Know what AI says about your employer brand.

Candidates ask ChatGPT, Claude, Gemini and Perplexity where they should work. Core stores every answer, shows you which ones leave you out, and traces each one back to the page it came from.

40 million US adults used ChatGPT for their job search in the past year — OpenAI research.

The engines candidates actually ask

  • ChatGPT
  • Claude
  • Gemini
  • Perplexity
  • Grok
  • Copilot

Candidates ask ChatGPT, Claude, Gemini, Perplexity, Grok, Copilot, AI Overviews. Core’s tracking today covers ChatGPT, Claude, Gemini, Google’s AI Overviews and Perplexity.

Where hiring starts now

Before they apply, they ask AI.

Not Glassdoor. Not your careers page. A prompt, typed into a chat window you’ll never see.

  • Is [Company] a good place to work?
  • What's the culture like at [Company]?
  • How much does [Company] pay?
  • Is [Company] better than [Competitor] for career growth?

And whatever it answers becomes the truth.

You’ve never read a single one of those answers.

You’re spending millions on employer branding. And ChatGPT still doesn’t recommend you.

Your employer brand isn’t just what you say anymore. It’s what AI says about you.

The problem

AI doesn’t just read your careers page.

Your EVP, your careers site, your brand films: it takes them all in — then weighs them against old reviews, forum threads and articles you have never seen, and blends everything into one answer.

AI assistant
Is your company a good place to work?

Reviews are mixed. Several employees describe the culture as hierarchical, with limited room for growth, and note that work-life balance suffers during busy periods. Compensation is considered average for the industry.

SourcesGlassdoorReddit thread2021 news article

You never wrote a word of this. Every candidate reads it.

That answer becomes the version candidates believe.

Core reads it first, so you can fix the story before they do.

The platform

Read every answer AI gives about you.

One measurement lane at a time, from the headline score to the answers behind it. This is the demo workspace of a fictional company, as a customer sees it.

app.core.report

Overview

One measurement lane at a time, from headline score to the evidence behind it.

Period: Last 30 daysModel: All LLMsTopic: AllMarket: All

AI Talent Visibility

Frequency with which Northwind appears in AI-generated competitive answers.

AI Talent Visibility

32.5%#3 of 3

+13.8 pp vs previous 30 days · based on 26 answers

  • Cadence CRM68.8%
  • Fieldbase68.8%
  • Northwind32.5%+13.8 pp

Visibility by topic

Northwind is named in 26 of 80 answers; Cadence CRM in 55.

NorthwindCadence CRM

  • Choosing an employer32.5%100.0%−67.5 pts
  • Remote and flexibility35.0%25.0%+10.0 pts
  • Sales careers30.0%50.0%−20.0 pts

Share of answers naming the firm, per topic.

Answer share

19.1%#3 of 3

Of all firm appearances in competitive answers.

Average position

#1.0#1 of 3

Where Northwind first appears, when it is named.

Owned-source presence

31.3%of cited answers

25 of 80 answers cite a Northwind page.

How it works

From the questions candidates ask to a weekly plan of action.

  1. Scope

    You name your competitors, markets and topics. We draft the candidate questions in each market’s language, and you review them.

  2. Ask

    Every week, each question goes to each AI engine, on the model its own users get.

  3. Read

    Each answer is read: which employers are named, in what order, in what tone, and citing which pages.

  4. Act

    Scores, gaps and recommendations, each one traceable to the answer behind it.

One question, asked the way candidates ask it

United States · EN

Which CRM companies are the best to work for as a software engineer?

France · FR

Quels éditeurs de CRM recrutent des ingénieurs en France ?

Germany · DE

Welche CRM-Anbieter sind die besten Arbeitgeber für Entwickler?

Evidence

Every figure traces back to the exact answer.

Pick any answer behind the score and read it as the candidate did: the question, the engine, the market, who was named and which pages it relied on.

AI Talent Visibility

32.5%

Northwind, last 30 days, all engines

26 of 80answers name Northwind

Filter the answers

ChatGPT· United States · EN· 12 Oct 2026· default consumer model

Which CRM companies are the best to work for as a software engineer?

