What a strong interview answer sounds like
Last updated: November 2026
Strong answers are not longer or more polished. They name one real case, say why you acted as you did, and say what happened. Below are weaker and stronger versions of short answers from four kinds of work. Every one is made up to show the rules. None is a real candidate.
What the scoring listens for
In an Aethel interview, each area's score comes from three ratings of your answers: the situation, your reasoning, and the result. The highest level of each, in the published words:
- Situation: a specific case, with where it was and how big.
- Reasoning: two or more options compared, with why each was rejected.
- Result: measured and then followed up over time, or adopted beyond the case.
Without a specific case, an area is at most 60, however good the rest sounds. That is why the first example is about the situation. Every level, the scale and the caps are in how AI interviews are scored.
The situation: name one specific case
A common weak answer describes how you usually work. It may be true, but there is nothing in it to rate. Name one case instead, and say where it was and how big it was, in words that fit your work: how many clients, tables, beds, tickets or accounts.
Weaker
“I always keep a close eye on my clients, and if something seems off I let the nurse know right away. Safety comes first with me.”
How the rules read it: Situation 0: no specific case, only general practice. However true it is, the area can score at most 60.
Better
“One morning a client seemed more confused than usual, so I called the nurse, and she came out to check on him.”
How the rules read it: Situation 1: a case, but without its context. We do not hear where it was or how big the work was, so the area is still at most 60.
Stronger
“This spring I had a morning visit with an older man who lives alone, one of five clients I see each week. He usually does the crossword with me, but that day he could not follow it, and he had not touched the water left by his chair the evening before. Before helping him with anything else, I called the agency nurse from his kitchen and stayed with him until she called back.”
How the rules read it: Situation 2: a specific case, with where it was and how big: a morning home visit, one of five clients on a weekly schedule. It still names no one. "An older man who lives alone" is enough.
The reasoning: say why, and what else you weighed
Once the case is clear, the next question is why you did what you did. A reason helps. Saying what else you could have done, and why you did not, helps more, because it shows judgment rather than habit.
Weaker
“On a Friday night the grill backed up, so I moved the salmon over to my sauté station and we got through the rush.”
How the rules read it: Reasoning 0: what they did, without why. The case is there, a Friday rush, but nothing says why moving the salmon was the right call.
Better
“On a Friday night I moved the salmon over to my sauté station, because the grill was overloaded and the fish was starting to stick.”
How the rules read it: Reasoning 1: a reason, or the cause they found. An overloaded grill explains the move, but nothing else was weighed.
Stronger
“I could have held the salmon tickets until the grill cleared, but six tables would have waited past twenty minutes. I could have asked the grill cook to push harder, but the fish was sticking because the grill was overloaded, so pushing would only have wasted plates. So I moved the salmon to my sauté pans, which had room, and told the expo why the sear would look different that night.”
How the rules read it: Reasoning 3: two options compared, holding the tickets and pushing the grill, with why each was rejected. Telling the expo why is also something the communication area can quote.
The result: say what changed, with a number
Many good stories end at "it went well". Say what changed, with a number if you have one, and who confirmed it. If you checked again later, or other people started doing it your way, say that too.
Weaker
“The new reconciliation checklist I wrote made month-end a lot smoother for everyone on the team.”
How the rules read it: Result 0 or 1: vague, or at best approximate and without a number. "A lot smoother" is a feeling, not a measure.
Better
“With my checklist, we closed in six working days instead of eight, and the controller signed off without sending anything back.”
How the rules read it: Result 2: measured, eight days down to six, and confirmed by someone else, the controller's sign-off.
Stronger
“With my checklist, we closed in six working days instead of eight, and the controller signed off without sending anything back. Four months later we still close in six, and the payroll team now uses the same checklist for its own accounts.”
How the rules read it: Result 3: measured and then followed up over time, still six days four months on, and adopted beyond the case by another team.
Communication: describe explaining, not just answering clearly
Every answer shows how clearly you speak, and a clear, on-topic answer with a concrete example does count toward communication. On its own, though, it counts only up to 80. A higher communication score needs you to describe explaining, writing or presenting to people: a customer, a patient's family, a coworker at handover, a manager.
Clear, but only clear
“The unit kept shutting off. I found the float switch had tripped because the condensate line was clogged, so I cleared and flushed the line and ran the system for half an hour to make sure the pan stayed dry.”
