I used to think the candidates who struggled most in interviews just weren't smart enough, or hadn't worked hard enough. Then I started actually talking to them about how they prepped, and I had to completely revise my view.
Most of the people who bomb interviews have put in real hours. They've read books. They've made flashcards. They've done mock interviews. They've practiced answers in the mirror. They are trying. The problem isn't effort — it's that the standard methods are broken in ways that aren't obvious until you look at the research on how learning actually works.
Here's the central problem: most interview prep is passive or blocked-practice. You read about STAR, then you practice STAR. You study "tell me about yourself," then you practice "tell me about yourself." You work on one type of question until you feel good at it, then move on.
This feels productive. It also doesn't match how skills transfer to new situations. Learning science has a name for it: "desirable difficulties." The most effective practice is harder in the short term and more flexible in the long term. It includes:
Almost no interview prep is structured this way. The result: candidates who've practiced extensively can't generalize. They can answer the exact questions they prepped, but struggle when the question is worded 30% differently, or when a follow-up takes the conversation somewhere unexpected.
The second issue: most prep has no honest feedback. Mock interviews with friends produce encouragement, not insight. Recording yourself is useful but uncomfortable to review, so most people don't. You end up optimizing against an imaginary version of the interview that's much friendlier than the real one.
"Good feedback has to be specific, accurate, and fast. Interview prep almost never delivers all three at once."
Real feedback — the kind that actually changes behavior — has to be specific (not "that was good, but maybe a bit long"), accurate (not softened by the relationship), and fast (ideally within seconds of the behavior you're trying to improve). Interview prep almost never delivers all three. Human coaches can be specific and accurate, but they're expensive and the feedback comes after the session. Friends can give fast feedback but rarely specific or accurate.
There's a cognitive bias called the fluency illusion — the feeling that you know something well because it sounds familiar when you encounter it. Interview prep is a breeding ground for this. You read the STAR method, it clicks, it feels intuitive. You answer a few practice questions and they go smoothly. You feel ready.
Then you're in the actual interview, someone asks a question you didn't see coming, and the framework you thought was automatic... isn't. Because you never actually tested it under conditions that resemble the real thing: unfamiliar questions, time pressure, someone watching you, social stakes.
The familiar always feels like the ready. They're not the same.
Run mock interviews with strangers, not friends. Use platforms where you get scored by someone who doesn't know you. If you're going to practice out loud, do it standing up, dressed as you would be, under a timer. The closer the practice conditions are to the real thing, the better the transfer.
Don't drill behavioral questions for an hour, then switch to technical. Mix them. Force your brain to shift gears constantly. It's harder, which means it's working. Blocked practice feels smoother — that smoothness is the problem, not the goal.
Close your notes and try to tell the story. Then check what you forgot. Don't re-read your notes and feel good about it — that's passive. Make yourself retrieve. The difficulty is the point.
The ideal feedback loop gives you data as it's happening. This is why coaches are useful — they're watching in real time and can tell you the moment you went too long on context, or the moment your energy dropped. When you don't have a human coach available, real-time tools can partially fill that role.
Deliberately practice what happens when you blank or stumble. Stop mid-answer, pause, say "let me take that from the top," and recover. If you've never practiced recovery, the first time you blank will feel like a catastrophe. If you've practiced it twenty times, it's just a moment you know how to handle.
Here's a view I've come to hold: the gap between preparation and in-session performance is partly a transfer problem, and partly a stress problem. Good preparation closes the transfer gap. But the stress problem — the cognitive overload, the retrieval failures, the anxiety spiral — has to be addressed in the session itself.
Athletes have coaches in real time. Surgeons have checklists and scrub nurses. Pilots have co-pilots and checklists for emergencies. The expectation that a job candidate, in one of the highest-stakes conversations of their career, should go in completely alone is a bit of an artifact of tradition rather than good design.
Real-time tools don't replace preparation — they supplement it. They're most useful when you've done the preparation and genuinely know the material, but the session conditions are making it hard to access. That's actually the majority of cases where people perform below their capability.
Before you add more to your prep, figure out what's actually broken. If you're not getting interviews: your resume and positioning need work. If you're getting interviews but failing early screens: your "tell me about yourself" story isn't landing, or your communication is unclear. If you're getting to late rounds but not offers: you're close, but probably losing on culture fit signals or specific technical gaps. If you're getting offers but below your target: that's a negotiation problem, not an interview problem.
Each diagnosis points to a different fix. More general practice doesn't solve all of them. Knowing which one you have is the starting point.
UnJam gives you quiet, in-session coaching — so the work you put into prep actually shows up when it counts.
Try it free