Managed services provider
Genpact
Genpact, an NYSE-listed Bermuda company, provides finance and accounting services and offers agentic AI suites for the close, payables and FP&A.
Function
- Ownership
- PublicSourceGenpact Limited, Annual Report on Form 10-K for the fiscal year ended December 31, 2025
Common shares (G) listed on the New York Stock Exchange; organized under Bermuda law, per its Form 10-K for 2025.
- Headquarters
- BermudaSourceGenpact Limited, Annual Report on Form 10-K for the fiscal year ended December 31, 2025
Based on the 10-K, which gives Hamilton, Bermuda as the principal executive office and registered office; Genpact's press releases carry a New York dateline.
- SEC reporting
- YesSourceGenpact Limited, Annual Report on Form 10-K for the fiscal year ended December 31, 2025
- Audit firm
- No
No statutory audit practice is described in the sources read; the flag is false on that basis, not on an independently sourced negative.
- Website
- SourceGenpact Record-to-Report Suite
- Subprocessor list
- Not stated
- Contract terms
- Not stated
Offering · AI agent
Genpact AP Suite
Genpact's AP Suite applies agentic AI to accounts payable through four modules: AP Capture, AP Advance, AP Trace and AP Assist. SourceGenpact AP Suite
Genpact says its agents read and validate invoices, manage matching, exceptions and auto-posting, route to people when needed, and learn from feedback over time. SourceGenpact AP Suite
AP Trace looks for duplicate invoices and unusual patterns, and AP Assist handles supplier queries in several languages. SourceGenpact AP Suite
Genpact says client data stays in the client's environment under the agreed deployment model and is not used to train models for other clients. SourceGenpact AP Suite
Where it fits
An accounts payable agent layer that Genpact says sits over existing ERP and cloud AI platforms rather than replacing them. SourceGenpact AP Suite
CoE analysis Auto-posting and duplicate detection touch payment controls, so the thresholds and approval points deserve as much scrutiny as the touchless rates.
What it does
- Offering type
- AI agentSourcesGenpact AP SuiteGenpact Limited, Annual Report on Form 10-K for the fiscal year ended December 31, 2025
- Workflow areas
- Accounts payableSourceGenpact AP Suite
- Customer size
- Not stated
AI
- Kind of AI
- SourceGenpact AP Suite
- Agents and automated actions
- Established machine learning and rules
- Suggests or acts
- Suggests and actsSourceGenpact AP Suite
- Availability
- Not stated
No availability, edition or regional terms on the product page.
- Models named
- Not stated
Control
- Who controls it
- SourceGenpact AP Suite
- Professional services firm
- Customer finance team
Commercial
- Pricing model
- Not stated
- Public price
- Not stated
Evidence
- Independent outcomes
- None found in the sources read.
- Sources read
- Vendor-stated 5
- Independent market evidence 4
- Independently verified outcomes None read
- Pricing None read
- Known limitations None read
- Coverage
- Directory record
- Last verified
Delivery and lock-in
- Delivery model
- Not stated
- Run service offered
- Not stated
- Team qualifications published
- Not stated
Not stated in the sources read.
- IP and post-engagement license
- Genpact's 10-K says its solutions often include proprietary methodologies, software and reusable knowledge capital, and that "our agreements with our clients govern the ownership of such intellectual property." No terms for this product, or for use of its agents after an engagement ends, are published in the sources read. [C-5108]SourceGenpact Limited, Annual Report on Form 10-K for the fiscal year ended December 31, 2025
- Savings claims (firm-stated)
- SourceGenpact AP Suite
Stated by the firm, not verified
up to 99% of discount opportunities captured
A share of available discount opportunities, not a monetary amount; marked on the page as a projected estimate. No baseline, population, period, or treatment of service fees is stated.
- Alliance partners
- SourceGenpact and Google Cloud Expand Alliance to Bring Agentic Solutions to the Office of the CFO (May 7, 2026)
- Google Cloud
Security and control
- Single sign-on
- Not covered in this record
- User provisioning
- Not covered in this record
- Role-based access
- Not covered in this record
- Administrator controls for AI
- Not covered in this record
- Audit logs
- Not covered in this record
- Certifications
- Not covered in this record
- Data retention
- Not covered in this record
- Data residency
- Not covered in this record
- Training on customer data
- Not covered in this record
AI features
As the vendor or firm describes them. Notes in small print record what the sources leave open.
AP Capture
Agentic AI extracts supplier data across invoice formats and classifies, enriches and cleans it.
- Kind
- Established machine learning and rules
- Suggests or acts
- Not stated
- Who controls it
- Professional services firm
- Availability
- Not stated
Control: Not stated.
SourceGenpact AP SuiteAP Advance
Three-way matching, exception handling and non-PO invoices, with AI-led auto-posting and recommended actions.
- Kind
- Agents and automated actions
- Suggests or acts
- Suggests and acts
- Who controls it
- Professional services firm
- Availability
- Not stated
Control: Not stated.
SourceGenpact AP SuiteAP Trace
Algorithms identify duplicate invoices and unusual patterns, using feedback to reduce false positives.
