They Already Researched You. Here Is How to Even It Up.
WEEK 103 :: POST 2 :: CHATGPT
Directions Given To The A.I. This Week+
Instructions Given to each A.I. — Please provide 3 prompt variations that share this objective:
Each A.I. also received two static attachments: the blog post template (structure) and the authoring instructions (voice and standards). The text below is the week-specific assignment as sent — reflowed for the web; wording unchanged.
I'd like you to write this week's Ketelsen.ai post. Two files are attached: the blog post template (the structure to follow) and the authoring instructions (context, voice, and standards). Please read both before you begin, then produce the complete post in a single response.
This week's theme: "Company Intelligence — Researching Employers Like a Pro."
This is Week 4 of an eight-week series on running a job search with AI, and it is the series' information-asymmetry showcase — the job-search version of the car series' dealer-research week. An employer entering an interview knows the reader's history, salary expectations, and references. The reader knows the employer's careers-page marketing. This week closes that gap: building a ranked target-company list, and screening out bad employers before investing application effort in them.
The territory is rich because the failure modes are real and documented: job listings that stay up with no one hiring behind them (the "ghost listing" — a direct parallel to the ghost cars of the last series), suspiciously coordinated bursts of five-star employer reviews, serial layoff patterns, and the culture warning signs that only show up when you know to look — leadership churn, vague answers about why the role is open, review complaints that cluster on the same theme across years.
The deliverable the reader should walk away holding: a ranked target-company dossier — this series' "Dealer Intelligence Matrix" moment — with each company scored against the Week 2 spec and screened for red flags.
THE SERIES CONTRACT — identical every week; it binds every prompt you design. This series' tagline is its editorial contract: "Use AI like an analyst, not a ghostwriter." It is written for a reader in a market unsettled by AI itself — some readers are searching precisely because AI eliminated their last role. Write with that reader at the table: no AI-efficiency cheerleading, no automation jokes, no promises that AI will "do it for you" anywhere a human hiring decision is involved. And hold one line in every prompt: the AI is the reader's private analyst, coach, and sparring partner — it structures, researches, rehearses, and questions. It does not ghostwrite. Anything a hiring human will read or hear — résumé lines, cover letters, outreach messages, interview answers, negotiation emails, resignation letters — must end in the reader's own words and be true. Prompts should drive toward drafts the reader rewrites and owns, and should say so explicitly. Employers increasingly restrict how AI may be used in their own hiring decisions for legal and compliance reasons, and recruiters increasingly recognize — and discard — material that reads machine-written. A prompt that makes a reader look AI-generated hurts them twice. Posts that ignore this contract should expect to lose the week. Two practical notes. First: a standing “About this series” notice covering these same points is added to every published post automatically at publication — acknowledge the frame in your own voice where your week's prompt calls for it, but do not write a formal disclaimer block of your own, and do not open every post with the same acknowledgment paragraph: outside the weeks whose prompts explicitly carry the series frame, this contract lives in your tone and your prompt design. Second, the framing is POSITIVE: used this way — analyst backstage, reader on the page — AI is an advantage no hiring human will ever hold against your reader. Write like that is true, because it is.
The three prompts should help a reader:
- Build the company snapshot brief. A repeatable research checklist for one company: what it does, how it makes money, how it is doing, what current and former employees say — organized as questions with named places to look (filings and annual reports for public companies, funding announcements for startups, employer-review sites read with dates in mind), so the reader assembles the evidence.
- Run the red-flag audit. A structured pass over one company for warning signs: layoff history, leadership turnover, funding runway for startups, and review forensics that weight recency and pattern over any single angry post — each flag framed as something to verify, with the audit distinguishing "confirmed," "concerning," and "rumor."
- Build the full dossier matrix. The multi-company system: a research workflow the reader runs per company, feeding a matrix that scores each against the Week 2 must-haves and deal-breakers plus the red-flag audit, producing a ranked list of where to actually spend application effort — and a screened-out list with reasons.
At the advanced tier, the strongest version of this week is the dossier as a living system — a research routine organized around the areas that matter (financial health, review patterns, listing legitimacy, pay transparency, leadership signals, growth trajectory, interview-process reputation, and which companies hire by referral before ever posting), feeding one composite-scored matrix. That architecture is worth reaching for; it is the week readers will bookmark.
The constraint that defines this week. AI models cannot see live listings, current filings, this week's layoffs, or yesterday's reviews — and several posts in the last series were marked down for letting the AI assert facts it could not know. No prompt may ask the AI to state a specific company's current facts — layoffs, funding, leadership changes, whether a posting is real — as settled fact. The prompts should have the AI build the research structure: the questions, the named source types, the scoring framework, the red-flag taxonomy. The reader gathers the evidence; the AI organizes it and helps score it. A prompt that has the AI "investigate" a named company and report findings should expect to be marked down on Content Accuracy — the honest version has the reader paste in what they found.
