Escalation Done Right

Negotiation Basics was about reaching agreement with someone directly, across a table. But sometimes you can’t reach it — the other person won’t move, the decision is above your pay grade, or a risk is growing that no one with the power to act can even see. That’s when you escalate: you take the problem up a level to someone who can resolve it. And for most people, that word lands with a thud. Escalation feels like tattling, like going over someone’s head, like admitting you couldn’t handle it yourself.

This page rewires that instinct. Done well, escalation is not a weapon and not a confession — it’s a normal, healthy tool for unblocking work. The difference between escalation that helps and escalation that poisons a relationship comes down to a few concrete habits. This page teaches them, so you can escalate to solve a problem rather than to assign blame.

What escalation actually is

Strip away the emotional weight and escalation is simple: bringing a decision or a blocker to the level that can resolve it. That’s all. It’s routing a problem to where the authority or the information to fix it actually lives.

Notice what that definition does not say. It says nothing about fault. It says nothing about reporting a person. It doesn’t imply you failed. Most of the dread around escalation comes from smuggling those meanings in — treating “I need a decision from above” as if it meant “I couldn’t cope” or “someone else messed up and I’m telling on them.” Peel those off and what’s left is a logistics problem: this thing is stuck at my level, and it needs a higher level to move.

Every organization runs on escalation constantly, and healthy ones barely notice it. A nurse pages a doctor because a patient’s situation exceeds what the nurse is authorized to decide. A barista calls the manager because a customer is demanding a refund larger than the barista can approve. A warehouse lead flags to the operations director that two departments both need the same forklift on the same morning. None of that is tattling. It’s work reaching the level that can act on it. The goal of this page is to make your escalations feel exactly that ordinary.

The common trap — treating escalation as a nuclear option

Because it feels aggressive, many people hoard escalation until they’re desperate — they let a blocker sit for three weeks, tell no one, and then escalate in a panic when the deadline is already blown. By then it is explosive: it’s late, it’s emotional, and it lands on the higher-up as a crisis rather than a decision. The irony is that treating escalation as a last resort is exactly what turns it into a weapon. Escalate a small problem early and it’s a two-minute conversation. Hoard it and you’ve built the bomb yourself.

When to escalate

If escalation is just routing a problem to the right level, the real question is: which problems belong at a higher level? Three do, cleanly.

• A genuine blocker. Something outside your control has stopped your work, and you’ve exhausted what you can do about it directly. You need another team’s data and they’ve gone quiet for a week. A budget is frozen and no one will unfreeze it. A dependency is late. You are stuck through no fault of effort, and only someone with more reach can free it.
• A decision above your authority. The choice in front of you is real and needs making, but it isn’t yours to make. Cutting a feature to hit a date. Spending beyond your limit. Changing a commitment to a customer. Trying to decide it yourself would be overstepping; sitting on it stalls everything. The fix is to hand a clean decision to the person who owns it.
• A risk others need to see. Nothing is blocked yet, but you can see something coming that the people above you can’t — a deadline that’s quietly slipping, a safety concern, a corner being cut that will hurt later. Here escalation is early warning. Staying silent to avoid being “that person” is how small risks become big surprises.

Before any of these, run one honest check: is this actually stuck, or am I just avoiding a hard conversation? If you haven’t yet tried to resolve it directly with the person involved, you’re not ready to escalate — you’re ready to talk to them. Escalation is for what direct effort can’t move, not a shortcut around the discomfort of a direct conversation.

Should I escalate?  →  Ask in order:

1. Have I tried to resolve it directly?     → No:  go do that first
2. Is it genuinely stuck, or a real
   decision / risk above my level?          → No:  keep working it
3. Have I warned the person involved?        → No:  warn them first
4. Do I have a recommendation ready?         → No:  build one, then escalate
   ————————————————————————————————————————
   All yes → escalate, cleanly and early

In practice — the data that never came

A project lead needs a report from the finance team to finish her forecast. She asks; nothing. She asks again; a vague “soon.” A week passes and her own deadline is now three days out. This is a genuine blocker: she’s done everything in her power, and finance doesn’t report to her. She messages finance’s manager and her own — together, not behind anyone’s back: “I need the Q3 spend export to close the forecast by Thursday. I’ve asked twice and haven’t got a date. Can we agree one today?” No blame, a clear ask, a clear deadline. It’s resolved in an hour. Had she waited until Thursday, it would have been a missed deadline and a fight.

