Verify public web identity — a practical checklist

DeepSearch Name is public-web research — not a background check and not FCRA-compliant. Use it for professional context with sources you can verify; use a regulated consumer reporting agency for employment, housing, or credit decisions. Read the full distinction.

Most people-research mistakes are identity mistakes: the right name, the wrong human. This checklist helps you verify public web identity before outreach, meetings, CRM updates, or publication. It is a research hygiene practice — not a background check, not legal advice, and not a substitute for FCRA processes when those apply. If your decision is employment, housing, or credit eligibility, use a compliant CRA workflow instead — see people search is not a background check.

Verification is the habit that makes every other people-research skill safe to use: Google operators, email clues, AI briefs, and CRM enrichment all fail in the same way when you skip it.

Interactive checklist

Check items off as you confirm them for the person in front of you. Use it as a gate before outreach, CRM merges, or meeting talking points — especially when the name is common or the only lead is an email thread.

The verification loop

Every material claim should survive a source click. If the brief or a directory says “VP at Acme,” open a page that actually supports it. If you cannot, downgrade confidence or keep looking. Summaries — human or AI — are maps. Maps are not territory.

Loop in practice: claim → open source → compare to known clues → accept, reject, or park as uncertain → only then act (email, meeting talking point, CRM field, publish). Skipping the open-source step is how confident mistakes happen.

Why verification fails

Same-name collisions

Common names produce parallel careers in the same city. Filters help; explicit candidate comparison helps more. Guide: common name disambiguation. Two people can share employer industry and metro without being the same human — check timeline and role specificity.

Copied bios and SEO mirrors

Aggregators reprint the same paragraph across domains. Ten copies are still one signal. Prefer employer sites, conference organizers, and original publications. If every page uses identical wording, treat them as one source with many URLs.

Stale titles

People change roles faster than some pages update. Look for a recent corroborating source when the meeting depends on current scope. A 2022 PDF that says “Head of X” is a historical signal, not proof of today’s org chart.

Motivated matching

When you need the meeting to go well, your brain fills gaps. Write clues down first; force yourself to list disconfirming evidence. Ask: “What would prove this is the wrong person?” If you cannot name a falsifier, you are not verifying — you are hoping.

Photo overconfidence

Faces feel definitive and travel poorly across the web. Stock-adjacent headshots, old photos, and lookalikes mislead. Use photos as a weak secondary check after employer and role align — never as the primary key.

Single-platform trust

One polished LinkedIn headline can be outdated, aspirational, or simply about someone else with the same name. Cross-check with Google advanced search or an employer page before you personalize.

Signal strength guide

Strong signals can still be wrong in isolation. Strength describes relative trust, not a license to skip a second check when the name is common or the stakes are high.

Checklist in prose form

1. Name spelling matches an authoritative public page (watch hyphens, diacritics, maiden/married variants).
2. Employer or industry aligns with what you already know from the invite or email domain.
3. Career timeline is consistent across at least two sources — or you have a documented reason for the gap.
4. Location or role is plausible for the meeting context (remote roles still leave public traces).
5. You opened at least two source links — not just skimmed a summary.
6. Conflicts are noted before you act (outreach, CRM, publish).
7. Personalization claims trace to a page you actually read.

Worked mini-cases

Case 1: Common name, clear company

Invite says “Sam Patel, Northwind.” Search yields three Sam Patels in tech. You open Northwind’s team page — one match — then a conference PDF with the same affiliation. Two independent signals: proceed. You do not need a deep personal dig.

Case 2: Email-only lead

You have alex@brightfield.io and no name certainty. Domain → company graph → name hypothesis → candidate confirm. Follow find a person by email, then return to this checklist before outreach.

Case 3: Conflicting titles

LinkedIn says VP; company blog from last month says Director. You note the conflict, prefer the employer’s recent page for the meeting, and avoid arguing titles out loud. Verification includes knowing what you will not claim.

Case 4: Thin footprint

Almost nothing public beyond a sparse profile. Options: ask the introducer, request a URL, keep the meeting, skip speculative personalization. Forcing a “verified” identity from weak data is how wrong-person emails happen.

Workflow by situation

Before outreach

Confirm identity, then personalize from a source you actually opened. Wrong-person email is costly and awkward. Sales context: prospect research.

Before a meeting

Pair this checklist with the ten-minute prep flow in research someone before a meeting. Identity first; talking points second.

Before CRM updates

Do not merge records because names match. Attach source URLs in notes when your process allows it. A bad merge pollutes every future touch.