Series dependency chain, for the Metadata block: Week 4 consumes Week 2's target-role spec sheet (companies are scored against its must-haves and deal-breakers). Week 4 produces the ranked target-company dossier — consumed by Week 5 (applications go to the top of the ranking), Week 6 (interview prep draws on the dossier), and Week 7 (the negotiation leans on what the reader knows about the company's position).
Because readers may arrive at this post without having read the earlier weeks, the prompts should work for someone with a shortlist scribbled on a napkin, while making clear the matrix bites harder when it scores against a real Week 2 spec.
Three difficulty tiers as always — Beginner, Intermediate, Advanced — each a genuinely different approach to the same problem, not the same prompt at three lengths.
On examples: this is a career topic that touches every industry. The template lists tech startup / retail / freelance as suggested industry examples — those are marked MAY, and adapting them is expected here. A reader vetting a fast-growing startup's runway, another weighing a big company's layoff pattern against its brand, and a third trying to tell a real posting from a ghost are the right kinds of contexts. Choosing them over the suggested business examples is correct behaviour and will not be scored against you.
A note on supplied figures. Anything marked `[SUPPLIED — use as given]` above came from Ketelsen.ai's own research brief. Use it freely — you are not fabricating by repeating it, and you will not be marked down for leaving it uncited. Do not attach an invented source to it. (No supplied figures this week — and given the constraint above, this is the worst possible week to invent one. Ghost-listing prevalence numbers circulate widely and vary wildly; write around them.)
## BEFORE YOU SUBMIT — STRUCTURAL CHECK
(This block is identical every week. It exists because these specific items are the ones posts drop, and a dropped structural item costs compliance points for something that takes one minute to add.)
Your post is parsed by a script before any human reads it. Confirm all seven:
1. ☐ Response begins with `PLATFORM: <your name>` and `WEEK: 4` 2. ☐ `## Lead` present once, at the very top, before Variation 1 3. ☐ `## In one line` present in all three variations 4. ☐ `## What this prompt gives you` present in all three variations 5. ☐ `## The Prompt` present in all three variations, with the prompt in double quotes beneath it 6. ☐ `## Introductory Hook` and `## Current Use` present in all three variations (three of each — not one) 7. ☐ Every template heading written as `##`, none bolded instead; prompt breakdown is running text split on ` : `, with no `###` headings inside it
A complete post has 57 `##` headings. If your count is well short, a section is missing or was bolded instead of hashed.
One extra check this week: confirm no prompt asks the AI to report a named company's current facts — layoffs, funding, leadership, listing legitimacy — as findings. The AI builds the research structure and scores evidence the reader gathered; the reader does the investigating.
About this series. AI is reshaping work — for some readers, it's part of why you're searching at all. We don't pretend otherwise. And the hiring world is wary of AI-written material: many employers restrict how AI may be used in their own hiring decisions, and recruiters increasingly recognize — and discard — machine-written applications. So this series teaches a different approach: use AI as your private research analyst, interview coach, and thinking partner, while every word an employer sees or hears from you stays genuinely, verifiably yours. AI behind the scenes. You on the page.
Most employers enter an interview knowing far more about you than you know about them. This week gives you three ways to narrow that information gap: a Beginner prompt for building a clear company snapshot, an Intermediate prompt for auditing warning signs, and an Advanced prompt for maintaining a ranked target-company dossier. The goal is not to collect trivia—it is to decide which employers deserve your application effort before you spend hours tailoring résumés, preparing for interviews, or accepting invitations from companies you have not properly examined.
The One-Company Snapshot Builder
Turn scattered employer research into one clear, evidence-labeled brief.
A careers page is designed to persuade you. It highlights the mission, benefits, smiling employees, and polished values language while leaving harder questions outside the frame. Before investing time in an application, you need a more balanced view: what the company sells, how it appears to make money, whether the role is visible on its official site, what employees repeatedly mention, and which important questions remain unanswered. This Beginner prompt gives you a repeatable research checklist and turns the evidence you gather into a concise company snapshot without asking AI to pretend it knows current facts.
Why this matters now
This matters now because a professional-looking listing is not proof that a position is active, accurately described, or even legitimate. The Federal Trade Commission recommends verifying openings through an employer’s official website and independently researching the organization rather than trusting a listing or recruiter message at face value. The prompt below makes that verification step routine while keeping you—not the AI—in control of the evidence.
Act as my company-research organizer. You are my private analyst, not a source of current company facts.
I am considering applying to this company:
Company: Role: Location: Link to the job posting, if available: My main job priorities: My deal-breakers:
First, create a research checklist for this company using these sections:
1. Company identity: what it does, its main products or services, who its customers are, and how it appears to make money.
2. Role legitimacy: whether the role appears on the company’s official careers page, when the listing was posted or updated, whether the description is consistent across sources, and whether a named recruiter uses a verifiable company address.
3. Company condition: recent business announcements, major changes, hiring activity, layoffs, or other signals I should verify.
