The Right Way To Iterate: How to Improve Without Starting Over

Read Time: 3 minutes.

Stop Agreeing to Disagree

I’m noticing a pattern with a range of leaders that is holding their teams back.

It comes in two parts:

1. There is no initial agreement

2. There is no systematic update of the agreement when things change

Instead, people operate by intuition and constantly hit reset when it fails them.

• It can happen on your team.

• It can happen between your team and an adjacent one.

• It’s why consulting firms invented the RACI matrix (even though the solution is far simpler).

The solution comes down to this:

1. Do we have an agreement of how we work together? This means we agree on the standard for passing work between us. We have a shared number or a shared understanding of how our numbers relate.

2. Do we have enough trust to know we’ll need to revise the agreement? Business is dynamic. Priorities change. Mistakes lead to learning if we let them. We’ll have to thoughtfully amend this agreement at some point.

Honestly, you can do this on a post-it note. The key is being explicit. And then being disciplined enough to compound your learnings into the agreement instead of contently starting over.

Wining iteration builds on previous agreements. It compounds.
Fake iteration are expensive do-overs disguised as learning. It confounds.

Let’s get it right.

Why Teams Start Over Instead of Iterate

The Perfectionism Trap: If it doesn't work perfectly, the whole thing must be wrong.

The Shiny Object Appeal: New approaches feel easier than fixing current ones.

The Agreement Amnesia: Teams forget what they originally agreed to try and why.

The Painful Result: You lose all accumulated knowledge and relearn basic lessons.

The Real Iteration Framework

Step 1: Make a Clear Agreement

Before starting anything, document:

• What you're doing

• Why you're doing it

• How you'll measure success

• Who is doing what

• When you'll review and decide

Example: 

• What: Weekly one-on-ones for 6 weeks.

• Why: Catch issues early.

• Success: 90% completion + team satisfaction >7/10.

• Who: Employee leads. Manager coaches.

• When: Review March 15th.

Step 2: Execute and Learn

Do the work while capturing insights:

• What's working better than expected?

• What's worse than expected?

• What assumptions were wrong?

• Who should do what differently?

Key distinction: Separate execution problems from design problems.
Are you not doing it consistently, or does the approach need adjustment?

Step 3: Review Against Original Agreement

At your review date, ask:

• Did we achieve what we agreed to measure?

• What did we learn?

• What should we keep?

• What needs adjustment?

Critical rule: Don't evaluate based on perfection.
Evaluate based on direction and learning.

Step 4: Revise, Don't Replace

Build on what's working. Fix what's not.

Refinement: One-on-ones work but 30 minutes is too long.
New agreement: 20 minutes with agenda.

Expansion: One-on-ones effective but need follow-up.
New agreement: Adding shared notes and action tracking.

Reduction: Weekly too frequent for senior staff.
New agreement: Bi-weekly for VPs, weekly for Directors.

Pivot: Individual meetings don't scale.
New agreement: Team huddles + monthly individual check-ins.

The key to leadership success?

Simple systems that help you make good on your good intentions.

We’ve taught our systems to more than 1300 leaders.

And now we’re using AI to make them even more efficient.

Grab one of the last 9 remaining seats for 2025. 

The Compound Learning Effect

Proper iteration creates compound learning:

Cycle 1: Learn what works and what doesn't
Cycle 2: Learn how to make what works better
Cycle 3: Learn how to scale what works
Cycle 4: Learn how to systematize what works

Don’t automate until it’s working efficiently.
It’s wasteful to engineer something that shouldn’t exist in the first place.

Common Iteration Killers

No Clear Agreement: "Let's try this and see how it goes"
Fix: Always document what you're testing and how you'll measure it.

Evaluating Too Early: Judging after one week
Fix: Agree on timeline upfront and stick to it.

Perfectionism: "Didn't work perfectly, need something new"
Fix: Measure progress, not perfection.

Forgetting the Goal: Getting lost in the solution
Fix: Connect every iteration back to the original problem.

Your Next Move

Pick someone you’re struggling to work with effectively. Apply the framework:

1. Write out current agreement: Who is doing what and why?

2. Agree on a review date: When will you re-evaluate?

3. Capture learning: What's working, what isn't?

The goal isn't getting it right the first time.
The goal is getting it more right each time.

Because iteration isn't about starting over.
It's about getting better at getting better.

What You Missed This Week

Our Sunday AM posts:

📌 9 Future-Proof Career Cheat Codes (Dave on LI)
📌 Why is Pays to Read Ahead (Mar on LI)
📌 8 Most Common Management Challenges (+Solutions) (Dave on X)

And here are our most popular posts last week:

🔥 How to Stay Calm in the Chaos (Dave on LI)
🔥 How to Build Enduring Trust (Dave on X)
🔥 11 Quiet Coaching Lesson to Ignite Change (Mar on LI)

Our goal is to build a community of 1 million thoughtful, curious leaders.

You can help us by reposting anything that resonates with you.  

Please Vote: Is Less Actually More?

We recently got feedback from a few readers that the MGMT Playbook was getting too long and in-depth and we should focus on solving more bite-sized problems faster.

What do you prefer?

Thank you for reading. Appreciate you!

Dave