for growth marketers who feel stuck deciding between ten different ideas. The Channel Basics: Offers & Creative course gives you the framework to stop the guesswork and start moving key metrics.
Mini Case
Sam had three ideas: a new landing page headline, a different pricing tier, and a referral bonus. By scoring them, she found the referral bonus had the highest potential impact and lowest effort. She ran that test first and saw a 15% lift in new user acquisition within 7 days. The other ideas? They went into the backlog for later.
Do This Now (5 Steps)
- List your top 3 experiment ideas. Write them down. No more than three.
- Score each on Impact (1-10). How much could it move your key metric? Be honest.
- Score each on Confidence (1-10). How sure are you it will work? Use past data.
- Score each on Effort (1-10). How much time and resources does it need? 10 is a huge lift.
- Calculate: (Impact + Confidence) / Effort. The highest score wins. That's your next experiment. It's like a recipe for focus.
Avoid These Traps
- Chasing shiny objects. That cool new platform your competitor is on? Score it. It probably has low confidence and high effort.
- Analysis paralysis. Don't spend 3 days perfecting the scores. Use your gut, assign numbers in 10 minutes, and commit.
- Ignoring the 'Confidence' score. A high-impact, low-effort idea is useless if you have zero data to support it. That's just a hope.
- Forgetting the 'Effort' score. A massive 6-month project will kill your team's momentum. Start with quick wins.
- Working on everything at once. You only have one team. Pick one experiment to run cleanly.
- Not defining 'done'. Before you start, decide what metric and what lift means success. Is it a 5% increase in sign-ups?
- Letting HiPPOs decide. The Highest Paid Person's Opinion shouldn't override the score. Let the system work.
- Skipping the backlog. Good ideas that don't win go here. Review it monthly. Your future self will thank you.
Your Win by Friday
By this Friday, you will have one prioritized experiment ready to launch. You'll have killed the debate about what to do next. Your team will know exactly what they're building and why. Go run your winning test. The data (and your sanity) are waiting.