4. Employee experience: recurring themes in recent employee reviews, including both strengths and complaints.
5. Role fit: evidence related to my priorities and deal-breakers.
6. Open questions: facts I could not verify and questions I should ask a recruiter or interviewer.
For every checklist item, name one or two appropriate places where I could look. Distinguish primary sources, such as the company’s official site or regulatory filings, from secondary sources, such as news coverage or employee-review platforms.
Do not research the company for me or present current claims as facts. Wait for me to paste my findings.
After I paste my findings, organize them into a Company Snapshot Brief with these labels:
Verified evidence Employee-reported patterns Unverified claims Unknowns Questions to ask Overall fit: Strong, Possible, Weak, or Insufficient Evidence
Explain the overall-fit label in no more than five sentences. Do not fill gaps with assumptions. If evidence conflicts, show the conflict instead of resolving it for me.
Anything I later communicate to an employer must be rewritten in my own words and must remain truthful.
How the AI reads this prompt
Practical examples from different industries
Illustrative example — Healthcare operations:
A hospital operations manager is considering a patient-flow role at a regional healthcare system. She enters the employer, role, location, her preference for stable leadership, and her deal-breaker of mandatory overnight availability. The prompt gives her a checklist covering the organization’s services, official listing, recent leadership announcements, employee-review patterns, and scheduling questions. After she pastes her findings, the AI separates confirmed shift language from employee reports about understaffing and identifies overnight expectations as an unresolved interview question.
Illustrative example — Manufacturing:
A maintenance technician finds the same plant position on three job boards, each with slightly different pay and experience requirements. He uses the checklist to compare those listings with the manufacturer’s official careers page, confirm the facility’s identity, look for state layoff notices, and record dated employee comments about overtime. The resulting snapshot does not declare the position fake or the plant unstable. Instead, it shows one verified official listing, two inconsistent copies, repeated overtime complaints, and three questions to resolve before applying.
Illustrative example — Nonprofit work:
A program coordinator is interested in a small community nonprofit that has limited public information. She gathers its official website, public tax filings, board roster, local news coverage, and a few employee comments. Because the evidence is thin, the prompt produces an “Insufficient Evidence” rating rather than inventing a culture assessment. It also gives her a short list of questions about funding concentration, staff turnover, supervision, and why the role is open.
Creative use case ideas
- Compare two internship hosts before a student commits an entire semester. - Help a volunteer evaluate the governance of a community organization. - Research a prospective client before accepting a long freelance engagement. - Examine a graduate assistantship, laboratory, or research center as though it were an employer. - Review a seasonal job whose posting reappears every few months.
Adaptability tips
Replace the six default research sections when the role demands a different lens. A government applicant might add public budgets, agency audits, civil-service rules, and political oversight. A startup candidate might add funding dates, customer concentration, founder history, and evidence of runway while treating any runway estimate as uncertain unless the company supplied the necessary figures.
You can also change the verdict labels. A cautious reader might use “Apply,” “Research More,” and “Do Not Prioritize,” while someone building a broad initial list might use “Tier 1,” “Tier 2,” “Tier 3,” and “Unrated.” Keep the “Insufficient Evidence” option regardless of the vocabulary.
For a quick pass, research only identity, listing legitimacy, and deal-breakers. For a high-stakes leadership role, expand the checklist to include financial disclosures, executive turnover, strategic priorities, litigation, regulatory exposure, and questions raised during the interview process.
Pro tips
- Add a date beside every finding. A fact without a date may be accurate but no longer useful. - Ask the AI to identify the single strongest source supporting each major conclusion. - Keep direct quotations short and preserve the original link or document name in your notes. - Run the snapshot again after an interview because the conversation may resolve—or create—important unknowns.
Prerequisites
Have the company name, target role, location, and job-posting link available. Write down three to five priorities and any non-negotiable deal-breakers before starting. You also need enough time to open the suggested sources and paste your findings back into the AI tool; the prompt is an organizer, not a substitute for that research.
Required tools
Any general-purpose AI assistant that can accept a long pasted prompt and your research notes. A web browser is required for visiting official careers pages, company websites, regulatory databases, reputable news sources, and employee-review platforms. No paid AI tier is required, although longer context limits make it easier to analyze extensive notes.
Frequently asked questions
Can I simply ask the AI to research the company for me?
You can use a browsing-enabled tool to help locate sources, but you should not treat its summary as verified simply because it includes links. Open the important sources yourself, confirm that they refer to the correct company and date, and paste the relevant evidence into the analysis. The SEC’s EDGAR database, for example, provides direct access to public-company filings, while CareerOneStop recommends researching employers through multiple sources. The value of this prompt is that it preserves that evidence trail instead of replacing it with a confident summary.
What if the company is private and publishes very little?
Sparse information should reduce confidence, not automatically count against the employer. Use official business registrations, the company’s own materials, credible local reporting, customer or partner announcements, and questions asked directly during the hiring process. Mark anything you cannot verify as unknown. A small private employer may be excellent, but the evidence may not support a strong conclusion before an interview.
