The third article in the systems series. Read this when a system is clearly not working — or when it's technically running but producing weaker results than it should.
The Problem with Jumping to Solutions
The instinct when a system underperforms is to fix it. Immediately. Change the schedule, add accountability, try a new tool, recommit with more intensity. This almost always fails — not because the fix is wrong, but because it's applied before the diagnosis is done.
As Harvard Business School professor Michael Tushman puts it: "Instinctually, most leaders go straight to solutions. You must do a thorough diagnosis first, determine the root causes of your gap, and only then can you move to an integrated intervention."
That's true for organizations and it's equally true for personal systems. Applying the right fix to the wrong problem produces temporary improvement followed by the same failure, usually faster. The audit comes first. The fix comes after.
Step One: Distinguish Underperforming from Failing
Before you start the audit, get clear on what you're actually dealing with. These are different problems.
A failing system has stopped running. You're skipping it entirely, the habit has collapsed, the routine is gone. The previous article ("When Your System Breaks Down") addresses this. The immediate move there is to restart quickly at a reduced intensity, then audit.
An underperforming system is still running but producing worse results than it should. You're showing up, putting in the time, going through the motions — but something isn't clicking. Progress is slower than expected, results are flat, or the system feels increasingly like a grind with no payoff.
The audit process below is primarily for underperforming systems — ones where the execution is happening but the outcomes aren't matching the effort. That's where the real diagnostic work lives, because the problem is hidden inside something that appears to be working.
Step Two: Measure the Gap Before You Explain It
The first job in any audit is to quantify the underperformance as specifically as possible. Vague dissatisfaction ("this isn't working") can't be diagnosed. A specific gap can.
Ask yourself:
- What result was I expecting from this system by now?
- What result am I actually getting?
- How long has this gap existed?
- Is the gap growing, stable, or slowly closing?
Write this down. The act of putting numbers or specific observations on paper forces precision and often reveals that the gap is either smaller than you felt (which is a different problem — expectation calibration) or larger and longer-running than you'd admitted to yourself (which tells you the system needed attention sooner).
This step also tells you whether you have a performance problem (the system is running but producing insufficient results) or an expectation problem (the system is running and producing reasonable results, but you expected more, faster). These require completely different responses. Many "failing systems" are actually reasonable systems with unrealistic timelines attached to them.
Step Three: Apply the 5 Whys to the Gap
Once you have a specific gap defined, the most effective tool for finding its root cause is a simple method developed by Toyota called the 5 Whys. The principle: when a problem occurs, ask "Why?" and then ask "Why?" again to the answer you just got — repeat this five times (or however many it takes) until you reach a cause you can actually address.
The method works because most people stop at the first plausible explanation. That explanation is almost always a symptom, not a root cause. Fixing a symptom produces temporary improvement. Fixing a root cause fixes the system.
Here's how to apply it to a personal system:
Example: Exercise system is underperforming
- The problem: I'm working out three times a week as planned but haven't gained any strength in eight weeks.
- Why? My workouts feel unfocused — I'm not pushing hard enough.
- Why? I'm always rushed when I get to the gym.
- Why? I go after work and I'm already depleted by then.
- Why? My afternoons are consistently high-stress and cognitively draining.
- Why? I've taken on too many commitments and haven't protected any recovery time in my schedule.
The root cause here isn't the exercise system. It's the schedule and energy management. Changing the workout program won't fix it. Shifting workouts to morning and protecting recovery time will.
The 5 Whys works best when you're brutally honest with each answer and resist the urge to stop at the first layer. The first answer is almost always just another symptom. Push past it.
Step Four: Run the Four-Domain Audit
The 5 Whys tells you where to look. This audit tells you what to look at. Every personal system can fail in four distinct domains, and each requires a different fix.
Domain 1: Design
Is the system built correctly in the first place? A poorly designed system will underperform regardless of effort or commitment.
Audit questions:
- Is the goal this system is working toward clear and specific enough to be measurable?
- Does the system have the right inputs? (Enough time, the right time of day, the right environment?)
- Is the system sequenced correctly? Are you doing the right things in the right order?
- Is there too much in the system? Overcomplicated systems fail because of their own weight.
Design problems are the most important to fix first because everything built on a bad design will inherit the flaw.
Domain 2: Friction
This is about how hard or easy the system is to execute. Justin Sung's principle from the first article applies directly: if executing the system requires willpower, it's too high-friction to be sustainable. A well-designed system should be the path of least resistance, not a daily act of discipline.
Audit questions:
- What are the three most common reasons I skip or shortcut this system?
