Think of the person on your team you rate lowest. What if the problem is partly the system around them?
The 30-second version
Problem: When someone keeps falling short, it is tempting to decide the person is the problem.
Shift: people perform the system they’re in, so the same person can look weak or strong.
Move: before you judge the person, check their conditions, fix one thing you control, and keep the standard.
The Spark
The same person can look weak in one setup and strong in another. This is a four-step check to run before you write a rating or start a replacement search. It keeps the standard high, and it makes sure you have looked at the part you control.
Sponsored resource 90% of AI Runs Through This CompanyThe biggest AI wins often come from companies you don’t hear about every day.
|
Why this matters
When results fall short, most of us look at the person first. Were they capable? Did they try hard enough? Those questions are fair, but they leave out the place where the person works.
The best-known story is the car plant in Fremont, California. GM closed it in 1982 after years of poor quality and absenteeism. It reopened a couple of years later as NUMMI, a joint venture with Toyota, and much of the same workforce came back. Under a different way of working, quality and attendance improved a lot. The people had not become new people. The system had changed what their effort produced.
A few more signals point the same way:
We blame traits and overlook the situation. Psychologists call this the fundamental attribution error. "Unreliable" can really mean unclear priorities and three people giving conflicting orders.
Ratings say something about the rater. A well-known 2000 study of multi-rater feedback (Scullen, Mount and Goff) found that the rater explained more of the score than the person being rated. It looked at developmental feedback, so do not stretch it to pay decisions. Still, who rates you matters.
Systems beat reminders. In Michigan ICUs, Peter Pronovost's team cut central-line infections with a short checklist, a supply cart, and permission for nurses to stop a procedure. They did not ask people to try harder.
Stars do not always travel. Boris Groysberg's research on star Wall Street analysts found that performance often dropped after a move to a new firm.
If you are a new engineering manager, this keeps you from blaming the wrong thing in your first year. If you are growing toward director, it is how you stop repeating the same hire-and-replace cycle. And it does not remove accountability. People own their behavior. You own the conditions.
Step 1 · Describe the conditions
30 minutes
Goal: write down what it is like to do this person's job, before you judge how they do it.
Priorities: could they name their top three this week, and would you agree?
Context: do they know why the work matters and who uses it?
Decisions: which ones can they make alone, and which wait for approval?
Support: what did their onboarding look like, and who do they ask when stuck?
Feedback: how quickly do they learn they are on or off track?
ARTIFACT · Conditions note
Role and team:
Top 3 priorities (as they would say them):
Decisions they own / decisions that need approval:
Onboarding and support they got:
How fast they get feedback:
What changed just before the problem showed up:Done when: the note could be read by someone who has never met the person.
Watch-out: writing it from memory. Ask the person to fill in their half too.
Step 2 · Split what is theirs from what is yours
15 minutes
Goal: sort the problem into what the person controls and what you control.
List the gaps you have seen, in plain facts ("three missed handoffs"), not labels ("not a team player").
Mark each one as mostly the person's, mostly the system's, or shared.
Be honest about shared. If you are unsure, treat it as shared until you have asked.
ARTIFACT · Two-column sort
What I observed (fact):
Mostly theirs (effort, behavior, judgment):
Mostly the system's (unclear goals, approvals, tools, docs):
Shared:
What I own fixing:Done when: your "what I own fixing" line has at least one real item.
Watch-out: using "the system" as an excuse to skip a hard conversation. If someone harms colleagues or ignores clear expectations, say so directly.
Step 3 · Look for patterns
One hour
Goal: find out if this is one person's problem or the system's.
Check for repeats. Do other capable people struggle with the same process, approval or manager?
Check your last hires. If every strong hire needs far longer than expected to become effective, look at onboarding and documentation.
Find your "load-bearing" people. Who is the only one who knows a system or absorbs every emergency? That is a risk to fix, and a burnout risk for them.
Review what the rating led to. Did a low label mean fewer chances and less context, which then "proved" the label?
ARTIFACT · Pattern check
[ ] Others struggle at the same point
[ ] New hires take longer than they should
[ ] One person is the only one who knows a critical system
[ ] The same person gets every hard problem
[ ] A past rating changed the work this person is givenDone when: you can say whether this looks like one person, a repeating pattern, or both.
Watch-out: treating a resignation as only a personal event. When a load-bearing person leaves, ask what knowledge and decisions left with them.
Step 4 · Fix one condition, keep the standard
Next two weeks
Goal: change one thing you control, and set a clear bar for what good looks like.
Pick one condition to change: a clear priority, a decision they now own, a documented process, a mentor.
Say the standard out loud. Do not lower the bar. Explain what you are changing to help them meet it.
Set a review date in two to four weeks and decide beforehand what you will look at.
Reward the multipliers. Praise people who make others more capable through docs, mentoring and sharing knowledge, not only those who ship the most alone.
ARTIFACT · Expectation and support note
The standard (what good looks like):
What I am changing on my side:
What I am asking of you:
What we will look at on [date]:Done when: the person has read it and could repeat it back to you in their own words.
Watch-out: changing five things at once. You will not know what helped.
Questions to ask your team
What slows you down the most in a normal week?
Which decisions do you wait on that you could make yourself?
What do you wish you had known in your first month?
When something goes wrong here, what usually happens next?
What is one thing I could stop, start or change that would help you most?
Mistakes to avoid
Replacing the person and keeping the same conditions.
Treating a rating as a permanent identity instead of evidence about one period.
Lowering the bar because the system is imperfect.
Praising the hero who rescues every incident instead of fixing why incidents need a hero.
Assuming a star hire will perform the same in your setup without checking what they relied on before.
Your 5-minute action
☐ Open a new doc called Conditions note v0. Pick one person you are worried about and fill in just two lines: their top three priorities as they would say them, and one decision they cannot make without approval.
If you cannot fill in those two lines, you have already found something you can fix. That is your first spark.
Hit reply and tell me: what is one condition in your team that makes good people look worse than they are? I read every answer.
Sponsored resource AI’s next bottleneck AI is not just a software story.
|
Keep going, one small spark at a time.
Erwin
PROMO CONTENT
Can email newsletters make money?
As the world becomes increasingly digital, this question will be on the minds of millions of people seeking new income streams in 2026.
The answer is—Absolutely!
That’s it for this episode!
Thank you for taking the time to read today’s email! Your support allows me to send out this newsletter for free every day.
What do you think for today’s episode? Please provide your feedback in the poll below.
How would you rate today's newsletter?
Share the newsletter with your friends and colleagues if you find it valuable.
Disclaimer: The "Tiny Big Spark" newsletter is for informational and educational purposes only, not a substitute for professional advice, including financial, legal, medical, or technical. We strive for accuracy but make no guarantees about the completeness or reliability of the information provided. Any reliance on this information is at your own risk. The views expressed are those of the authors and do not reflect any organization's official position. This newsletter may link to external sites we don't control; we do not endorse their content. We are not liable for any losses or damages from using this information.