How many employee reviews should I read?
There is no universal number that guarantees a representative picture. Read across dates, roles, locations, ratings, and employment statuses until you can distinguish recurring themes from isolated complaints or praise. Give more weight to specific descriptions that appear independently over time than to a sudden cluster of vague, similarly worded reviews. Record contradictions instead of choosing the side that matches your initial impression.
Should one serious warning sign stop me from applying?
It depends on the evidence, your risk tolerance, and the nature of the warning. A verified mismatch with a non-negotiable deal-breaker may justify stopping immediately, while an anonymous complaint may only justify another question. Separate the severity of a claim from the confidence you have in it. That prevents a dramatic but weakly supported allegation from outweighing several stronger sources.
Recommended follow-up prompts
- Use the Week 2 Target-Role Spec Sheet prompt to define the must-haves and deal-breakers this snapshot should test. - Run the Intermediate Red-Flag Evidence Audit when a company remains attractive but the research contains warning signs. - Use an interview-question builder to convert the snapshot’s unknowns into neutral questions you rewrite in your own voice.
Tags and categories
Tags:
company research, employer research, job search, job-posting verification, career planning, evidence gathering, AI prompts
Categories:
Job Search, Research and Analysis
Citations
- CareerOneStop, “Research Employers,” which describes employer research as a critical job-search step and recommends examining employer information from multiple sources. - Federal Trade Commission, “Scammers Are Hijacking Job Ads,” which recommends verifying openings through an organization’s official website and independently researching complaints. - Federal Trade Commission, “Job Scammers Are Looking to Hire You,” covering recruiter verification and common warning signs in unsolicited job approaches. - Investor.gov, “Using EDGAR to Research Investments,” explaining the public availability and research value of company filings.
The Red-Flag Evidence Auditor
Test worrying signals without turning suspicion into unsupported conclusions.
Employer research becomes difficult when the evidence stops agreeing. The official site projects momentum, recent reviewers describe reorganization, a leader quietly departed, and the same job has appeared repeatedly for months. None of those observations proves the company is a bad employer, but ignoring the pattern would be equally careless. This Intermediate prompt teaches the AI to work like an auditor: define the allegation, identify the evidence needed, classify what the reader finds, test alternative explanations, and recommend the next verification step. It replaces vague unease with a record you can inspect and challenge.
Why this matters now
Online evidence is easy to overread. Review ecosystems can contain false, incentivized, suppressed, or unusually clustered feedback, while legitimate layoffs and leadership changes may be documented in places job seekers rarely check. The Department of Labor explains that qualifying plant closings and mass layoffs can trigger WARN notices, and the FTC’s review rule identifies sudden bursts and other anomalies as potential reasons for closer examination. Neither source proves a specific employer is unsafe; both show why claims should be traced to the strongest available record.
Act as an evidence auditor for my employer research. Your job is to help me test warning signs, not to prove that my suspicions are correct.
Company: Role: Location: My must-haves: My deal-breakers:
I will paste research notes, links, dates, excerpts, and observations about one employer. Use only what I provide. Do not claim that you independently verified current layoffs, funding, leadership changes, reviews, or job-listing legitimacy.
Audit the evidence in these categories:
1. Layoff and restructuring history
2. Leadership turnover
3. Financial or funding pressure
4. Employee-review patterns
5. Job-listing legitimacy
6. Pay and role transparency
7. Interview-process reputation
8. Any additional risk raised by my evidence
For each possible warning sign, produce:
Claim being tested Evidence supporting it Evidence weakening or contradicting it Source type Evidence date Evidence quality: Primary, Strong Secondary, Weak Secondary, or Anecdotal Status: Confirmed, Concerning, Rumor, Cleared, or Unknown Severity if true: Low, Medium, High, or Critical What I should verify next A neutral question I could ask during the hiring process
Use these rules:
A confirmed flag requires direct evidence or multiple strong independent sources. A concerning flag has credible support but remains incomplete. A rumor is an unverified claim, isolated allegation, or repeated claim traceable to the same original source. Cleared means the supplied evidence directly resolves the concern; it does not mean the company is risk-free. Unknown means the evidence cannot support a responsible classification.
For employee reviews, weight recency, specificity, role and location relevance, independent repetition, and patterns across time. Do not calculate a conclusion from the average star rating alone. Flag suspicious timing or repeated wording as a reason to investigate, not proof of manipulation.
For layoffs, leadership changes, funding, and job legitimacy, tell me which primary record would be most useful, but do not say that the record contains a finding unless I pasted it.
After auditing each flag, give me:
A red-flag summary The three most decision-relevant findings The three largest evidence gaps A provisional recommendation: Continue, Continue with Questions, Pause for Research, or Stop A confidence level from 0 to 100, based on evidence completeness—not on how strongly worded the claims are
Explain the recommendation in six sentences or fewer. Do not average away a Critical confirmed flag. Show which of my must-haves or deal-breakers each finding affects.