How the rules read it: A clear answer, in order, with a concrete example. It is good troubleshooting evidence, but for communication it shows only how clearly it was told, so on its own it counts toward communication only up to 80.
Describes explaining to people
“The tenant thought the unit was dying and wanted a new one, so I showed him the clogged line and the water in the pan, and explained that the switch shuts the unit off on purpose so it does not leak into the apartment below. Then I wrote on the work order that the line should be flushed every spring, with the cost, so the property manager could put it on the schedule.”
How the rules read it: This describes explaining to people: showing the tenant the cause, saying why the unit shut off, and writing a plan with its cost for the manager. That is the kind of evidence a communication score above 80 needs.
Say the important part in full
Scores are only as strong as the quotes behind them, and the quotes are your exact words. An area can score 90 or more only when one of its quotes is complete and at least 20 words long. When its best quote has fewer than 12 words, the area is at most 74. So say the key part of your answer as a full sentence, not as a shorthand.
Too short to quote
“I weighed both and picked the second.”
How the rules read it: 7 words, and it does not say what "both" were. A quote needs at least 8 words to count at all.
Long enough to carry a score
“I weighed holding the tickets against moving the salmon, and I moved it because six tables would have waited far too long for their food.”
How the rules read it: 25 words, ending at a full stop, and it shows the reasoning on its own. A complete quote of at least 20 words is what an area needs to score 90 or more.
These snippets are short on purpose. In a real interview, one answer often covers all three ratings at once, and follow-up questions build on what you said.
Common ways a good story loses points
- "I always" or "we usually": general practice rates situation 0, however true it is.
- Only "we": saying what the team did, without your own part, leaves little to quote about you. For a teamwork area, naming your team is not evidence on its own.
- Stopping at what you did: without the why, reasoning rates 0.
- Ending at "it went well": a result with no number and no one confirming it rates low.
- Two stories in one answer: neither gets its full context, so each is harder to rate.
- Names and private details: leave out the names of patients, clients, children and customers. They add nothing to a score, and in every field except software engineering the interviewer is told not to ask for them.
Full worked examples by field
Each field page has a full worked example for that field, with the four areas it is scored on and its practice questions. Pick yours:
- Healthcare support and caregiving interviews
- Retail, food and hospitality interviews
- Childcare and early education interviews
- Customer service and support interviews
- Skilled trades and facilities interviews
- Administration and office support interviews
- Finance and accounting interviews
- Education and training interviews
- Law and compliance interviews
- Software engineering interviews
Not listed here? See the list of fields Aethel interviews for.
Questions about strong answers
Is it bad to say "we" in an interview answer?
No, but say "I" for your own part. Saying "we" or naming your team is not, by itself, evidence of teamwork. Your words need to show you working with someone: agreeing on a plan, settling a disagreement, or helping a coworker.
Can I talk about a mistake I made?
Yes. A mistake you noticed, owned and fixed can be a strong case, because it has a situation, a reason for what you did next, and a result. Say what you did about it and what changed afterwards.
What if my result is only an estimate?
Say that it is an estimate, and how you got it. In the scoring rules, an "approximate, hedged" result sits one level below a "measured" one. If someone else confirmed it, such as a supervisor, a nurse or a customer, say so: "confirmed by someone else" sits at the same level as "measured".
Does a longer answer score higher?
No. Length is not rated. A long answer without a specific case is still at most 60 for that area. A shorter answer that names the case, says why, and gives the result can rate much higher.
Should I memorize my answers word for word?
No. Follow-up questions build on what you said, so a script rarely fits. Know your cases well instead: where and how big, what you did and why, and what happened. Then tell them in your own words.
More from the interview guide
- How AI interviews are scored
How the Aethel AI interview scores answers: what counts as evidence, the three ratings, the fixed scale, the caps, and what is never scored.
- How to prepare for an AI interview
A step-by-step plan for any AI interview: pick three real cases, practice out loud, check your tech, choose accommodations, and know what to do if it breaks.
- How the AI interview works: the full interview guide
What to expect on the day, the tech check, accommodations, and what happens after.
Try it with your own cases
Warm up with your field's practice questions, then take the free interview and see the quotes behind every score.