Skip to main content
PDF4me APIIntegrationsKnowledge BaseExtensionsAPI TesterBlog

AI-Marriage Certificate Parser in Power Automate

PDF4me AI-Marriage Certificate Parser is a Power Automate action that reads a marriage certificate PDF or image and returns over 30 structured fields covering spouse details, officiant information, and certificate identifiers. Use it to digitize vital records, verify identity documents, or screen certificates for fraud without manual data entry.

What this action does

PDF4me AI-Marriage Certificate Parser extracts structured data from marriage certificate PDFs and images directly inside your Power Automate flow. Feed it binary content from SharePoint, OneDrive, email attachments, or Dropbox and receive back 30+ structured fields covering certificate identifiers, both spouses' personal data, officiant and witness details, official seal and signature presence, and optional AI-powered fraud and authenticity detection. The structured output slots directly into SharePoint lists, Dynamics 365 records, Excel workbooks, SQL databases, or any downstream Power Automate step without manual data entry.

Related Blog Posts
No blog post yet for this feature — coming soon.
In the meantime, browse the PDF4me blog for tutorials and workflows across every platform.
Visit the blog →

How Do I Authenticate My Power Automate Flow?​

Every PDF4me action in Power Automate requires a valid API key. Create one in the PDF4me developer dashboard and enter it when the PDF4me connector prompts for credentials on first use.

Get your API Key

Important Facts You Should Not Miss​

File name with extension is required even for binary inputs
Always set Marriage Certificate Name to the actual filename including its extension such as certificate.pdf or scan.png. The action uses this string for format detection and AI model selection. Omitting the extension or passing a generic placeholder name reduces extraction accuracy.
Authenticity verification adds processing time but enables fraud detection
Setting verifyAuthenticity to Yes activates the fraud and tamper detection pipeline, populating authenticityIssues, fraudIndicators, and authenticityRecommendation in the response. For high-volume batch flows where throughput matters and authenticity is verified downstream, leave it as No.
Check the warnings array before writing to your records system
When the AI encounters low-confidence fields or difficult handwriting, it populates the warnings array and sets fallbackUsed to true. Add a condition step after this action to route certificates with non-empty warnings to a manual review approval flow before committing extracted data to your records system.

Parameters​

Required fields: Marriage Certificate File Content and Marriage Certificate Name are mandatory for every request. Body/verifyAuthenticity and Body/customFieldKeys are optional; omit them entirely when their functionality is not needed.

AI-Marriage Certificate Parser action in Power Automate showing Marriage Certificate File Content, Marriage Certificate Name, Body verifyAuthenticity, and Body customFieldKeys fields

Full Parameters view of the AI-Marriage Certificate Parser action in Power Automate.

ParameterRequiredWhat it doesExample
Marriage Certificate File ContentRequiredBinary data of the marriage certificate. Map from the output of a preceding step such as Get file content (SharePoint), Get file content (OneDrive), or an email attachment retrieval. Accepts PDF, PNG, JPG, and JPEG formats.body('Get_file_content')
Marriage Certificate NameRequiredFilename of the certificate including its extension. Used for format detection and AI model selection. Must include the correct extension matching the actual file type: .pdf, .png, .jpg, or .jpeg.marriage_certificate_2024.pdf
Body/verifyAuthenticityOptionalEnable AI-powered authenticity and fraud detection. When Yes, the response includes authenticityIssues, fraudIndicators, and authenticityRecommendation. Adds processing time. Omit or set to No when speed is the priority.No
Body/customFieldKeysOptionalOptional array of custom field keys to extract beyond the standard certificate fields. Each string names a custom field the AI will attempt to locate and return inside the extractedFields object of the response.["registrationRegion", "notaryId"]

Authenticity Verification Options​

Controls whether the action runs AI-powered fraud and tamper detection on the certificate.

NoDefault: skip authenticity checks
Authenticity verification is skipped. The action extracts all certificate fields without running fraud analysis. Use this for high-volume batch processing flows or when document authenticity is validated by a separate downstream step. Faster response times.
YesEnable fraud and tamper detection
Activates the full authenticity analysis pipeline. The response includes authenticityIssues (specific issues found), fraudIndicators (detected fraud signals), and authenticityRecommendation (overall assessment). Use for identity verification, legal document processing, immigration, and insurance workflows where certificate legitimacy must be confirmed.

