🎯 A job title tells you who has power on paper; the process tells you who can actually change the outcome. When you look at an organization, do not ask “who is the biggest.” Ask “whose hands does this pass through from start to finish.”
I. The org chart tells you rank, not outcomes
When people first judge an organization, their instinct is to open the org chart: the boss on top, then the vice presidents, then managers, supervisors, employees. Follow that chart and the conclusion takes one sentence — the boss decides.
That is not wrong, but it is nearly useless.
Because in real life, a specific thing is rarely decided by one person from start to finish. Someone else prepared the proposal, someone else interprets the standards, someone else approved the budget, and someone else executes it. There is only one useful question:
Who does this pass through, from the beginning to the end? Who among them can change the outcome?
That is power flow: not looking at who has the biggest title, but at who screened, who interpreted, who blocked, and who executed the thing while it moved through the organization.
The thirteen kinds of power below are grouped into four sets by where they appear in a single piece of work. Start with the overview:
| Type of power | What it can do | Where you can see it |
|---|---|---|
| Final sign-off | Make the last choice among the remaining options | Who signs; who says “this is settled” in the meeting |
| Information control | Decide what decision-makers see | Who prepares the reporting deck; who filters the data |
| Interpretive right | Define what vague words mean | Who decides what counts as “excellent” or a “special case” |
| Veto | Stop the thing from moving | Whose refusal halts it |
| KPI control | Decide what everyone prioritizes each day | Which metrics are on the review form |
| Reward and punishment | Decide which behavior gets reinforced | Who gets promoted, who takes the bonus, who is actually punished |
| Execution control | Decide what the rule becomes in practice | What frontline staff actually do |
| Resource control | Decide who gets money, people, traffic | Who approves the budget; who allocates the slots |
| Expertise | Give a judgment on technical questions that others cannot refute | When someone says “non-compliant,” nobody dares push ahead |
| Access | Get heard by decision-makers more easily | Who can talk to the boss alone at any time |
| Relationship | Borrow someone else’s power to change the result | Whose single phone call can change a plan |
| Personnel | Decide who makes the decisions | Who promoted the people in key posts |
| Exception | Bypass the process | Who can skip the process |
💡 There is a comparable split in the academic literature. Social psychologists John French and Bertram Raven proposed five bases of power in 1959: legitimate, reward, coercive, expert, and referent power, and in 1965 Raven added information power. The thirteen below fit everyday work better and can be read as this framework unfolded inside a real process.
II. On the decision chain: who filters, who interprets, who can call a stop
1. Final sign-off: the person who signs only owns the last step
An internet company is changing office locations, and the CEO makes the final call: “We lease Building A.” It looks like the CEO decided this.
But trace it backwards. The admin department first looked at twenty buildings, and the budget cut ten of them. HR thought several locations were too far out, and cut five more. Finance considered three of them too expensive on rent. By the time the material reached the CEO’s desk, only options A and B were left.
So he did make a decision, but the eighteen options that were cut, he never saw.
Final sign-off looks like the largest power, but it only handles the last tap. What actually shapes the outcome are the people earlier in the chain who decide which options can be seen at all.
2. Information control: who decides what reaches the top
A factory’s orders are sliding. The sales director knows exactly how bad it is — three large customers are all preparing to cut their purchases. But at the weekly meeting he says: “There has been a bit of fluctuation in the market recently, overall still fairly stable.” He is afraid the boss will conclude that the sales team is incapable.
The boss judges on the basis of that sentence: “No need to adjust for now, let us watch for two more months.” Two months later orders are down 30%, and the boss is furious: “Why did no one tell me earlier?”
The problem is not that nobody knew. The information simply never reached him.
1) Information control shows up in four actions
- Which numbers go into the report and which stay on your own computer
- Which problems get escalated and which get a “let us keep an eye on it”
- Which customer complaints get dropped from the summary
- Which risks get written up as “small issues”
2) Why middle managers filter on purpose
Filtering is often not malice but self-protection: when bad news goes up, the first reaction is usually “how did you let it get like this.” So in many organizations, reporting good news is cheap and reporting bad news is expensive.
