<img height="1" width="1" style="display:none;" alt="" src="https://px.ads.linkedin.com/collect/?pid=1611884&amp;fmt=gif"> Skip to main content
NewVouched is live in Epic MyChartLearn More
Talk to salesStart free
  1. Solutions
  2. Use cases
  3. Match & dedupe records

Use case: Match & dedupe records

One person, one record.

Names change, addresses move and typos multiply. When your matching system flags a possible duplicate, Vouched compares the faces on each record and returns a likelihood of duplication, so your team decides with evidence.

Talk to salesStart free
Two records compared by faceYour system flags Jonathan Reyes, 12 Elm St, from 2019, and Jon Reyes, 12 Elm Lane, from 2024, as a possible duplicate. Vouched compares the face from a selfie on one record with the ID photo on the other and returns a high likelihood of duplication, marked for review. POSSIBLE DUPLICATE · FLAGGED BY YOUR SYSTEMRECORD AJonathan Reyes12 Elm St · 2019FROM SELFIERECORD BJon Reyes12 Elm Lane · 2024FROM ID PHOTOVSFaces comparedNames and addresses differ; the face does notLIKELIHOOD OF DUPLICATIONHighREVIEW

The problem

Duplicates split people in two.

In healthcare a duplicate splits a patient’s history; in marketplaces and finance, duplicate accounts hide fraud rings and abuse.

Mortality

5×

More likely to die in hospital with duplicate medical records

Prevalence

5–10%

Of U.S. patients thought to have duplicate records

Safety

Top 10

ECRI threat to patient safety: misidentification

Fraud

8%+

Of account creation attempts flagged as suspected fraud in 2025

Sources: BMJ Group study, 2026 · ECRI, via Altera Digital Health, 2024 · Sift data, via NHIMG, 2025

How it works

Your system finds the candidates. Vouched compares the faces.

Vouched doesn’t search your records for duplicates. It adds face evidence to the candidates your matching system already flags.

  1. 1

    Flag

    Your EMPI, MDM or registration system flags records that may belong to the same person.

  2. 2

    Compare

    Vouched compares the face from a selfie or ID photo on each record, whatever the names and addresses say.

  3. 3

    Decide

    You get a likelihood of duplication, so teams merge true duplicates and keep different people apart.

Vouched compares faces only, not ID document data, names or addresses, and records don’t need prior Vouched verification.

Compared

How record matching methods hold up.

Demographics drift over time. A face doesn’t.

Reset methodSurvives name and address changesAdds independent evidenceEffort per case
Demographic matchingNoNames and addresses changeNoAutomated, but error-prone
Manual reviewPartlyDepends on the reviewerPartlyWhat the records containMinutes to hours per case
Probabilistic matching scorePartlyWeights changed fieldsNoAutomated, with a review queue
Vouched face comparisonYesFaces don’t change with an addressYesA biometric likelihoodSeconds per pair

Who it’s for

Wherever one person can become two records.

Health records, customer files and marketplace accounts.

Health systemsHIM teamsResolve duplicates and overlays in the patient record.Patient matching →Telemedicine and digital healthPatient recordsKeep one record per patient across brands and states.Telemedicine and digital health →Gig economyMarketplace accountsSpot one person behind many worker or seller accounts.Gig economy →Financial servicesCustomer filesCatch duplicate customers before they become duplicate risk.Financial services →

Built on

Built on Vouched Assure.

Matching uses the same face technology as recovery and step-up.

Vouched AssureMatching and dedupFace comparison across flagged records.Vouched VerifyVerified registrationStop new duplicates by verifying at sign-up.Vouched AssureRe-authenticationRe-prove returning users against the right record.

The proof

Face evidence, in seconds.

The same accuracy that powers verification powers every comparison.

Accuracy

99%

ID document authentication accuracy

Completion

97%

First-attempt completion at registration

Source: Vouched product data

Related use cases

Other jobs Vouched does.

Each use case runs on the same verified identity, so one integration covers them all.

Use caseOnboard a new userVerify new customers, patients or employees at sign-up.Read more →Use caseRecover an accountGet locked-out users back in with a selfie, not a help desk call.Read more →Use caseReset a passwordSelf-service password and MFA resets that can’t be talked around.Read more →Use caseStep up a loginRe-prove the person before high-risk actions.Read more →Use caseMeet KYC / AMLIdentity, data and sanctions checks in one flow.Read more →Use caseGovern AI agentsDetect agents and tie every action to a verified person.Read more →

Trust

Built for the security review.

Everything your security review asks for, from the SOC 2 report to the BAA, is on one page.

Visit the Trust Center
HIPAA and BAABuilt to support HIPAA compliance; BAA in place
NIST SP 800-63A IAL2Kantara-approved workflow
SOC 2 Type IIReport available under NDA
ISO 27001Information security management
Zero access to clinical recordsIdentity data only
Biometric privacyNotice and consent on every selfie

FAQ

Record matching, answered.

Straight answers about how Vouched helps you match & dedupe records.

Contact us with a question
Does Vouched find duplicate records?

No. Your EMPI, MDM or registration system finds candidate duplicates. Vouched compares the faces on the flagged records and returns a likelihood of duplication.

What does Vouched compare?

Only faces, from a selfie or an ID photo on each record. It doesn’t compare ID document data, names or addresses.

Do records need prior Vouched verification?

No. Vouched can compare any records that hold a selfie or ID photo.

Why do duplicate records matter in healthcare?

A 2026 BMJ Group study found patients with duplicate records were almost five times more likely to die in hospital, and 5% to 10% of U.S. patients are thought to have duplicates.

Does this work outside healthcare?

Yes. Marketplaces and financial services use face comparison to spot one person behind many accounts.

How do we prevent new duplicates?

Verify identity at registration, so each new record starts from a verified person rather than typed demographics.

Who makes the final merge decision?

Your team. Vouched returns a likelihood; your matching system and reviewers decide whether to merge or keep records apart.

How fast is a comparison?

Seconds per pair of records.