Output Fields​

FieldTypeWhat it contains
certificateNumberStringUnique certificate or registration number extracted from the document.
dateOfMarriageStringDate the marriage ceremony took place. Returned in ISO 8601 format (YYYY-MM-DD).
placeOfMarriageStringLocation or venue where the marriage ceremony was conducted.
dateIssuedStringDate the certificate was issued by the authority. Returned in ISO 8601 format (YYYY-MM-DD).
issuingAuthorityStringName of the government body or agency that issued the certificate.
spouse1FullNameStringFull legal name of the first spouse as it appears on the certificate.
spouse1AgeNumberAge of the first spouse at the time of marriage.
spouse1BirthDateStringBirth date of the first spouse in ISO 8601 format (YYYY-MM-DD).
spouse1BirthplaceStringBirthplace of the first spouse as extracted from the certificate.
spouse1ResidenceBeforeMarriageStringResidential address of the first spouse prior to the marriage.
spouse2FullNameStringFull legal name of the second spouse as it appears on the certificate.
spouse2AgeNumberAge of the second spouse at the time of marriage.
spouse2BirthDateStringBirth date of the second spouse in ISO 8601 format (YYYY-MM-DD).
spouse2BirthplaceStringBirthplace of the second spouse as extracted from the certificate.
spouse2ResidenceBeforeMarriageStringResidential address of the second spouse prior to the marriage.
officiantNameStringFull name of the officiant who performed the ceremony.
officiantTitleStringTitle or role of the officiant, for example Minister, Judge, or Registrar.
officiantLicenseNumberStringLicense or registration number of the officiant, where present on the certificate.
witnessesStringNames and details of witnesses present at the ceremony.
officialSealStringDescription or indicator of the official seal or stamp on the certificate.
signaturesPresentStringIndicator of which signatures are present: spouses, officiant, and witnesses.
authenticityIssuesArrayIssues found during authenticity verification. Populated only when verifyAuthenticity is Yes.
fraudIndicatorsArrayPotential fraud or tampering indicators detected by the AI model. Populated only when verifyAuthenticity is Yes.
authenticityRecommendationStringOverall recommendation on certificate authenticity from the AI analysis.
extractedFieldsObjectKey-value pairs of any custom fields requested via Body/customFieldKeys.
warningsArrayWarning messages for fields that had low confidence or could not be parsed reliably.
fallbackUsedBooleanTrue when a secondary extraction method was used due to limitations of the primary method.
rawOcrTextStringFull raw OCR text extracted from the document before structured field parsing.
jobIdStringUnique identifier for the AI processing job. Use for support and debugging.
jobIdExtStringExternal job identifier for linking to third-party tracking or case management systems.
successBooleanTrue when the extraction completed successfully; false when a processing error occurred.
messageStringHuman-readable summary of the extraction result or description of any error encountered.

How Do I Set Up AI-Marriage Certificate Parser in Power Automate?​

  1. Log in to Power Automate at make.powerautomate.com and create a new flow or open an existing one.
  2. Choose a trigger that delivers the certificate file: SharePoint (When a file is created), Outlook (When a new email arrives) for attachment-based intake, or OneDrive (When a file is created) for folder-based intake.
  3. Add a Get file content step (SharePoint, OneDrive, or Dropbox) to retrieve the binary content of the certificate file, or use Get attachment content for email-based flows.
  4. Click New step, search for PDF4me, and select the AI-Marriage Certificate Parser action. Enter your PDF4me API key when the connector prompts for credentials.
  5. Set Marriage Certificate File Content to the binary output of the previous Get file content step using the dynamic content picker.
  6. Set Marriage Certificate Name to the certificate filename including its extension, for example certificate.pdf, or use a dynamic expression from the trigger to pull the filename automatically.
  7. Set Body/verifyAuthenticity to Yes if your workflow requires fraud or tamper detection. Leave it as No for higher throughput batch flows where authenticity is not a primary concern.
  8. Optionally populate Body/customFieldKeys with an array of custom field names if you need to extract fields beyond the standard certificate data.
  9. Add downstream steps that consume extracted output fields using dynamic content: certificateNumber, spouse1FullName, dateOfMarriage, authenticityRecommendation, and others as needed.
  10. Add a Condition step that checks whether the warnings array is non-empty and routes those certificates to a manual review approval flow before writing to your records system.
  11. Test the flow with a sample marriage certificate and verify the extracted fields in the flow run history.