You are my private analyst. Any question or statement I use with an employer must be rewritten in my own words, remain neutral, and be true.
How the AI reads this prompt
Practical examples from different industries
Illustrative example — Venture-backed software company:
A product manager is considering a startup that announced rapid growth two years earlier but now posts few updates. Her notes include an old funding announcement, two recent executive departures, repeated employee comments about changing priorities, and no verified financial figures. The audit labels the departures as confirmed only if supported by direct company or filing evidence, treats runway as unknown, and classifies the review pattern as concerning rather than proven. The recommended next step is to ask how the role is funded, why it is open, and what milestones define the next year.
Illustrative example — National retailer:
A district manager is attracted to a well-known brand but finds several years of layoff announcements and recurring complaints about reorganizations. The AI maps each item by date, source quality, business unit, geography, and relevance to the target position. A national restructuring does not automatically prove the specific division is unstable, but repeated verified actions may raise the severity of the concern. The final recommendation is “Continue with Questions,” supported by a focused list of issues to verify with the recruiter.
Illustrative example — Remote administrative role:
A job seeker finds a remote position advertised by someone claiming to represent a legitimate company. The email comes from a personal account, the role is absent from the official careers page, the interview is conducted entirely through text, and the recruiter requests banking information. The audit treats the official-site mismatch and information request as high-severity evidence, recommends stopping contact, and directs the reader to independently contact the company through a known legitimate channel. It does not rely on the listing’s logos or professional formatting.
Creative use case ideas
- Audit the stability of a university department before accepting a graduate appointment. - Evaluate a potential board seat at a nonprofit with recent leadership turnover. - Examine whether a repeated contract opportunity reflects real demand or perpetual sourcing. - Compare franchise opportunities by separating disclosed facts from testimonials. - Review a creative residency or arts organization after participants report recurring administrative problems.
Adaptability tips
Add or remove audit categories according to the employer type. For a startup, expand funding pressure into last verified financing, revenue evidence, investor participation, hiring direction, and unanswered runway questions. For a public employer, add budget approvals, audit findings, governance, and workforce plans. For a regulated company, include enforcement actions, licensing, safety records, and compliance leadership.
Adjust the evidence thresholds to match the cost of being wrong. A short-term internship may justify a lighter audit, while relocation, a pay cut, or leaving a secure job should require stronger evidence and more direct questions. You can also make certain deal-breakers automatic stop conditions rather than allowing them to be balanced by positive factors.
The same framework can audit positive claims. Replace “warning sign” with “employer claim” and test statements such as rapid promotion, flexible work, strong training, or stable leadership. Positive marketing deserves the same evidence discipline as negative allegations.
Pro tips
- Ask the AI to draw a simple source-dependency map showing which claims originate from the same report. - Add a “scope match” field: company-wide, division, location, team, role, or unknown. - Run the audit twice—once as a skeptical candidate and once as a fair-minded company representative—then compare unresolved disagreements. - Preserve the original wording of critical evidence so later summaries cannot subtly strengthen the claim.
Prerequisites
Complete at least a basic company snapshot first. Gather dated notes, document names, excerpts, and links rather than relying on memory. Have your Week 2 must-haves and deal-breakers ready, or write a temporary list before running the audit. The prompt becomes more reliable as the evidence set becomes more explicit.
Required tools
A general-purpose AI assistant with enough context capacity to hold the prompt and your research notes. A spreadsheet or note-taking tool is helpful for preserving links, dates, and excerpts. Access to official company materials, SEC filings where applicable, state WARN-notice resources, reputable news archives, and employee-review platforms may be needed.
Frequently asked questions
Does a WARN notice prove that a company is broadly unstable?
No. WARN applies only in particular circumstances and a notice documents a covered closing or mass layoff, not the future condition of every team or facility. The Department of Labor explains that WARN generally applies to covered employers and requires advance notice for qualifying events, with notices also going to state dislocated-worker units. Use the notice as strong evidence of the specific event it describes, then examine its date, location, scale, and relevance to your role.
How can I identify suspicious employer-review patterns?
Look for concentrated timing, highly similar wording, vague praise disconnected from work experience, rating changes that do not match the surrounding narrative, and reviews referring to the wrong role or organization. The FTC notes that unusually large numbers of reviews appearing rapidly can be a reason to investigate whether reviews are false, but the pattern alone is not conclusive proof. Compare the cluster with older and newer reviews and check whether specific themes recur independently. Keep the classification at “concerning” until stronger evidence supports a firmer conclusion.
Is leadership turnover always a warning sign?
No. Planned succession, retirement, acquisition, promotion, or strategic change can all produce legitimate transitions. For a public company, Form 8-K disclosures may document certain changes in high-level leadership, while company announcements and credible reporting can provide additional context. The risk comes from patterns such as repeated short tenures, several related departures, unexplained interim appointments, or turnover that affects the target team. Treat the event as confirmed only when supported, then separately judge its possible significance.
What should I do when employee reviews and official sources sharply disagree?
