Most candidates approach pre-interview research one of two ways: they over-research (spend six hours, exhaust themselves, and arrive with a memorized timeline of the company's founding) or under-research (skim the homepage twenty minutes before the call and sound generic in every answer). Both fail the test interviewers are actually applying.

This post is the 90-minute version of pre-interview research that hits what interviewers notice and skips what they don't.

What interviewers actually notice

Structured interview rubrics across most major hiring frameworks weight four things from a candidate's company research, in roughly this order:

A specific opinion about a product or strategy choice. The single highest-value research output. A candidate who says "I noticed you launched the SMB tier in March and I'd love to ask how that's tracking against the enterprise motion" is doing something most candidates don't: showing they read past the homepage, formed an opinion, and have an actual question. This is what separates serious candidates from generic ones.

Knowledge of recent strategic moves. Recent acquisitions, funding rounds, leadership changes, product launches. Mentioning one of these naturally in an answer ("I imagine the recent acquisition of $X means the platform team's roadmap got more complicated this quarter") signals preparation without making preparation the point.

Basic product fluency. What the product does, who uses it, what makes it different. This is the floor — necessary, not sufficient. Skipping it is a hard signal of low preparation.

Memorized trivia. Founding-year details, founder names, exact employee counts. These don't help. Interviewers don't reward memorization. They reward specificity in your thinking.

The research that produces good versions of the first two is the research worth doing. The research that produces only the second two is wasted time.

The 90-minute routine

A working pre-interview research routine fits in 90 minutes the day before. It has five parts:

Read the 'About' and product pages — 15 minutes. The basics. Skim the homepage, the product page, the 'about' page. Goal is conversational fluency — you can say what the company does in two sentences without searching for words.

Read the most recent funding or earnings news — 10 minutes. Crunchbase or the company's press page for private companies; the most recent 10-Q, earnings transcript, or analyst call for public ones. You're looking for the strategic narrative the company is publicly committing to right now. Growth area, restructuring, acquisition, expansion. This is the material interviewers are most likely to ask about indirectly, and the material your "do you have any questions" moment can hook into.

Read the JD again, twice — 15 minutes. The JD is the most important document about the role. Once for content, once for what isn't said. Mark the 3-5 must-haves. Identify the one or two responsibilities the JD over-emphasizes — those are usually the actual pain points the role exists to address.

Look up the interviewers on LinkedIn — 20 minutes. Each interviewer. Read their headline, current role, tenure, and a few recent posts. The goal isn't flattery — it's knowing who you're talking to. A 1:1 with someone three weeks into the company is a different conversation than a 1:1 with someone who's been there eight years.

Form one specific point of view about the company — 30 minutes. The most important step, and the one most candidates skip. Pick one specific thing — a product choice, a market position, a recent move — and form a real opinion. Doesn't have to be right; has to be specific. This is what becomes your "questions for us" question, and what makes your other answers feel grounded rather than generic.

90 minutes total. Done the evening before. The night-before scheduling matters — research done a week in advance doesn't stay fresh enough to surface naturally in conversation.

Why the "questions for us" moment matters more than candidates think

Structured interview rubrics consistently weight the candidate's questions as a meaningful chunk of the score — often 15-20% in formal frameworks. This is the part of the interview where the dynamic flips: the candidate is doing the asking, the interviewer is doing the answering, and the questions reveal what the candidate actually thinks the role is about.

Strong questions are specific and tied to real choices. Weak questions are generic.

Strong:

• "I noticed you launched the SMB tier in March. How is that tracking against the enterprise motion in terms of CAC?"
• "The JD mentioned the platform reorg. What's the biggest unsolved problem the team has six months in?"
• "You moved off the monolith last year, based on the engineering blog. What's the next architectural shift the team is anticipating?"

Weak:

• "What's the culture like?"
• "What do you like about working here?"
• "How do you measure success in this role?"
• "What does a typical day look like?"

The weak questions aren't wrong, exactly — they're just available to any candidate who didn't research. They don't differentiate. The strong questions require having read the company's blog, earnings call, or product release, which is the research the 90-minute routine produces.

What the routine specifically isn't

A few things the 90-minute routine does not require, and that candidates sometimes waste time on:

It doesn't require reading every Glassdoor review. A scan is fine. Reading 80 Glassdoor reviews mostly produces anxiety, not interview-useful insights.

It doesn't require memorizing the founder's biography. Interviewers don't ask about this. If they do, polite ignorance is fine ("I knew the company was founded around 2014 but I don't know the founding team well — I focused my research on the recent product moves").

It doesn't require reading every blog post the company has published. Pick the two or three most recent that touch the role's area. Skim the rest.

It doesn't require deep technical research on the codebase or product internals. A senior engineer can ask about technical choices in the interview itself. Pre-interview research is about the strategic and product surface, not the implementation.

What to do with the research during the interview

The research should show up in three places, in this order:

1. Your answers. "When I read about the SMB tier launch, the question I'd want to dig into is..." Tie one of your answers to one piece of research. Not all of them — that's overplaying it.

2. Your "tell me about why you're interested in this role" answer. This should reference one specific thing about the company. "What got me looking was the platform-engineering blog post about the migration off your monolith — that's the exact problem space I've been working in."

3. Your "questions for us" moment. Two specific questions tied to research, one open-ended question about the role. The specific questions are the differentiation; the open question gives the interviewer space to talk about what they care about.

For the broader interview-prep framework this fits into, see phone-screen-what-recruiters-evaluate and mock-interview-prep-timeline. For the specific question of how to use research to inform the "tell me about yourself" opener, see tell-me-about-yourself-90-seconds.

What this isn't

A few clarifications:

• It's not a license to over-prepare. 90 minutes is enough. Beyond that, you stop adding value and start sounding rehearsed.
• It's not the same routine for every interview round. The first recruiter screen needs less — 30 minutes is enough. The 90-minute routine is for hiring-manager and final-round loops.
• It's not a substitute for the actual interview prep on your own answers. Research the company; also practice your "tell me about yourself" and your behavioral answers. The research helps the answers land, but it doesn't write them.

The short version: 90 minutes, the day before, in five blocks. Goal is one specific opinion about the company that can become your "questions for us" hook. Memorized trivia isn't graded; specificity is.