Workflow Examples​

Workflow ExamplesCommon Power Automate flow patterns using AI-Marriage Certificate Parser.
Marriage certificate uploaded to SharePoint → extract data → create vital records entry → archive
  1. A SharePoint trigger fires when a new certificate is added to the Incoming Certificates document library.
  2. A Get file content step retrieves the certificate binary from SharePoint.
  3. AI-Marriage Certificate Parser extracts all fields with verifyAuthenticity set to Yes.
  4. A Condition step checks whether the warnings array is empty. Non-empty warnings trigger an approval request routed to the records supervisor.
  5. A Create item step adds a new row to the Vital Records SharePoint list with certificateNumber, spouse1FullName, spouse2FullName, dateOfMarriage, placeOfMarriage, issuingAuthority, and authenticityRecommendation mapped from the output.
  6. A Move file step archives the original certificate to the Processed Certificates library and tags it with the extracted certificateNumber as metadata.
  7. A Send an email step notifies the records management team with an extraction summary including the jobId for audit reference.
Certificate received by email → identity verification → cross-reference identity system → approval routing
  1. An Outlook trigger fires when an email with a marriage certificate attachment arrives in the verification mailbox.
  2. A Get attachment content step retrieves the certificate binary from the email.
  3. AI-Marriage Certificate Parser extracts spouse names, birth dates, and certificate details with verifyAuthenticity set to Yes.
  4. An HTTP action queries the identity management system with spouse1FullName and spouse1BirthDate to check for a matching identity record.
  5. A Condition step branches on authenticityRecommendation. Certificates with issues are routed to a security team approval request via the Approvals connector.
  6. A Create item step logs the extracted data and identity match result in a SharePoint verification log list.
  7. A Send an email step returns the verification outcome to the requester.
Certificate batch intake → fraud risk scoring → alert generation → fraud monitoring dashboard update
  1. A manual or scheduled trigger processes a batch of certificates from a SharePoint queue folder.
  2. An Apply to each loop iterates over each certificate file in the folder.
  3. AI-Marriage Certificate Parser processes each certificate with verifyAuthenticity set to Yes.
  4. A Condition step evaluates the length of the fraudIndicators array. High indicator counts set a risk tier variable to High.
  5. For High risk tier, a Send an email step alerts the fraud team immediately with the extracted certificateNumber and fraudIndicators list.
  6. A Create item step adds an investigation record to the Fraud Cases SharePoint list with all extracted fields and the calculated risk tier.
  7. An HTTP action posts the fraud screening result to a Power BI streaming dataset for real-time monitoring dashboard updates.
Legal document intake → compliance validation → legal records system entry → secure archive
  1. A SharePoint trigger fires when a certificate is uploaded to the Legal Documents library.
  2. A Get file content step retrieves the certificate binary.
  3. AI-Marriage Certificate Parser extracts all fields with verifyAuthenticity set to Yes.
  4. A Condition step validates that issuingAuthority, officiantName, and officiantLicenseNumber are all non-empty. Missing fields trigger a routing action to a legal clerk for manual review.
  5. A Condition step checks that dateOfMarriage falls within jurisdiction-specific validity windows stored in a SharePoint configuration list.
  6. A Create item step adds the certificate to the legal document management SharePoint list with all extracted fields.
  7. An Append to array variable step adds a timestamped audit log entry including the jobId and processing result for the compliance audit trail.
  8. A Move file step places the original certificate in the secure archive library with the extracted certificateNumber and dateIssued as metadata tags.

Frequently Asked Questions​

What file formats does the AI-Marriage Certificate Parser support?+
The action accepts PDF, PNG, JPG, and JPEG files. For best accuracy, use a high-resolution scan (300 DPI or higher) or a digitally-issued PDF. Blurry or low-resolution images will reduce field extraction confidence and may cause some fields to return as null or empty. The Marriage Certificate Name field must include the correct file extension so the action can select the right AI model for the format. Date fields such as dateOfMarriage and dateIssued follow <a href="https://www.iso.org/iso-8601-date-and-time-format.html" target="_blank" rel="noopener noreferrer">the ISO 8601 date format</a>, and certificate content generally follows the recordkeeping conventions described by the <a href="https://www.cdc.gov/nchs/nvss/marriage-divorce.htm" target="_blank" rel="noopener noreferrer">CDC National Vital Statistics System</a>.
What does verifyAuthenticity do and when should I enable it?+
When verifyAuthenticity is set to Yes, the AI model runs additional checks to detect signs of document tampering, forgery, inconsistent data, or missing security features. The results appear in three output fields: authenticityIssues (specific issues detected), fraudIndicators (potential fraud signals), and authenticityRecommendation (overall assessment string). Enabling this adds processing time to each request. Enable it for identity verification, immigration, insurance claims, and legal workflows where certificate legitimacy must be confirmed. Leave it as No for high-volume batch processing flows where speed is the priority and authenticity is validated separately.
Which output fields are always returned and which are conditional?+
The fields success, message, jobId, and rawOcrText are always present in every response. Certificate-specific fields such as certificateNumber, dateOfMarriage, placeOfMarriage, spouse1FullName, and officiantName are populated when the AI successfully extracts them from the document; they are returned as null or empty string when extraction is not possible. The authenticityIssues and fraudIndicators arrays are only populated when verifyAuthenticity is Yes; they return as empty arrays when verification is disabled. The extractedFields object is only populated when customFieldKeys are provided. The fallbackUsed boolean indicates whether the action used a secondary extraction method.
How do I handle certificates where the AI cannot extract some fields?+
Check the warnings array in the action output. When extraction confidence is low or a field cannot be parsed, the action populates warnings with descriptive messages and sets fallbackUsed to true. Add a Condition step after the action that checks the length of the warnings array. Route certificates with non-empty warnings to a manual review approval flow using the Approvals connector before writing data to your records system. This prevents low-confidence data from entering production systems while keeping high-confidence records fully automated.
How do I store the extracted data in a SharePoint list or Dynamics 365?+
After the action completes, add a Create item step targeting your SharePoint list or a Create a new record step for Dynamics 365. Use the dynamic content picker to map output fields: set the Title or Name column to the certificateNumber output, and map spouse1FullName, spouse2FullName, dateOfMarriage, placeOfMarriage, issuingAuthority, and authenticityRecommendation to their corresponding columns. For Dynamics 365, map to the Contact or custom entity fields as appropriate for your data model.

Same Task on Other Platforms​

Get Help​

PDF4me Connector DocsGet your API KeyPower Automate Documentation