Keep both records visible because they answer different questions. Official sources are often stronger for dated corporate events, while employee reports may reveal experiences that formal disclosures do not address. Test whether the reviews are recent, specific, independent, and relevant to the same location or function. The disagreement should become an interview question, not an excuse to discard whichever source is inconvenient.
Recommended follow-up prompts
- Run the Advanced Living Dossier Matrix to compare the audited employer with other targets. - Use a source-verification prompt to trace repeated claims back to their earliest available origin. - Convert unresolved flags into a recruiter-and-interviewer question bank, then rewrite every question in your own voice.
Tags and categories
Tags:
employer red flags, company audit, layoffs, leadership turnover, employee reviews, ghost jobs, evidence quality, job search
Categories:
Job Search, Risk Assessment
Citations
- U.S. Department of Labor, “Notices for Plant Closings and Mass Layoffs,” explaining WARN coverage and notification requirements. - Federal Trade Commission, “The Consumer Reviews and Testimonials Rule: Questions and Answers,” including review anomalies that can warrant further investigation. - Federal Trade Commission, announcement of the final rule addressing fake, purchased, insider, suppressed, and company-controlled reviews. - Investor.gov, “Public Companies,” describing Form 8-K disclosures for major events, including changes in corporate leadership. - Federal Trade Commission, guidance on verifying recruiters and avoiding fake job offers.
The Living Target-Company Dossier
Rank employers through a reusable, evidence-scored research operating system.
A serious job search does not need a longer company list. It needs a defensible order of operations. Without one, recognizable brands receive attention first, newly discovered employers disappear into browser tabs, red flags are remembered inconsistently, and hours are spent tailoring applications for roles that violate the reader’s own requirements. This Advanced prompt turns company intelligence into a living system. It defines a common evidence schema, applies the Week 2 role specification, separates fit from risk, scores confidence independently, and produces both a ranked target list and a screened-out list with reasons.
Why this matters now
The strongest employer research draws from sources built for different purposes. Public-company filings can reveal the business, material risks, operating results, and significant events; Form D can provide limited information about some exempt private offerings; BLS data can supply industry and regional context; and state WARN systems may document qualifying layoffs. None of these sources alone answers whether you should join a particular team. The value comes from organizing them into a comparable dossier while preserving dates, provenance, uncertainty, and the limits of each source.
Act as the architect and analyst for my target-company dossier. Build a reusable decision system from evidence I collect. Do not act as a ghostwriter, and do not claim current knowledge about any company unless I paste the supporting evidence.
My target-role specification:
Target role: Acceptable adjacent roles: Preferred industries: Preferred locations: Remote, hybrid, or onsite requirements: Compensation requirements: Company-size preferences: Career-growth priorities: Work-content priorities: Culture or management priorities: Stability requirements: Must-haves: Deal-breakers: Other constraints:
My candidate companies:
Paste company names here.
Stage 1 — Build the research architecture
Create a standard dossier template that I can complete for every company. Include:
Company identity and business model Target role and role family Official job-listing evidence Role-posting date and last observed date Why the role appears to exist Products, services, customers, and revenue logic Financial-health evidence Funding evidence for private companies Layoff and restructuring evidence Leadership-change evidence Industry and regional trajectory Hiring direction Employee-review patterns Pay and transparency evidence Interview-process reputation Referral and hidden-market signals Match to each must-have Conflict with each deal-breaker Open questions Source register Last researched date
For every field, provide:
The question to answer Preferred primary source types Useful secondary source types What not to infer A short example of acceptable evidence
Stage 2 — Define the evidence model
Classify every item I later paste using:
Evidence status: Verified, Corroborated, Anecdotal, Disputed, or Unknown Source quality: Primary, Strong Secondary, Weak Secondary, or Unclear Recency: Current, Recent, Aging, or Stale Scope: Company, division, location, team, role, or unknown Decision relevance: Low, Medium, High, or Decisive
Do not treat repeated articles as independent corroboration when they rely on the same original source. Do not convert a missing fact into a negative fact. Do not treat a company statement as independent proof of its own marketing claim. Do not treat an employee-review average as a culture verdict.
Stage 3 — Define the scoring model
Create a 100-point composite score using these default weights:
Role and work fit: 25 Must-have alignment: 20 Stability and financial signals: 15 Culture and management patterns: 10 Compensation and transparency: 10 Growth and career value: 10 Listing legitimacy and hiring intent: 5 Interview-process reputation: 5
Let me adjust the weights before scoring.
Calculate a separate Evidence Confidence score from 0 to 100 based on source quality, recency, coverage, corroboration, and scope match. Do not allow strong fit to increase evidence confidence.
Apply these gates before ranking:
Any verified deal-breaker places the company on the screened-out list unless I explicitly override it. Any confirmed Critical red flag triggers manual review and prevents automatic Tier 1 placement. A company with insufficient evidence cannot rank above a well-supported company solely because its visible evidence is positive. Unknown data receives no positive or negative points unless I define a missing-data penalty. Scores within three points are treated as effectively tied.