Before publication

Journalists and marketers: affiliation errors are public and sticky. Prefer on-the-record bios and direct confirmation when stakes are high — for journalists.

Before internal intros

When you connect two people over email, verify both sides enough that you do not introduce the wrong Sam. A mistaken intro wastes social capital for everyone on the thread.

Using DeepSearch Name inside the checklist

Name-first search → candidate confirmation → sourced brief → click citations → chat for follow-ups. The product accelerates collection; you still own verification. Learn more about source-linked profiles and how it works. Lookups stay private — private people search — but privacy of the lookup does not reduce the need to confirm identity.

Practical rule: if you cannot click a citation for a claim you plan to use, delete the claim from your notes. Fluency without sources is not verification.

Team habits that scale

• Require two-source confirmation before CRM “current employer” edits
• Store URLs, not vibes, in shared notes
• Separate research prep from regulated screening artifacts
• Re-verify when a deal or story sits idle for months
• Praise teammates who pause on conflicts — speed without accuracy is expensive

What this checklist is not

• Not an FCRA consumer report or employment screen
• Not permission to access private accounts or hacked data
• Not a substitute for asking a human when public sources are thin
• Not a guarantee — public web coverage varies by person
• Not legal advice for your jurisdiction or industry

Read people search is not a background check when stakeholders confuse categories. Deeper comparison: people search vs background check.

If you cannot verify

Options: ask the introducer, request a LinkedIn URL, postpone personalization, or keep the meeting but avoid factual claims you cannot support. Acting on a weak match is optional; owning the mistake later is not fun. “I’m not sure I have the right person — are you at Northwind?” is a professional sentence. Sending a paragraph about the wrong career is not.

Confidence labels you can actually use

Labels beat vibes in shared systems. “Looks right” is not a label. If your team cannot agree what “confirmed” means, reuse the two-independent-sources rule until you have something better.

Verification under time pressure

Real calendars do not always give you twenty quiet minutes. A minimum viable verify under pressure: (1) match employer or email domain, (2) open one primary page, (3) skip any claim you did not open. That is weaker than a full checklist and still better than ranking-based guesswork. After the call, upgrade probable matches to confirmed when you can.

For operator speed, keep a short query kit from Google advanced search for people. For meeting rhythm, pair with pre-meeting people research.

Red flags that should stop you

• Two current employers with no date explanation
• Photo match with conflicting career timeline
• Only scraper directories, zero primary pages
• Introducer details contradict every public page
• You feel tempted to “just send it” despite a nagging mismatch

Stopping is a skill. Speed culture rewards sending; accuracy culture rewards the pause. Wrong identity errors are almost never fixed by typing faster.

Documenting verification without bureaucracy

You do not need a legal memo for every LinkedIn glance. You do need a lightweight trail when the name is common or the output will be reused. Three lines suffice: anchors (employer + role), two URLs, confidence label. Paste that into the CRM note or the meeting doc. Future you — or a teammate — can re-open the links in seconds.

Avoid screenshots without URLs. Avoid “verified by intuition.” Avoid merging records because the first names matched and the industries felt adjacent. Documentation is how verification survives handoffs; without it, every new owner reintroduces motivated matching.

If you use DeepSearch Name, the sourced brief already nudges toward this trail — still copy the two strongest primary links into whatever system your team actually reads. Product pages: people search by name.

For email-derived leads, verify only after you have a name hypothesis worth testing — find a person by email. For meeting prep, verification is step one of the ten-minute ritual, not an optional epilogue. Make the checklist a team default whenever a name is ambiguous; optional only when identity is already obvious from a trusted primary page you opened yourself.

Verification when sources disagree on dates

Title lag and stale bios are normal. Prefer the newest independent primary page when dates conflict, and write the conflict into your notes instead of averaging two careers into one fictional timeline. If employer A and employer B both look current, treat them as unresolved until a primary source settles it — or ask a human introducer. Acting on a blended timeline is how wrong-person outreach and wrong CRM merges start. For name collisions that cause most date conflicts, start with common name disambiguation.

Bottom line

Verify public web identity with two independent sources, open the links behind claims, and document conflicts before you act. Use the interactive checklist above every time the name is ambiguous or the stakes are non-trivial. When you want help gathering candidates and citations, start a DeepSearch Name search — and still click through. Identity first; narrative second; action last. The checklist is short on purpose: if you cannot finish it, you are not ready to personalize, merge, or publish. Slow is smooth; smooth is fast when the alternative is a wrong-person email.