The Digital Footprint Audit

What this is for

Someone is going to look you up before they hire you, hire your firm, or take a meeting with you — and increasingly they're not using Google to do it. This is for finding out what that lookup actually surfaces, including the old accounts and outdated bios still contradicting how you position yourself now.

Why this has to run in AI tools, not Google

A recruiter, client, or due-diligence analyst doing a background check today is as likely to open Perplexity or ask ChatGPT with browsing as they are to type your name into Google. Those tools don't just list links — they read across sources and summarize a conclusion about you ("appears to be a marketing executive with a background in..."). That summary is what forms someone's first impression, and it can be wrong, outdated, or built from a source you forgot existed. You need to see that summary yourself before someone else does.

What these tools can and can't actually see

Be clear-eyed about this before you start:

• They can see what's public and indexed: news mentions, published articles, public LinkedIn/company bio pages, conference listings, public social posts, old company "about us" pages still live on the web.
• They generally cannot see: private social media accounts, content behind logins, most people-search/data-broker sites in full detail (though they may reference that such listings exist), and anything that isn't crawled or indexed at all.
• They can be wrong. AI search tools summarize and sometimes misattribute — two people with your name can get merged into one profile, or an old role can be reported as current. Treat every result as a lead to verify, not a fact.
• Results differ by tool and change over time. What Perplexity surfaces today and what it surfaces in three months won't be identical — this is a spot-check, not a permanent record.

The prompts

1. What a recruiter finds

Use Perplexity, or ChatGPT with browsing turned on (in ChatGPT, this happens automatically when a prompt requires current web info — no toggle needed in current versions, but confirm your model supports browsing before running this).

ROLE: You are a recruiter doing a routine background look-up on a
candidate before an interview.

CONTEXT: The candidate's name is [YOUR FULL NAME]. They currently work
in [INDUSTRY/FUNCTION, e.g. "B2B marketing, Northeast US"] and have
previously worked at [1-2 PAST EMPLOYERS, to help disambiguate from
other people with the same name].

CONSTRAINTS:
- Search for and summarize what you find about this person in the
  first 2 minutes of a normal background look-up.
- Note anything that looks outdated (old job titles, old employers
  listed as current) or contradictory across sources.
- If you find information about someone else with the same name, say so
  explicitly rather than blending it in.

OUTPUT FORMAT: A short paragraph, as if reporting back to a hiring
manager, followed by a bullet list of every source URL you used.

2. What a client finds

Same tool as above. Run this separately from Prompt 1 — a client is evaluating credibility, not employability, and the AI's framing changes with the role you give it.

ROLE: You are a prospective client doing due diligence on a consultant
or vendor before a first meeting.

CONTEXT: The person's name is [YOUR FULL NAME], and they present
themselves professionally as [YOUR CURRENT POSITIONING, e.g. "an
independent operations consultant for mid-size manufacturers"].

CONSTRAINTS:
- Search for and report what you find that would inform whether to trust
  this person with a paid engagement: credibility signals, past client
  or employer names, any public reviews or testimonials, anything that
  looks inconsistent with the stated positioning above.
- Explicitly flag anything you find that contradicts or undercuts the
  stated positioning.

OUTPUT FORMAT: A short credibility summary (as if writing a note to a
colleague deciding whether to proceed), plus source URLs.

3. What an adversary finds

Same tool. This is the least comfortable prompt and the most useful one — run it last.

ROLE: You are someone trying to find leverage or embarrassing
information about a specific person, for competitive or personal
reasons.

CONTEXT: The person's name is [YOUR FULL NAME], located in
[CITY/REGION], working in [INDUSTRY].

CONSTRAINTS:
- Search for anything publicly available that could be used against
  this person professionally or personally: old controversial posts,
  litigation records, negative press, financial or property records
  that are publicly indexed, or contradictions between public claims
  and public records.
- Do not fabricate anything. If nothing turns up, say so plainly instead
  of stretching a weak result.

OUTPUT FORMAT: A blunt list of findings, each with the source URL, or a
single line stating nothing notable was found.

4. Old accounts and bios still live and contradicting you

Use Perplexity for this one — it tends to be better at listing distinct sources rather than folding them into one narrative summary.

ROLE: You are a researcher cataloging every public profile and bio
associated with a specific name across the web.

CONTEXT: Name: [YOUR FULL NAME]. Current professional positioning:
[1 SENTENCE, e.g. "Fractional CFO for early-stage SaaS companies"].
Past employers/roles for disambiguation: [LIST 2-3].

CONSTRAINTS:
- Find every distinct public profile, bio, or "about" page you can
  locate: old company team pages, conference speaker bios, alumni
  pages, old personal or company websites, directory listings, cached
  social profiles.
- For each, note whether it's still live, and whether the title,
  employer, or description on it conflicts with the current positioning
  I gave you.
- Sort the list with the most contradictory or most visible ones first.

OUTPUT FORMAT: A table — URL, what it says about me, live/dead if you
can tell, conflict level (high/medium/none).

5. What images of you are indexed

Use ChatGPT with browsing, or Perplexity — check whichever you're using actually returns image results, since this varies by tool version.

ROLE: You are a researcher finding publicly indexed photos associated
with a specific person's name.

CONTEXT: Name: [YOUR FULL NAME], plus disambiguating detail:
[e.g. "works in commercial real estate, based in Denver"].

CONSTRAINTS:
- Find and describe publicly indexed images tied to this name: headshots
  on old company or bio pages, conference or event photos, press photos,
  anything you can locate via public web search.
- For each, describe where it's hosted, roughly how old it appears to
  be, and whether it looks current or outdated versus a professional
  headshot from the last 2 years.

OUTPUT FORMAT: A list — where found (with URL), estimated age, current
vs. outdated assessment.

Where this goes wrong

• You run one prompt and call it done. Each of the 5 asks the AI to adopt a different vantage point, and each surfaces different results even for the same name — recruiter framing misses what an adversary framing catches, and vice versa. Run all 5.
• You treat every result as confirmed fact. AI search tools blend and sometimes misattribute information, especially for common names. Click through to the actual source URL before you act on anything, especially anything negative.
• You skip this because "nothing bad exists." The point isn't just finding bad information — it's finding the outdated bio calling you a "Manager" three titles ago, or the old company page still listing you as an active employee two jobs later. That's the more common problem, and it's the easiest one to fix.

The 2-minute version

Open Perplexity. Run Prompt 1 (recruiter view) with your name and current role. Skim for anything outdated or contradictory. If you have more time, run Prompt 4 to catch every stale bio still live — that's usually where the real cleanup work is.

---

Where you stand with AI, in 10 minutes. The free AI Readiness Quiz gives you a personalized roadmap instead of another tool list: genxcelerate.com

Weekly, on Tuesdays. The Gen Xcelerator — practical AI intelligence for people who already know how to do the job. Subscribe free

Experience Is the API. — GenXcelerate