How to search for matching customers with Myaza Trust
Search an authorised face photo against the eligible faces in your organisation. Use this for investigation and review, and keep it distinct from comparing two photos or proving a login.
Charles Archibong, Co-founder
· 3 min read
Key takeaways
- The searched photo has a documented permitted purpose.
- Results link to authorised source evidence.
- No-result, no-hit and possible-hit paths remain distinct.
- A possible match receives review before a consequential action.
Search an authorised face photo against the eligible faces in your organisation. Use this for investigation and review, and keep it distinct from comparing two photos or proving a login.
Before you begin
A clear face photo and permission to use it
Eligible verified faces or face lists in your organisation
Face search available for this deployment
Use a team member with permission for this control. Check the organisation and environment before changing a setting. Review the current price before a paid check or ongoing enrolment. Screenshots show an example workspace or the public integration reference; they do not show a real customer or a completed paid check.
Set up and use the capability
Step 1: Open standalone Face search
In the dashboard, choose your organisation, then open Verification → Spot Checks → Face search. The standalone tool is Production only: in Sandbox, the page explicitly says that the search is unavailable and keeps Search face disabled. Check your face-search permission before continuing. This is separate from duplicate detection inside a verification workflow.
Dashboard example: Open standalone Face search
Step 2: Prepare the photo and consent
Use a clear face photo you are permitted to search. Check the intended purpose and the relevant biometric consent before uploading. A face search needs a query face and an eligible comparison population. An empty or ineligible customer book cannot demonstrate a useful hit, and another organisation’s private customer book is not your population.
Dashboard example: Prepare the photo and consent
Step 3: Review the scope and current price
Read the Pricing panel before running a Production search. It is charged when the search answers; the page explains how unreadable-face submissions are handled. Check the workspace and consent checkbox. The search covers this organisation’s verified faces and its face lists. Do not invent filters the form does not offer. Workflow face search has a separate charging rule shown on the page.
Dashboard example: Review the scope and current price
Step 4: Run the search once and inspect the results
Submit the permitted photo in the Face search tool. Read the returned matched-account or face-list results and their recorded confidence. Inspect the linked source record before taking a customer action. A similarity percentage is comparison evidence, not an allegation of fraud, proof of legal identity or proof that an account is controlled by the person presenting the photo.
Dashboard example: Run the search once and inspect the results
Step 5: Review each possible account connection
Open the authorised customer or list entry linked from a finding. Check names, documents, prior verification and the context for a repeated face. A returning customer or mistaken second registration may be legitimate. Record your investigation rationale in the relevant case or review process instead of adding someone to a blocklist merely because the search returned them.
Integration reference: Investigation workspace
Step 6: Apply recurring detection in onboarding when needed
For continuous duplicate checks at capture time, open the onboarding workflow and enable Presence Intelligence → Rules → Face search. Configure each duplicate finding’s action separately, check Review routing, then save and publish. This protects new runs of that workflow. A standalone search result does not automatically configure or enforce all your onboarding journeys.
Dashboard example: Apply recurring detection in onboarding when needed
Get more from the capability
Use standalone search to answer a specific review question, not as a substitute for a subject-bound authentication session.
Use workflow duplicate detection for the onboarding flows you actually run, with deliberate actions for different names and documents.
Keep findings tied to the original customer or list evidence. A shared face finding does not itself change your application’s account state.
Test before relying on it
Run the supported Sandbox or simulator scenarios for this product. Where a tool is Production only, check access and scope first, then use only a permitted, deliberately reviewed Production test. A documentation screenshot or a successful page load is not an end-to-end integration test.
The searched photo has a documented permitted purpose.
Results link to authorised source evidence.
No-result, no-hit and possible-hit paths remain distinct.
A possible match receives review before a consequential action.
Troubleshooting
The tool is unavailable
Check deployment availability, organisation access and your permissions; do not use another organisation’s book.
No match appears
Check query quality and the eligible verified population. Absence of a hit is not proof of a clean identity.
A match was found but signup still passed
The application needs the published workflow policy and a result handler to enforce its own signup decision.
Open the setup and integration references
Back to Solutions. Choose the capability you want to activate, then follow its own guide.
Sources
Charles Archibong
Co-founder
Charles Archibong co-founded Myaza Trust. He writes about identity verification, financial technology, and the practical work of building trusted digital services.