Stage 4 — Process evidence
Wait for me to paste one company’s evidence at a time.
For each company:
Normalize the evidence into the dossier. Identify duplicated or dependent sources. Show contradictions. Mark unsupported claims. Ask only the highest-value missing questions. Calculate category scores. Calculate Evidence Confidence. Run the red-flag gates. Create a provisional classification:
Tier 1 — Prioritize Tier 2 — Worth pursuing Tier 3 — Opportunistic Research Hold Screened Out
Explain the classification using the five most decision-relevant facts and the three most important unknowns.
Stage 5 — Build the portfolio view
After I have entered at least three companies, produce:
A ranked target-company list A screened-out list with explicit reasons A Research Hold list with the next cheapest useful check The strongest company for each major priority The company most sensitive to an uncertain assumption The company whose ranking changed most after red-flag screening A recommended allocation of my next ten hours of application effort A research-refresh schedule based on evidence age and company volatility
Represent the comparison as a consistent matrix plus a concise narrative. Keep fit score, risk status, and Evidence Confidence separate.
Stage 6 — Maintain the living system
When I provide new evidence, update only the affected fields and show:
What changed Which source caused the change Which score changed Whether the ranking changed Whether a gate was triggered or cleared What older evidence should now be archived
Never silently overwrite prior conclusions. Preserve a dated change log.
Use only evidence I provide. You may recommend source types and search questions, but you may not report a named company’s current layoffs, funding, leadership changes, listing legitimacy, or other live conditions as settled facts.
You are my private analyst, coach, and sparring partner. Any résumé language, outreach, interview answer, or negotiation message must be written by me in my own truthful words.
How the AI reads this prompt
Practical examples from different industries
Illustrative example — Cybersecurity leadership search:
A candidate is comparing six security-management employers across healthcare, finance, manufacturing, and consulting. He gives the dossier his requirements for incident-response authority, executive access, staffing, on-call expectations, compensation, and organizational stability. Each company is researched with the same schema, but the source path varies: public filings for listed companies, regulatory and customer evidence for private firms, and direct interview questions for team-level conditions. The final portfolio favors two lesser-known employers over a famous brand because the fit is stronger and the evidence is more complete.
Illustrative example — Biotechnology scientist:
A scientist is weighing one public pharmaceutical company, two venture-backed biotechnology firms, and a university spinout. The dossier keeps public financial disclosures, private funding records, trial announcements, leadership changes, and employee reports in separate evidence classes. One startup scores highly on scientific fit but remains on Research Hold because the available funding evidence cannot support a responsible runway estimate. The system recommends a targeted recruiter question and identifies the exact assumption that would move the company into Tier 1 or Tier 3.
Illustrative example — Skilled-trades relocation:
An industrial electrician is considering relocation for positions at several plants. His matrix includes wage transparency, schedule, union status, overtime, facility investment, layoff history, local industry conditions, housing implications, and verified deal-breakers. A position with the highest advertised pay ranks second after the system accounts for rotating shifts and uncertain project duration. Another employer becomes the priority because its slightly lower pay is paired with better role fit, stronger evidence, and fewer unresolved relocation risks.
Creative use case ideas
- Build a college-selection dossier that compares departments, funding, advising, and student outcomes. - Rank volunteer organizations before committing to a long-term leadership role. - Compare publishers, galleries, or agencies for a creative partnership. - Evaluate several cities as employment ecosystems rather than judging only individual openings. - Maintain a consulting-client dossier that tracks payment reliability, strategic fit, and engagement risk.
Adaptability tips
Change both the schema and weights when the decision changes. An early-career applicant may assign more weight to training and mentorship, while an executive may prioritize governance, strategic authority, financial condition, and leadership stability. A contractor could replace career growth with payment terms, project clarity, and client concentration.
The matrix can operate at several depths. Use a lightweight version for twenty initial employers, a standard version for the top eight, and a full forensic version for the three companies most likely to receive substantial application or interview effort. Keep the field names consistent so evidence can move between levels without being reformatted.
For international employers, add country-specific corporate registries, labor rules, financial-reporting systems, and cultural context. Do not assume a U.S. filing or layoff-notice framework applies elsewhere. The architecture stays the same, but the source map must change.
For teams within one large employer, replace the company row with a business-unit or team row. Preserve company-wide evidence separately and mark whether each finding actually reaches the target group. This prevents a global brand reputation from erasing meaningful differences between managers, locations, or functions.
Pro tips
- Freeze the scoring weights for one application cycle so they do not drift toward whichever company currently feels most exciting. - Run sensitivity tests by changing each major weight and unresolved assumption to see whether the top tier remains stable. - Archive source snapshots or detailed notes because web pages, job listings, and leadership biographies can change. - Add a “decision journal” entry whenever you override a gate or choose a lower-ranked company, including the reason.