- At what point in the process do I most often stop or disengage?
- What would make starting the system easier? (Environment design, preparation, cues?)
- Are there unnecessary steps that add friction without adding value?
Friction problems are often invisible because you adapt to them. You stop noticing that setting up for the system takes 20 minutes, or that the cue is unreliable, or that the location is inconvenient. You just feel a vague resistance to starting.
Domain 3: Feedback
You can't improve what you don't measure. A system with no feedback loop is running blind — you won't know it's drifting until the drift becomes a collapse.
Audit questions:
- How do I know if this system is working?
- Am I tracking any metrics, even rough ones?
- How quickly do I find out when something goes wrong?
- When did I last deliberately review whether this system is producing results?
The fix for a feedback problem is almost always simpler than people think. You don't need a spreadsheet or an app. You need one measurable indicator reviewed once a week. Pick the simplest metric that reflects the system's health and look at it regularly.
Domain 4: Alignment
This is the deepest level. Is this system actually serving a goal you still have? Goals change. Values evolve. Life circumstances shift. A system that was perfectly aligned six months ago may be quietly misaligned now — and the underperformance is the signal.
Audit questions:
- Does executing this system still feel connected to something that matters to me?
- Has anything changed about my priorities, situation, or values that would change what I'm trying to accomplish?
- Am I doing this because I genuinely want the outcome, or because I built the habit and stopping feels like failure?
- If I were designing this system fresh today, would I build something similar?
Alignment problems are the most uncomfortable to acknowledge because they raise the possibility that you need to stop, not optimize. Sometimes the right outcome of an audit is deciding that this particular system is no longer worth running, and redirecting that energy somewhere else. That's not failure — it's recalibration.
Step Five: Prioritize One Fix
The audit will likely surface multiple problems across multiple domains. This is normal — systems that underperform rarely have exactly one thing wrong. The temptation is to fix everything at once.
Don't.
Making multiple changes simultaneously is both bad engineering and bad psychology. Bad engineering because if the system improves (or doesn't), you won't know which change caused it. Bad psychology because the cognitive load of implementing multiple changes at once increases friction and reduces follow-through.
Prioritize one fix using this logic: fix design problems first, then friction, then feedback, then alignment. Design is upstream of everything else — a friction reduction built on a bad design is pointless. Feedback is useless if the system design is wrong. And alignment can't be resolved until you know whether the system is actually capable of working.
Run the single fix for two to four weeks before evaluating. This is shorter than it feels. Most system improvements need at least two weeks to produce visible signal. Evaluating after five days is premature and will cause you to abandon improvements that were actually working.
Step Six: Confirm the Fix Worked — Then Look for the Next Constraint
After two to four weeks, measure the gap again using the same metrics you defined in Step Two. Three things can happen:
The gap closed. The fix worked. Document what you changed and why it helped — this becomes part of your systems knowledge. Then look for the next constraint in the system and repeat the process. This is continuous improvement: small, sequential fixes that compound over time.
The gap partially closed. You found a real problem but not the only one. The fix was correct but incomplete. Return to the audit, identify the next bottleneck, apply the next fix.
The gap didn't change. One of two things: either the fix was wrong (you treated a symptom, not a root cause) or the fix was right but two weeks isn't enough time to see results (some system improvements take longer — fitness is the obvious example). Give it another two weeks before concluding the fix failed.
The System Audit on One Page
When something's underperforming, work through this sequence:
- Classify it. Is it failing (stopped running) or underperforming (running but weak results)?
- Quantify the gap. What result did you expect? What are you actually getting? How long has the gap existed?
- Check expectations first. Is this a system problem or a timeline problem?
- Apply the 5 Whys. Ask why the gap exists, then why that answer is true, until you reach something actionable.
- Run the four-domain audit. Design, friction, feedback, alignment — which domain is the primary failure point?
- Fix one thing. The highest-priority fix in the highest-priority domain. Nothing else.
- Run it for two to four weeks. Then measure the gap again.
- Repeat. Continuous improvement, one constraint at a time.
A Note on Honesty
The audit process only works if you answer the questions honestly. The most common failure mode in system audits isn't using the wrong framework — it's letting yourself off the hook with comfortable explanations.
"I've been unusually busy" is almost always true. It's also almost always a reason not to redesign the system rather than a root cause. Push past the comfortable answer to the actual one.
The goal of the audit isn't to feel better about the underperformance. It's to find what's actually broken and fix it. Those are different objectives, and conflating them is what produces people who "audit their systems" regularly without ever improving them.
No comments:
Post a Comment