Escalate the problem, not the person

Here is the single habit that decides whether escalation builds trust or burns it: escalate the problem, not the person.

The tattling version puts a human at the center: “Marcus never replies to my emails and it’s dragging everything down.” Now the higher-up’s attention is on Marcus — is he lazy, is this a performance issue, who’s right? You’ve started a small trial, and Marcus, when he hears about it, will remember it for a long time. The work is no closer to moving.

The constructive version puts the problem at the center and describes it the way a neutral observer would: impact, then options.

• Impact — what is stuck and what it costs. Facts, not character. “The design sign-off has been open for eight days; until it closes, the build team is idle and we’ll miss the launch window.”
• Options — the realistic ways forward, so the higher-up is choosing a path, not hunting for a culprit. “We can extend the deadline a week, pull in a second reviewer, or ship the earlier version. I’d suggest the second reviewer.”

Same underlying situation. But one version asks “who’s to blame?” and the other asks “what should we do?” The second is faster, calmer, and it leaves every relationship intact — including the one with Marcus, who was never put on trial. You can name people as facts (“I’m waiting on design sign-off, which sits with Marcus’s team”) without making them the charge. The test: are you describing a bottleneck, or building a case against someone? Describe the bottleneck.

The evidence — blame is corrosive to how teams work together

Research on high-performing teams — Google’s multi-year Project Aristotle study is the best-known example — generally finds that the strongest predictor of team effectiveness is psychological safety: the shared sense that you can raise a problem or admit a gap without being punished or humiliated. Blame-shaped escalation attacks that directly. Every time a raised problem turns into someone being singled out, people learn to stay quiet — and the risks that most need surfacing are exactly the ones that then stay hidden. Escalating problems rather than people isn’t just politeness; it’s protecting the thing that lets a team surface trouble early at all.

Never let it be a surprise

The fastest way to turn a reasonable escalation into a betrayal is to spring it on the person involved. If the first Marcus hears of the sign-off problem is his boss asking him about it, you haven’t escalated — you’ve ambushed him. He will not forget it, and next time he’ll route around you.

So the rule is: give warning, and try direct resolution first. In order:

1. Try to solve it directly. Talk to the person. Most blockers dissolve here, and doing this first is what makes escalation legitimate rather than lazy.
2. If it’s still stuck, tell them you’re going to escalate — and why. “We’re not getting unblocked at our level and the deadline’s real, so I’m going to raise it with both our managers to get a decision. I wanted you to hear it from me first.” This one sentence changes everything. It’s respectful, it’s honest, and it often solves the problem on its own — people move when they know it’s about to go up.
3. Then escalate, ideally with them in the loop. Where you can, escalate together or copy them in, so it’s plainly a shared attempt to unblock, not a report filed against them.

The principle underneath all three: no surprises. The person involved should never learn about your escalation from someone above them. Warning them costs you a slightly awkward sentence and buys you a relationship that survives the escalation — and the reputation of someone who is straight with people even when it’s uncomfortable.

The common trap — the silent CC to the boss

The most relationship-damaging move in office life is quietly copying someone’s manager on a complaint email without telling them. It feels efficient and safe — you’re “just keeping people informed.” To the person CC’d, it’s a public accusation with their boss watching. It detonates trust instantly and teaches everyone around you to be careful what they put in writing near you. If a manager genuinely needs to be in the loop, tell the person first, or add them openly with a line explaining why. The secret CC is never worth what it costs.

Make it easy to act on

The last habit is what separates an escalation that gets resolved today from one that sits in someone’s inbox for a week: bring a recommended path and name the decision you need.

A higher-up is busy and context-poor. An escalation that says only “we have a problem with the sign-off, can you help?” hands them homework — they now have to investigate, gather context, invent options, and choose. That’s slow, and slow is often the whole failure. Your job is to do that work for them and hand over something they can act on in two minutes. A good escalation has four parts:

1. The situation   — one or two sentences of plain context
2. The impact      — what's stuck and what it costs (facts, not blame)
3. The options      — the 2–3 realistic paths forward
4. The ask          — the specific decision you need, and by when
                      (include your recommendation)

Written out, it looks like this:

To: my manager + finance manager

Situation: I need the Q3 spend export to close the forecast, due Thursday.
Impact:    I've asked finance twice over a week with no date; if I don't have
           it by Wednesday, the forecast slips and so does the board pack.
Options:   (a) finance sends it by Wed AM, (b) I use the Q2 figures with a
           caveat, (c) we push the board pack a week.
Ask:       Can we commit to (a) today? I'd strongly prefer it. If not, I'll
           fall back to (b) unless you'd rather (c). Need to know by 3pm.

Notice how little is left for the reader to do. The decision is teed up, the recommendation is stated, the fallback is clear, and there’s a deadline on the decision itself, not just the work. This is the difference between escalation that feels like a burden and escalation that feels like a gift — you’ve made the higher-up’s job easy, and busy people remember who does that.

Bringing a recommendation does one more quiet thing: it keeps you in the driver’s seat. You’re not surrendering the problem, you’re proposing a solution and asking for a decision on it. That reads as ownership, not helplessness — which is the exact opposite of the “I couldn’t cope” fear that makes people avoid escalating in the first place.

Case study — Toyota’s andon cord

On Toyota’s assembly lines, any worker who spots a defect can pull a cord — the andon cord — that signals a problem and can stop the line. What’s striking is the culture around it: pulling the cord is not treated as failure or as snitching; it’s treated as exactly the right thing to do, and a line that never gets stopped is seen as a line hiding problems, not a healthy one. The lesson for escalation is precise. When surfacing a problem early is normal, expected, and blame-free, problems get caught small — at the station, in minutes — instead of shipping downstream to become expensive. An organization’s health shows in how safe it is to pull the cord. Escalating well is how you make pulling it look easy.

Reading the response

Escalate cleanly and one of a few things happens, and each teaches you something. If the decision comes back fast, you’ve confirmed the problem really did belong at that level — good routing. If the higher-up hands it back — “work this out between yourselves” — that’s useful information too: it may mean you escalated too early, before trying hard enough directly, or that they trust you to solve it. Either way, don’t take it as a rebuke; take it as a recalibration of where the line sits. And if the same kind of blocker keeps forcing you to escalate, that’s a signal in itself: a recurring escalation usually points to a broken process, not bad luck. Fix the process and the escalations stop — which is the best outcome of all.

Try this

The next time something at work is stuck, before you complain about it or sit on it, write the four-part escalation for yourself: situation, impact, options, ask. Force yourself to describe the impact in facts a neutral observer would recognize — no character, no fault — and to bring at least two options with a recommendation. Then check the two gates before you send it: have I tried to resolve this directly, and have I warned the person involved? Even if you decide not to escalate after all, writing it this way will clarify what the problem actually is — and half the time, the act of laying it out shows you the direct conversation you should have first.

Reflect

1. What’s your gut reaction to the word “escalation” — tool, weapon, or admission of failure? Where did that reaction come from?
2. Think of a blocker you’re sitting on right now. Is it genuinely stuck above your level, or are you avoiding a direct conversation you could have this week?
3. Recall a time you escalated (or were escalated about). Was the problem at the center, or a person? How did that land?
4. Have you ever learned about an escalation “against” you from someone above you? What did that do to your trust in the person who raised it — and what does that tell you about your own escalations?
5. When you raise a problem upward, do you tend to hand over homework or hand over a decision? What would it take to always bring a recommendation?

Show reflections

1. Most people flinch — the word carries tattling and failure baggage from somewhere, often a boss who treated it as a threat. Naming the source helps you separate the baggage from the plain, healthy definition: routing a problem to the level that can solve it.
2. The honest answer is often “avoiding.” The test in the page is decisive: if you haven’t tried to resolve it directly, you’re not ready to escalate — you’re ready to talk to the person. Escalation is for what direct effort genuinely can’t move.
3. Person-centered escalations start a trial and damage relationships even when you “win.” Problem-centered ones — impact, then options — keep everyone intact and move faster. If a past escalation landed badly, look for whether a human was at the center of it.
4. A silent CC or an ambush is remembered for years and quietly teaches people to route around you. Feeling it from the receiving end is the strongest possible argument for the “no surprises” rule — warn the person first, every time.
5. Handing over a raw problem is slow and reads as helplessness; handing over a teed-up decision with a recommendation is fast and reads as ownership. Building the four-part habit (situation, impact, options, ask) is what makes the second one automatic.