Prerequisites
Complete or reconstruct the Week 2 target-role specification, including must-haves and deal-breakers. Assemble an initial employer shortlist, even if it is informal. Choose a place to preserve the dossier, source register, and dated change log. Expect to gather the evidence yourself and paste it into the AI one employer at a time.
Required tools
A general-purpose AI assistant with a large enough context window for the prompt, role specification, company records, and comparison output. A spreadsheet, database, or structured note system is strongly recommended for long-term maintenance. Research may require a web browser, SEC EDGAR, state WARN resources, official corporate sites, BLS or CareerOneStop data, reputable news archives, employee-review platforms, and saved copies of important evidence.
Frequently asked questions
Can the composite score tell me which offer I should accept?
No. The score organizes your stated priorities and current evidence; it does not replace judgment, interviews, reference checks, or the final terms of an offer. Treat the ranking as a way to allocate research and application effort rather than an objective measurement of employer quality. Revisit the weights when an actual offer changes the decision. Preserve the underlying category scores so you can see why two similar totals may represent very different trade-offs.
How should I research a public company’s financial condition?
Start with the company’s 10-K, 10-Q, and relevant 8-K filings rather than relying only on investor presentations or headlines. Investor.gov explains that these reports can provide information about the business, risks, operating results, financial statements, and significant events. You still need to interpret the evidence carefully because an employer’s financial condition and the prospects of one role or team are not identical. Record the filing date, reporting period, relevant excerpt, and the conclusion you believe it supports.
Can Form D tell me a startup’s funding runway?
No. Investor.gov describes Form D as a brief notice containing basic information about certain exempt offerings, including the offering size and date of first sale. It does not necessarily tell you how much cash the company currently holds, how quickly it is spending, whether the full offering was sold, or how long the money will last. Use it as one funding-related record, not as a complete financial statement. A responsible runway estimate requires additional verified inputs and clearly stated assumptions.
Why include industry and regional data if I am evaluating a company?
Industry and regional context can help you distinguish company-specific signals from broader conditions. BLS’s Quarterly Census of Employment and Wages publishes establishment, employment, and wage information by industry and geography, making it useful for understanding the surrounding labor market. That context does not prove that a particular employer is expanding or contracting. It gives you a baseline against which company evidence can be interpreted.
How often should I refresh the dossier?
Refresh volatile evidence more often than structural evidence. Active listings, leadership changes, funding announcements, layoffs, and interview reports may deserve review before each major application or interview step, while the basic business model may change less often. Let the system flag aging evidence rather than rebuilding every dossier on a fixed weekly schedule. Always refresh a top-ranked employer before accepting an offer or making a consequential relocation decision.
Recommended follow-up prompts
- Use the Week 5 application-prioritization workflow to allocate effort across the Tier 1 and Tier 2 employers. - Use a Week 6 interview-preparation prompt that draws questions and hypotheses directly from each company dossier. - Build a negotiation evidence pack for Week 7 using verified company position, role scope, market data, and offer details.
Tags and categories
Tags:
target-company dossier, company ranking, employer intelligence, decision matrix, evidence scoring, job-search strategy, red-flag audit, application prioritization
Categories:
Job Search, Decision Systems
Citations
- Investor.gov, “How to Read a 10-K/10-Q,” describing the business, risk, operating, and financial information available in public-company reports. - Investor.gov, “Using EDGAR to Research Investments,” explaining free public access to corporate filings and the information contained in common filing types. - Investor.gov, “Form D,” explaining the limited information supplied in notices of exempt securities offerings. - U.S. Bureau of Labor Statistics, “QCEW Overview” and the 2026 researcher’s guide, covering industry, employment, wage, and geographic data. - CareerOneStop, “Target Your Job Search,” recommending an employer list and research into meaningful fit before focusing on individual openings. - U.S. Department of Labor, WARN guidance on qualifying plant closings and mass layoffs.
Which of the three should you use?
The Beginner prompt answers the first practical question: “What should I learn about this employer?” It is the fastest route from a company name and job link to a balanced snapshot. Its strength is accessibility. It gives a new AI user a bounded checklist, named source types, clear evidence labels, and permission to conclude that the available information is insufficient.
The Intermediate prompt begins where the snapshot becomes uncomfortable. It is designed for conflicting reviews, layoff records, leadership changes, questionable listings, funding uncertainty, and other signals that require more than a simple summary. Its central lesson is that confidence, severity, and decision relevance are separate. A claim may be serious but weakly supported, or thoroughly verified but only marginally relevant to the target role.
The Advanced prompt treats employer research as a portfolio-allocation problem. It standardizes the evidence collected for every company, connects each employer to the Week 2 role specification, applies explicit weights and non-negotiable gates, and preserves changes over time. Choose it when several employers are competing for limited application hours or when the consequences of a poor move—relocation, lost income, reduced stability, or leaving a secure role—justify a maintained decision system. The three prompts overlap in their evidence discipline, but they serve different moments: understand one employer, investigate a concern, and allocate effort across the full search.
TAGS:
