Crush Your Goals with Feedback Loops

By Marwin Zoepfel

The Thermostat and the Timer

In the world of engineering, stability isn't an accident—it's the result of meticulous design. Take something as mundane as a thermostat in your home. It doesn't just blast heat indiscriminately; it operates on a principle that's both elegant and essential: the feedback loop. The thermostat constantly measures the room's temperature, compares it to your set point (say, 22°C), and adjusts accordingly—turning the furnace on if it's too cold or off if it's too warm. This closed-loop system keeps things balanced, preventing wild swings.

Now, contrast that with an "open-loop" system, like a basic kitchen timer hooked up to your heater. You set it for an hour, and it runs full blast, no questions asked. The room could turn into a sweltering sauna or remain a frigid icebox, but the timer doesn't care—it has no way to check the outcome. It's blind to reality, plowing ahead regardless of results.

This isn't just a lesson in household appliances; it's a mirror to how most of us navigate life. We set ambitious goals—"I want to be more productive," "I need to get fit," or "I'll build better relationships"—and then dive in with effort and enthusiasm. We hustle, grind, and push forward, but here's the catch: we rarely pause to measure if our actions are actually moving the needle. We're like that timer, operating in an open loop, assuming hard work alone will get us there. But without feedback, we're guessing, not guiding.

The good news? You can engineer your life like a well-designed system by installing your own feedback loops. This isn't about drowning in data or downloading a dozen apps—it's about stripping things down to the "bare metal," the raw essentials that make real change possible.


The Bare Metal Control Loop: A 5-Step Protocol for Real Change

Forget vague goals and relying on willpower. We're going to stop guessing and start engineering the results we want. This protocol is a simple control system for your life. It's about getting clear signals and making decisions based on data, not feelings. Here’s how it works.

Step 1: Isolate One Quantifiable Metric (Your Output Signal)

You cannot control what you do not measure. The first step is to translate your fuzzy goal into a single, undeniable number. This number is your metric. It must be something you can count, not just a feeling.

Goal: "Be less distracted by my phone."

  • Bad Metric: "Feeling less distracted." (This is subjective and unreliable.)
  • Good Metric: "Number of times I unlock my phone per day." (This is a hard number. You can find it in your phone’s "Screen Time" or "Digital Wellbeing" settings.)

Goal: "I want to get fit."

  • Bad Metric: "Working out more." (How much is "more"?)
  • Good Metric: "Number of workouts completed per week" or "Total minutes spent walking each day."

Goal: "I need to stop procrastinating on my project."

  • Bad Metric: "Making more progress." (What does that even mean?)
  • Good Metric: "Number of 25-minute 'Pomodoro' work sessions completed per day."

Action: Pick one goal and define your metric. This is the only number you will track.

Step 2: Establish a Baseline (Calibrate Your System) 📊

Before you try to fix anything, you need a map of where you are right now. You must measure the system in its current, unchanged state.

For three days, live exactly as you normally would, but at the end of each day, write down your metric in a simple notebook.

Example (Goal: Reduce phone distraction):

  • Day 1: 112 phone unlocks
  • Day 2: 121 phone unlocks
  • Day 3: 108 phone unlocks

Now, do the simple math: (112 + 121 + 108) / 3 = 113.6. Your baseline average is 114 unlocks per day. This number isn't good or bad; it's just the truth. This is the problem, quantified.

Step 3: Implement One Change (Your Control Input) ⚙️

Now we introduce a single, specific change to the system. The key is to change only one thing. If you try to change five things at once, you’ll have no idea which one was actually responsible for the result. We need a clean signal.

Example: The change we will test is a simple rule: "When I'm home, my phone lives in a drawer in the closet. I can use it, but it must go back in the drawer immediately after."

For the next seven days, you will live with this one new rule.

Step 4: Analyze the New Signal (Compare to Baseline) 📉

As you live with the new rule, continue to track your single metric every single day.

Example: After a week with the "phone in the drawer" rule, your log looks like this:

  • Day 4: 58 unlocks
  • Day 5: 65 unlocks
  • Day 6: 55 unlocks
  • Day 7: 61 unlocks
  • Day 8: 54 unlocks
  • Day 9: 62 unlocks
  • Day 10: 59 unlocks

Your new average is 59 unlocks per day. Now you can compare the two numbers directly. Your baseline was 114, and your new average is 59. The data, not your feelings, proves that this new system caused a 48% reduction in phone unlocks. This is an undeniable signal that the change worked.

Step 5: Decide: Persist or Iterate (Tune the Loop) 🔁

This final step is simple. You have clear data, so you can make a clear decision.

  • If your metric improved significantly: The change was effective. The decision is to persist. The "phone in the drawer" rule is no longer an experiment; it's just the way your system operates now. It's your new, optimized baseline.
  • If your metric didn't change (or got worse): This is not a failure—it is valuable data! The data proves that your hypothesis was wrong. The rule you implemented was not the right solution for the problem. Now you can iterate. Form a new hypothesis (e.g., "I will delete social media apps from my phone for one week") and run the 5-step experiment again.

This is debugging. You test, you analyze the data, and you iterate until you find the solution that works. This is how you build a life that is not accidental, but engineered.