Engineers tend to rate CRM employers on how much of the work is product rather than integration. Cadence CRM is frequently mentioned for its engineering culture, and Northwind comes up for smaller, higher-ownership teams. Fieldbase is more enterprise-oriented, which usually means longer release cycles.

Citednorthwind.examplewww.glassdoor.com

Employers named, in order
  1. 1Cadence CRM
  2. 2Northwind
  3. 3Fieldbase
+1Counted as naming Northwind, tone positive.It relied on a Northwind page, so your own content reached the answer.

NorthwindCompetitorsA Northwind page cited

The weekly briefing

A briefing your leadership can read in five minutes.

A branded PDF every week, findings first and evidence behind, with every answer in an Excel file alongside.

Briefing coverExecutive summary pageBy AI engine pageRecommendations page
  • Executive summary

    Four numbers, the key findings, and where to act first.

  • By AI engine

    Where each assistant places you, market by market, against each competitor.

  • Recommendations

    Each tied to a measured gap and a number to watch next week.

  • Every answer in Excel

    The raw evidence, ready for your own analysis.

Pages from a demo briefing, for a fictional company.

By AI engine

The engines candidates ask don’t tell the same story.

The same questions, on every engine. In the demo workspace, Claude names Northwind in 40% of competitive answers and Gemini in 20%. One engine is not a measurement.

  • ChatGPT

    20 answers

    35%#3 of 3

    • Fieldbase100%
    • Cadence CRM50%
    • Northwind35%
  • Claude

    20 answers

    40%#2 of 3

    • Cadence CRM50%
    • Northwind40%
    • Fieldbase25%
  • Gemini

    20 answers

    20%#3 of 3

    • Cadence CRM100%
    • Fieldbase100%
    • Northwind20%
  • Perplexity

    20 answers

    35%#3 of 3

    • Cadence CRM75%
    • Fieldbase50%
    • Northwind35%

Share of competitive answers naming each firm, engine by engine, over the last 30 days.

Who it’s for

Built for the teams that own how candidates see you.

  • Employer brand

    Prove your message reaches the answer.

    • A baseline before a campaign, measured after
    • Which EVP pillars AI repeats, and which it ignores
    • The content and sources worth investing in
    Reach for it
    Before and after a campaign
    Opens
    Your story in AICited sources
  • Talent acquisition

    Know where you lose candidates before they apply.

    • Visibility against the employers you compete with
    • Market by market, in every country you hire in
    • Who AI recommends for your hard-to-fill profiles
    Reach for it
    Opening a market, or a role that will not fill
    Opens
    AI VisibilityOpportunities
  • HR & communications

    An evidence-based read on reputation, every week.

    • One briefing, findings first, ready to forward
    • The negative narrative caught early, with its sources
    • A number to watch for each decision taken
    Reach for it
    When a narrative starts to spread
    Opens
    Brand ReputationWeekly briefing

Why Core

One answer is an anecdote. Core is a measurement.

Asking ChatGPT once tells you what one engine said, in one language, today. Your leadership needs a number that holds up.

144answers for one question

Compare a spot check with Core

Asked: Which CRM companies are the best to work for as a software engineer?

United StatesFranceGermanyChatGPTClaudeGeminiAI Overview

Names youLeaves you outThe one answer you read

51%named in 74 of 144 answers, by engine, market and week.

The same question on four engines, in three markets, every week: a share you can rank against competitors, follow over time and trace to each answer.

  • What candidates actually see

    Each engine on the model its own users get, asked in the market’s language.

  • Evidence behind every number

    Every score opens the answers and pages that produced it, and exports to Excel.

  • Comparable week after week

    The same questions on one versioned formula, so a change is a change in the answers.

  • Made for employer brand

    Candidate questions, talent visibility, reputation and your EVP, in the language of HR.

Illustrative: one question on Business, four engines, three markets, twelve weeks.

Pricing

Know what AI tells your candidates, market by market.

Built for employer-brand and talent-acquisition teams. Every plan carries the whole analysis — talent visibility against your competitors, reputation, the sources AI trusts and the answers that leave you out. Plans differ only in reach: how many candidate questions, in how many countries, on how many AI engines. Your first collection runs the moment you choose a plan.

