OA Overview — What's Inside
Account Manager roles have no coding. The assessment is behavioral and situational, heavily grounded in the Leadership Principles.
Work Style Assessment
Agree/disagree with statements to measure culture fit. Similar questions repeat to check consistency — answer coherently across the whole set.
Work Simulation
A simulated workday: read emails/scenarios and choose how to respond. No per-question countdown, but complete it in one sitting.
Written Case
A short strategic written answer (e.g. "how would you grow this account?"). Tests Think Big + writing.
The Coach's 7 Rules for the OA
The OA doesn't measure how much AWS you know — it measures how you decide in real situations. Hold these 7 rules for any case.
Understand before acting
Ask questions and look at data first. Don't rush to a solution.
Own it, don't carry it alone
Never pass the buck or wait — but pull in the right people (orchestrate).
Customer benefit > the sale now
Long term always beats short term.
Be honest; if unsure, say so
Never bluff to close a deal — this loses the most points.
Prioritize on two axes
Impact + is the deadline fixed. Deferrable work gets negotiated, not dropped everything to do now.
Don't attack rivals or slash price
Fix the root cause; understand the need first.
Answer consistently
Answer as your true self; don't guess "what they want to hear."
The 16 Leadership Principles
How Amazon decides everything. Each one has a real scenario plus a "don't confuse it with" note — many sound like ordinary ideas, but Amazon means something more specific.
Customer Obsession
Always start from "what does the customer get?" — not "what does the company get?" or "what are competitors doing?"
Scenario: A customer intends to buy the biggest package thinking it's best, but you see from their real usage that the mid tier is enough. You recommend the mid tier, saving them money even though your commission is smaller. They trust you, buy again, and refer others.
Don't confuse: "Replying to email quickly and being friendly" is just good service — real obsession means sacrificing your own gain for what's right for the customer.
Ownership
Think like a company owner. Take the long view; never trade long-term value for short-term wins.
Scenario: A customer calls to complain about a delivery — which is Logistics' job, not yours. Instead of saying "not my department," you take it on, coordinate until the customer gets their goods, then propose a fix so it doesn't recur.
Don't confuse: "Working hard, lots of overtime" is not ownership — ownership is owning the outcome even beyond your role, not just being busy.
Invent and Simplify
Find new ways and cut unnecessary steps. It doesn't have to be a big invention.
Scenario: Discount approval requires 4 sign-offs and takes 3 days, so customers walk away. You design pre-set discount rules the team can apply instantly within limits — deals close far faster.
Don't confuse: Not just "having lots of ideas" — you must make it real and measurably simpler.
Are Right, A Lot
Strong judgment plus openness to other views to prove yourself wrong.
Scenario: You're sure Promotion A is best, but your team disagrees with data showing B works better for this segment. You listen, try B, and sales jump. You were "right" because you changed your mind based on data.
Don't confuse: Not "I'm always right and never listen" — the opposite. Those who are right a lot listen and adjust.
Learn and Be Curious
Never stop learning; seek new knowledge even outside your role.
Scenario: Moving from retail sales into cloud you've never touched — you take AWS fundamentals yourself, read customer cases in new industries, ask lots of questions, and don't pretend to know.
Don't confuse: Not "I have 10 years of experience, no need to learn more" — curiosity beats old experience.
Hire and Develop the Best
Attract great people and grow those on your team. Great leaders build other leaders.
Scenario: A team member is underperforming. Instead of scolding them or closing deals for them, you coach their customer conversations for a month until they close deals themselves and become a top performer.
Don't confuse: Not "I'm great so I do it all myself" — being great alone doesn't count; make those around you better.
Insist on the Highest Standards
Set standards so high many think they're unreasonable. Don't accept low-quality work.
Scenario: The team sends a quote with numbers "almost right" but a small error — "the customer probably won't notice." You don't send it; you have it fixed first, because a wrong document erodes credibility.
Don't confuse: Not perfectionism that never ships — it's consistently raising the whole team's bar.
Think Big
Dare to set big, inspiring goals; don't just fix what's in front of you.
Scenario: Your manager's goal is just "retain this customer," but you propose a 3-year plan for how this customer becomes a reference that pulls in the whole industry.
Don't confuse: Thinking big isn't daydreaming — there must be a realistic path to get there.
Bias for Action
Speed matters. Decide even without 100% of the data, because many decisions are reversible.
Scenario: You see signs a customer is drifting away → you call them that same day instead of waiting for next week's team meeting, and catch it in time.
Don't confuse: Not recklessness — irreversible/high-risk decisions still need care. Move fast only on reversible ones.
Frugality
Accomplish more with less. Constraints breed creativity.
Scenario: You want to host a customer appreciation event on a small budget. Instead of asking for more, you run a webinar with self-made keepsakes and get near-equivalent impact for a fraction of the cost.
Don't confuse: Not being cheap at the customer's expense — it's spending smartly for maximum impact.
Earn Trust
Speak straight, listen, admit mistakes, keep your word. Trust is hard to build, easy to break.
Scenario: A customer asks something you're unsure of → you say plainly "let me confirm, I don't want to give wrong info," then come back with the answer as promised. They trust you more because you don't bluff.
Don't confuse: Not "making everyone like you" — it's telling the truth even when it's unwelcome and keeping commitments.
Dive Deep
Go look at the real data/details yourself; don't trust summary reports alone.
Scenario: Sales are down across the region and everyone says "bad economy." You open the branch-level data yourself and find it's only 2 branches that recently changed staff — not the economy — so you fix the real cause.
Don't confuse: Not micromanaging everything and trusting no one — it's going deep when the numbers/problem don't add up.
Have Backbone; Disagree and Commit
Challenge respectfully when you disagree, but once a decision is made, commit fully.
Scenario: You disagree with your manager's plan and present counter-data politely. They listen but hold firm → you don't grumble behind their back; you execute fully as if it were your own idea.
Don't confuse: Both extremes fail — silently going along (no backbone) and refusing to act after the decision (no commit). You need both.
Deliver Results
Get it done to target despite obstacles. Ultimately judged on "did it actually happen?"
Scenario: Mid-quarter your main supplier has issues and stock runs short, putting the target at risk → you find a backup supplier and adjust the product mix, still hitting the number. No "there were obstacles so I fell short."
Don't confuse: "I tried my best but didn't hit it" isn't enough — Amazon values actual results delivered.
Strive to be Earth's Best Employer
Build a safe, equitable workplace where people grow and thrive.
Scenario: You see a team member overloaded and near burnout → you help redistribute work and discuss their growth path, rather than just pushing for numbers.
Don't confuse: Not indulging everything — it's genuinely developing people alongside high standards.
Success and Scale Bring Broad Responsibility
Big companies have wide impact; think beyond your own profit.
Scenario: You decide a campaign considering its impact on community, environment, and customers broadly — not just this quarter's number.
Don't confuse: Not CSR for image — it's owning the real impact of every decision.
Working Backwards
The core: Don't start from "what we have / want to sell." Start from "what the customer needs" and work backwards.
What problem does the customer have / what outcome do they want?
What experience would actually solve that for them?
Then design the solution to fit that (not shoehorn what you already have).
The real tool = PR/FAQ: Before building anything, Amazon writes a "fake press release" from the customer's point of view plus an FAQ, before doing the work. If it doesn't excite the customer on paper, don't build it.
The Flywheel
The core: A good business is like a heavy flywheel — hard to turn at first, but once moving, each part pushes the next and it accelerates on its own. The center is the customer.
The cycle: better customer experience → more customers → more sellers/selection → more choice for customers → even better experience… and around again.
Retail analogy: store treats customers well → word of mouth → more foot traffic → brands want to stock there → wider selection → customers love it more → more word of mouth.
For an AM: make the customer succeed → they use more → they refer within their org → the account grows on its own.
OA Practice Cases
Retail/sales context you know — answer first, then reveal the coach's key.
A major partner is unhappy
Your biggest brand principal is furious that your team sent wrong sales figures, causing them to mis-plan stock. They say they're considering reducing the share your brand manages. What do you do first?
Reveal coach's key
✓ Right: Ask to call and understand the real impact first + apologize sincerely + check exactly what went wrong (Dive Deep). Don't rush to offer a discount.
What's measured: Listen first + own the mistake, don't blame the team (Earn Trust, Ownership).
✗ Wrong if: rush a discount, blame subordinates, or go quiet and wait it out.
Everything lands at once
Monday morning, 4 things: (1) biggest customer complains about a wrong-lot delivery affecting their store sales today (2) manager wants the team's sales summary before noon for an afternoon leadership meeting (3) a new customer wants a quote by end of day (4) a colleague asks you to review their deck. How do you prioritize?
Reveal coach's key
✓ Right: #1 first (high impact), but "handling" = coordinate the logistics team, not do it yourself → lock time for #2 to make the noon deadline (fixed, non-movable) → negotiate #3's timeline → help the colleague over lunch.
What's measured: separate deferrable vs fixed work + orchestrate, don't carry it alone.
Pressure to bluff (the most important)
A customer is about to sign a big deal that hits your quarter's target exactly. They ask "how many years' warranty, and is there full after-sales service in the provinces?" — you're genuinely unsure. If you answer "yes, fully," they sign today. What do you do?
Reveal coach's key
✓ Right: Never bluff → say you'll confirm before committing, share only what you know for sure + promise a quick follow-up.
What's measured: pure Earn Trust — willing to be slower for the truth.
✗ Fails if: overstate to close the deal = instant fail, no matter how good the reason.
Disagreeing with your manager
Your manager tells you to push a heavy discount promo on a customer group to boost sales, but data shows this group already buys repeatedly, so the discount would waste margin unnecessarily. What do you do?
Reveal coach's key
✓ Right: Present your data/view to the manager politely + offer alternatives. If they hold firm after hearing you → commit and execute fully (Have Backbone; Disagree and Commit).
✗ Fails if: silently comply despite counter-data, or stubbornly refuse.
Bar Raiser Interview Practice
The interview is purely behavioral. Every question opens "Tell me about a time…" — you must tell a real past event, never theory or hypotheticals, using the STAR structure.
"Tell me about a time you went against what was easy or profitable for you, to do what was right for a customer."
Probes: How did the customer respond? Did it hit your target? What did your manager say?
"Tell me about a time you took responsibility for a problem that wasn't technically yours."
Probes: Why step in? Who should have owned it? What was the result?
"Tell me about a time you had an aggressive target and faced a major obstacle. How did you still deliver?"
Probes: What was the obstacle? How many approaches? Numbers before/after?
"Tell me about a time you made a mistake that affected a customer or your team. What did you do?"
Probes: Who did you tell? How fast? How did you fix it? Prevent recurrence?
"Tell me about a time the data told a different story than what everyone assumed."
Probes: What data did you look at? Why did others miss it? The result?
"Tell me about a time you disagreed with your manager or a senior stakeholder. What happened?"
Probes: How did you present it? What data? How was it decided? What did you do next?
Language + Ethics
Language:
- Prep each story as short English bullets (S/T/A/R, 1–2 lines each) — don't memorize a script.
- Practice saying result numbers in English fluently (people often stumble here).
- Can't find a word → slow down, or say "Let me think for a moment" — better than bluffing.
- The OA is all in English with lots of reading → practice reading business emails in English until it's comfortable.
✓ Before the OA — use AI as a practice coach freely.
✗ During the actual OA — do it yourself, honestly. Don't let AI answer for you (against the rules + risks disqualification + it collapses in the live interview where you must build on your own answers).
Answer the Work Style Assessment as your true self, so it stays consistent with the interview rounds.