The result is that someone in a low-ranking position who sits at the information entrance can influence decisions more than many directors.
3. Interpretive right: one written rule, three departments, three sets of standards
A company has a written policy: “Employees who perform excellently may be promoted early.” It sounds clear until you ask what “excellent” means.
Manager A thinks hitting 120% of target is enough. Manager B thinks that on top of results, they must be able to onboard new hires. Manager C requires two consecutive quarters rated A. Same policy document, three sets of standards in practice.
The problem is not how well the policy is written but who is qualified to interpret it. So when you hit a vague rule, keep asking: who defines “excellent”? who recognizes a “special case”? who decides what counts as a “major problem”? who has the final say on whether something is “reasonable”? A lot of real power hides inside words like these.
4. Veto: cannot make a thing happen, but can make it stop
A product team wants to ship a new feature. Requirements are set, the boss approved, design is finished, it looks ready to launch. The security lead says one sentence, “This design carries a data leak risk, I do not agree to launch,” and the project stops on the spot.
He has no ability to make sure the feature ships, but he can decide that it cannot ship right now.
Similar cases: finance refusing a budget, legal refusing to sign off, engineering saying it cannot be built, procurement saying the supplier does not qualify. These people are not necessarily the highest-ranking, but they sit on the critical nodes. When you analyze anything, always ask one more question: who has the ability to say no?
III. When rules land: the rules on paper versus the rules that actually work
1. KPI control: the review form is the real instruction manual
An e-commerce company repeats “customer experience first” every day — it is on the wall, in the meetings, in the training. But the support team’s review criteria are: handle 150 tickets a day, average call length under 3 minutes, hourly reply volume on target.
So a customer calls in with a complicated problem that would take about 10 minutes to solve properly. The agent listens for a while and starts to rush, because another minute and the average will go over target. He says: “Let me log this for you, someone will contact you later.” Then he hangs up. The problem is unsolved, but the metric is safe.
The rule the company says out loud is “customer experience first.” The rule that actually works is “your review score must not drop.” To find out what an organization really values, do not listen to slogans. Look directly at who sets the KPIs and what the KPIs reward.
⚠️ Economics has a name for this: Goodhart’s Law — “When a measure becomes a target, it stops being a good measure.” Once a metric is tied to bonuses and promotions, people optimize the number itself rather than what the number was meant to represent.
2. Reward and punishment: watch who is promoted, who gets paid, who is actually punished
A sales company repeatedly stresses “do not overpromise just to close a deal.” But the person publicly praised every month is always the top seller by revenue, and bonuses are paid on revenue too.
One salesperson, to sign the contract, promised “this feature will definitely ship next month,” even though engineering had no such plan. The customer signed, and he took the bonus. Two months later the customer found the feature missing and started complaining. The company held another meeting about selling honestly, and the next month the sales champion took the award as usual.
After a few rounds, everyone understands that the real rule is not “do not overpromise” but as long as you can sell it, the rest can wait.
✅ To judge what an organization really encourages, three questions are enough: who got promoted? who took the biggest bonus? who was actually punished after making a mistake? These three answers are usually closer to the facts than any values slogan.
3. Execution control: one set at headquarters, another at the counter
A chain company’s headquarters sets the rule “no-reason refunds within 7 days of purchase,” sends out the document, and considers the matter settled. But one store manager thinks refunds are hurting the store’s numbers and tells staff: “Try not to refund if you can avoid it.”
A customer comes in to return something. Staff say the packaging is opened so it cannot be refunded. The customer points out that the website says it can. Staff reply: “Our store is different.”
So headquarters has one set of rules, and the customer meets another. The person who writes the rule and the person who applies it are usually not the same person, and what ordinary people experience day to day is almost always the executor’s version.
IV. Four kinds of influence that do not depend on rank
1. Resource control: whoever controls money, people, and traffic becomes a dependency
Marketing says it needs two designers, product says it needs three engineers, and on the surface the boss decides. But the boss asks finance first: “How many people can we still hire this year?” Finance answers: “At most three.” That one sentence has already locked most of the outcome in place.
Resources can be money, people, customers, traffic, data, servers, supply chain, inventory, advertising budget. Whoever holds those, others have to come to them. A purchasing manager may not rank highly, but if every critical raw material in the company passes through them, their real influence is not small.
2. Expertise: someone above you in rank may still have to take your word for it
A medical device company is launching a new product and the general manager is in a hurry: “This has to launch this month.” The engineer in charge of compliance says one certification item is still missing and it cannot be done yet. The general manager asks whether they can launch first and get certified later. The engineer answers: “Selling now could be a direct violation.” In the end it has to be postponed.
The general manager ranks higher, but on this question he has no judgment to offer. Network security, legal, finance, medical affairs, engineering, data, tax — as long as the question is specialized enough, people in high positions must depend on specialists. The source of this power is not rank but the fact that others do not know and you do.
3. Access: easier to be heard, easier to change the result
A strategy department has an ordinary employee, not high-ranking, who prepares the CEO’s weekly materials and therefore often gets to speak with him alone.
Once the company was considering cutting a project, and other departments could only wait for the formal meeting. While organizing the materials he mentioned casually: “Two large clients have been testing this project recently. Cutting it outright might be a bit of a waste.” The CEO said: “Really? Show me the data.” The next day the project was saved.
He has no formal decision authority at all. He is simply heard earlier and more easily than others. This is why chief of staff, secretaries, and strategy staff often have more influence than their rank suggests.
4. Relationship: borrow someone else’s power to change the result
An ordinary salesperson at a company has a middling rank but an excellent relationship with the largest distributor, which contributes 20% of the company’s annual revenue. The company plans to transfer him to another region, and the distributor’s owner calls the leadership directly: “If he is not our contact, we will have to reconsider next year’s cooperation.” The transfer was cancelled that same day.
He himself has no authority over personnel, but he holds a critical relationship. One way to think about relationship power: how many people with power are willing to use it for you.
V. The two most direct signals of status: personnel power and exception power
1. Personnel: deciding “who decides”
A vice president at a company rarely interferes with frontline matters and does not seem to have much presence. But the heads of three key departments were all promoted by him: the marketing head takes orders from him, the sales head is his former subordinate, and the operations head he recruited himself.
Many things do not need him to say a word — the department heads simply follow his thinking. Because you do not need to decide 100 things yourself; you only need to decide which 10 people decide those 100 things.
So to see someone’s real position, look at: who can promote people? who can replace a department head? who can put people they trust into key posts?
2. Exception: see whether rules stay hard in front of them
Company policy says any purchase over 100,000 must go through three quotes. Ordinary staff follow the process honestly. One day an executive says: “I know this supplier, sign it directly.” Purchasing asks whether the quotes are still needed, and he says: “This project is special, I will take responsibility for it.” And that was that.
Power differences inside an organization often show up on exactly this line:
| Position | What they can do when facing the process |
|---|---|
| Ordinary staff | Must follow the process |
| People with some authority | Can interpret the process |
| One level higher | Can modify the process |
| The top level | Can decide to skip the process this time |
Whoever makes the rules go soft is not in an ordinary position.
VI. How to use this: draw one thing as a process and ask at each link
1. Take “a raise” as the example
1) First draw the chain
Most people only think of one question: will the boss approve it? But the real process usually looks like this:
Employee requests a raise → does the direct supervisor recommend it → does the department head approve → HR checks the salary band → finance confirms budget → senior leadership signs off → salary is finally adjusted
Any link in this chain can stop the thing. The boss agreeing does not help if finance says there is no budget this year.
2) Then ask questions against the chain
- Who gives me my performance rating?
- Who is qualified to recommend a raise?
- Who can veto?
- Who controls the budget?
- Who can make an exception?
- Whose words does the leader ultimately take in?
Once you have asked these, what you see is not an org chart but the actual path power takes.
2. A general order of follow-up questions
Replace “who is the biggest” with “who decides this thing,” then keep breaking it down: who supplies the information? who interprets the rules? who executes? who controls the resources? who can veto? who can make an exception? who decides rewards and punishments? who can influence the person with the final say?
🎯 A job title tells you “who has power on paper.” The process tells you “who can actually change the outcome.” Walk one specific thing from start to finish and there is nowhere left to hide how the organization really works.