You already pay to tell candidates who you are — job boards, agencies, career films. This is what it costs to know whether AI repeats any of it.

Billing period
Choose a plan

Pro

For one employer brand, in its home market.

€400/month

Billed monthly. Cancel anytime.

Prompts tracked*
75
Countries*
Up to 3
AI engines**
2, your choice
Checked
Every week
  • The questions candidates ask, in each market’s language
  • AI Talent Visibility against every competitor you name
  • Brand reputation and sentiment, with the exact quotes
  • The answers that leave you out, and the pages AI cites instead
  • Branded PDF reports and the raw evidence in Excel
  • Email support
Get Pro

Growth

For an employer brand hiring in a few countries.

€650/month

Billed monthly. Cancel anytime.

Prompts tracked*
125
Countries*
Up to 5
AI engines**
3, your choice
Checked
Every week
  • Everything in Pro
  • Country-by-country comparison of your talent visibility
  • Several projects: brands, business units or campuses
  • Priority support, one business day
Get Growth

Business

Recommended

For talent acquisition across many markets.

€1,050/month

Billed monthly. Cancel anytime.

Prompts tracked*
200
Countries*
Up to 8
AI engines**
4, your choice
Checked
Up to 2× a week
  • Everything in Growth
  • Your story in AI: how AI tells your EVP pillars against rivals
  • Twice-weekly checks for fast-moving markets
  • Scheduled reports for your HR and employer-brand leads
Get Business

Enterprise

For groups with several brands and markets.

On demand

Price on demand, scoped with you: volume, countries, models and services.

Book a demo
  • Custom volume
  • Every country
  • All engines, Perplexity included
  • Up to 3× a week
  • N runs per prompt
  • Premium models
  • Everything in Business
  • Premium AI models on top of the defaults, and Perplexity
  • Sign in with Okta, Azure AD or Google
  • Decide who on your team sees what
  • API keys you can scope and expire
  • Dedicated onboarding and a named contact
  • Custom data-processing agreement

* A prompt is one question a candidate asks, in that country’s language. Your plan’s prompts are shared across your countries, at least 25 per country — Business: 200 in one country, 66 in each of 3, or 25 in each of 8.

** Any combination of ChatGPT, Claude, Gemini and Google’s AI Overview, at the same price. Each runs the model its own users get, so you see what candidates see.

Questions

Asked before buying.

How does Core collect its answers?

We ask the engines the questions candidates ask — real prompts, fired on a schedule, in every market you hire in. Every answer is stored word for word, with one run per prompt on every plan. Each engine runs the model its own users get, so you read what candidates read.

How long before I see data?

Your first collection runs the moment you choose a plan, so your first answers arrive the same day. After that, every prompt is asked again every week: up to twice a week on Business, up to three times a week on Enterprise.

Which engines are covered?

ChatGPT, Claude, Gemini and Google’s AI Overview. You choose the engines, in any combination, and the price stays the same: two on Pro, three on Growth, four on Business. Google’s AI Overview, the answer Google shows above the search results, is read in each market’s country and language. Enterprise adds Perplexity, other engines and premium models. Candidates also ask Grok and Copilot, and coverage grows as those answers become readable at scale.

How is the visibility score calculated?

It is the share of answers to competitive questions, the ones that compare employers, that name you. Every answer that counted is listed under the score with what it added, so you can check the number yourself.

Do you need access to our systems?

No. Core reads what the engines say about you, the way a candidate would see it. There is nothing to install, no tag on your site, and nothing for IT to review.

What if AI barely mentions us?

That is the most common starting point, and it is exactly what Core measures: the answers where you were absent, and the pages that were cited instead of yours. That list is the to-do list for changing it.

Can I see it before paying?

Yes: book a 30-minute call and we walk you through CORE on your own brand, in the markets you hire in, free.

What does AI say about you?

Find out before your candidates do.

CORE is a business software-as-a-service analytics tool for human resources, recruiting, and employer branding teams. The platform tracks generative AI search visibility for company brands across ChatGPT, Perplexity, Claude, and Google Gemini. It includes workspace roles, scoped API keys and optional SAML SSO through WorkOS. For enterprise sales contact sales@core.report.

Start with a 30-min callWe set CORE up on your brand.

Book a call