If you've ever come away from a "career chat" feeling more confused than when you walked in, the problem probably wasn't the person you talked to. It was the question you brought. Most career advice is bad because the question was vague, too big, or designed to be agreed with rather than answered.
You can fix that without becoming someone different. The shape of a good ask is small, specific, and surprisingly easy to write down once you know what you're looking for. Here's the version of this we wish someone had handed us a few years earlier.
It's not the advice. It's the ask.
When senior people give bad career advice, it's almost never because they don't know what they're talking about. It's because they've been asked a question that only has bad answers. "Should I move to product management?" is an unanswerable question from a stranger. "I'm a backend engineer with three years at a fintech, I have an offer to lead a small ML team, and I'm torn because I like writing code more than running standups - what would you trade off here?" is a different question entirely.
Same person. Same expert. Two completely different conversations.
What a bad career question looks like
Most useless career questions share a few traits. They're easy to spot once you know what you're looking at.
- Too abstract. "How do I grow in my career?" There's no situation, no constraint, no decision being made. The only honest answer is "it depends," which is exactly what you'll get.
- Too big. "Should I switch industries?" is a year of decisions wrapped into a sentence. Nobody can answer that for you in a 20-minute conversation.
- Yes/no without trade-offs. "Should I take this offer?" assumes the advisor knows what you're optimising for. They don't. Tell them.
- Looking for permission. If you've already decided and you just want someone to validate it, ask for a stress test instead. It's more honest and you'll get more out of it.
- No deadline. Without a deadline, the answer can't be prioritised. "Long-term, what should I focus on?" gets you a list. "I have to pick a stretch project by Friday - which one teaches me the most?" gets you an answer.
The four ingredients of a good ask
Almost every useful career question we've watched a Seeker bring to a Guide has four parts. They don't have to be in order, but they all have to be there.
1. The situation, in three lines.
Where you are now, what you've been doing, and what's just changed. "Three years into backend engineering at a Series B fintech, mostly Go and Postgres, just got an internal offer to lead a four-person ML team I haven't worked with before."
2. The two or three real options.
Not "what should I do?" but "I'm choosing between A and B, and B-prime is a fallback I'd take if A falls through." Naming the options forces you to do the first 40% of the thinking on your own.
3. What you're optimising for.
Money, scope, learning, optionality, location, sanity. You don't have to pick one - but you have to be honest about which one wins when they conflict. The advisor isn't a mind reader.
4. The deadline.
Give it a date, even if the date is loose. "I want to decide by next Friday" gets you a different conversation than "in the next quarter."
Where to ask - and where it always goes wrong
Where you ask matters as much as how you ask. A good question in the wrong venue still loses.
Public forums and Reddit threads are great for finding the table-stakes answer to a generic question. They're terrible for context-heavy decisions, because the people answering have to assume the median version of you.
LinkedIn DMs to senior strangers work about 5% of the time, and only when the message respects their time: short, specific, no "can I pick your brain?" phrasing. Most "no replies" aren't rudeness; they're triage.
Friends and former managers are the most generous and the least objective. They know you, but they also have a stake in how the conversation goes. Use them for emotional sanity-checks, not for the cold call on whether the move is right.
Paid 1-on-1 conversations with someone who's done the thing are usually the best return on time per rupee, especially for big decisions where you don't have the right person already in your network. You're not paying for wisdom in the abstract; you're paying for thirty minutes of focused attention from someone with the relevant context.
A template you can actually use
Copy this and edit it. It's deliberately short.
Situation: [3 lines about your role, recent context, what just changed]
Options I'm weighing: A) ___, B) ___, fallback) ___
Optimising for: [pick one or two; note which wins when they conflict]
Deadline: [a date, even a loose one]
What I'd love your take on: [the one specific thing - not "everything"]
A useful answer. Maybe a question you hadn't thought of. Sometimes a "talk to this other person instead" - which is itself a great outcome. Almost never silence, because the message respects the reader's time and signals you've already done the thinking they'd otherwise have to do for you.
What to do with conflicting answers
If you ask three people, you'll often get three answers. That's not a failure of the process - it's the process working. Here's how to read the spread.
- Look for what they agree on, not what they disagree on. The agreement is the table-stakes truth. The disagreement is usually about values, risk tolerance, or context they don't share.
- Map each answer to the advisor's situation. A founder will tell you to bet bigger. A risk-averse manager will tell you to optimise for stability. Both are giving you their advice, calibrated to who they are. You're calibrated to who you are.
- Pick the one whose constraints look most like yours. Not the most senior. Not the most successful. The one whose context is closest to yours is the one whose advice transfers best.
Want a focused 20 minutes with someone who's already made the call you're about to?
Amigzo lets you book per-minute time with Guides who've actually done the thing - not generic coaches. When we launch, you'll be able to ask the kind of question this post is about, and get the kind of answer this post is about.
The follow-up that earns more advice next time
The single highest-leverage thing you can do after getting good advice is to close the loop. Two weeks later, three lines: "Took option B. Here's how it's going. Thanks again." Almost nobody does this. It's the cheapest way to earn the right to ask again.
The same person who gave you a thoughtful answer the first time is now mildly invested in how it turned out. They'll answer faster the second time, and their advice will be sharper - because they've seen one round of how their input actually played out.
The 90-second version
- Most "bad career advice" is a response to a bad ask. Fix the ask first.
- Make the question small, specific, and bounded by a deadline.
- Name two or three real options, not an open question.
- Tell the advisor what you're optimising for. They can't read your mind.
- Conflicting answers are signal, not noise. Pick the one whose constraints match yours.
- Always close the loop. It earns you the next conversation.
Aanya K.
Editorial Lead, Amigzo
Aanya runs Amigzo's editorial. She's spent a decade between product and writing, and has done more "20-minute conversations" than she'd care to count - on both sides of the table.