- Kind
- Established machine learning and rules
- Suggests or acts
- Suggests
- Who controls it
- Professional services firm
- Availability
- Not stated
Control: Not stated.
SourceGenpact AP SuiteAP Assist
Handles supplier queries across languages, identifies intent and automates responses.
- Kind
- Agents and automated actions
- Suggests or acts
- Acts
- Who controls it
- Professional services firm
- Availability
- Not stated
Control: Not stated.
SourceGenpact AP Suite
Questions to ask
For this offering
- Which steps do agents or automation complete without a person approving them (for example posting entries, clearing reconciliations or releasing payments), and how are those thresholds set, changed and evidenced?
- Which invoices can AP Advance post without human approval, and how do AP Trace alerts block or release a payment?
- Will you provide a SOC 1 Type 2 report for the outsourced finance processes, and does its scope include the AI platform and agents used to deliver them, including changes to models and prompts?
- Is our data used to train, tune or improve any model, rule or shared knowledge base, including one used for other clients, and where is it processed and retained?
- Which model providers and sub-processors sit behind the AI features, and how will you notify us when they change?
- Can you show the population, baseline and period behind the outcome figures you publish, and will you commit to any of them in the contract as service levels?
From the diligence question set
- DQ-31 Who, by name and role, will design, build, test and run the AI in our finance process, and what AI has each of those people personally put into production?
- DQ-32 How does your firm decide that someone is qualified to build or govern AI for financial reporting, and who outside the delivery team reviews model risk before anything goes live?
- DQ-33 When the engagement ends, what will we own and be able to run ourselves, and what will work only with your firm? On what terms are your accelerators licensed afterward, and what does it cost to leave a run service?
- DQ-34 What will this cost in years two to five, including run fees, platform licenses, AI usage fees and change requests, and how does that compare with the savings case once the internal oversight roles we keep are counted?
- DQ-35 Do you receive fees, rebates, resale margins or other incentives from the software you are recommending, and do you, or will you, audit our financial statements?
- DQ-36 What results from clients like us can you show that someone other than your firm has verified, and by what method?
Offering · AI agent
Genpact Finance Actions Hub
Genpact describes Finance Actions Hub as agentic AI-assisted decision support for FP&A. SourceGenpact Finance Actions Hub
It names four agents: forecasting and scenario, insight and diagnostics, anomaly detection, and an insight narrative agent that drafts commentary for reporting packs. SourceGenpact Finance Actions Hub
Genpact says human-in-the-loop review is built in at every step and that the product uses a proprietary, industry-specific fine-tuned model rather than a generic large language model. SourceGenpact Finance Actions Hub
Where it fits
CoE analysis Decision support for FP&A that drafts analysis and commentary for people to review, not a planning system of record.
CoE analysis Teams would still need to establish which data the model was tuned on and how generated narratives are checked before they reach a reporting pack.
What it does
- Offering type
- AI agentSourcesGenpact Finance Actions HubGenpact Limited, Annual Report on Form 10-K for the fiscal year ended December 31, 2025
- Workflow areas
- SourceGenpact Finance Actions Hub
- Planning and FP&A
- Finance data and analytics
- Reporting and disclosure
- Customer size
- Not stated
AI
- Kind of AI
- SourceGenpact Finance Actions Hub
- Agents and automated actions
- General-purpose assistant
- Established machine learning and rules
- Suggests or acts
- SuggestsSourceGenpact Finance Actions Hub
- Availability
- Not stated
The page invites contact with an expert; outcomes are to be validated during project initiation.
- Models named
- Not stated
Control
- Who controls it
- Professional services firmSourceGenpact Finance Actions Hub
Genpact says human-in-the-loop review is built in at every step; who can change configuration, prompts, data connections or behavior is not stated.
Commercial
- Pricing model
- Not stated
- Public price
- Not stated
Evidence
- Independent outcomes
- None found in the sources read.
- Sources read
- Vendor-stated 3
- Independent market evidence 4
- Independently verified outcomes None read
- Pricing None read
- Known limitations None read
- Coverage
- Directory record
- Last verified
Delivery and lock-in
- Delivery model
- Not stated
- Run service offered
- Not stated
- Team qualifications published
- Not stated
Not stated in the sources read.
- IP and post-engagement license
- Genpact's 10-K says its solutions often include proprietary methodologies, software and reusable knowledge capital, and that "our agreements with our clients govern the ownership of such intellectual property." No terms for this product, or for use of its agents after an engagement ends, are published in the sources read. [C-5108]SourceGenpact Limited, Annual Report on Form 10-K for the fiscal year ended December 31, 2025
- Savings claims (firm-stated)
- SourceGenpact Finance Actions Hub
Stated by the firm, not verified
up to 40% reduction in manual reporting effort
Labeled "indicative targeted outcomes only"; the page says actual potential outcomes must be validated during project initiation. No baseline, population, period, or treatment of service fees is stated.
- Alliance partners
- SourceGenpact and Google Cloud Expand Alliance to Bring Agentic Solutions to the Office of the CFO (May 7, 2026)
- Google Cloud
Security and control
- Single sign-on
- Not covered in this record
- User provisioning
- Not covered in this record
- Role-based access
- Not covered in this record
- Administrator controls for AI
- Not covered in this record
- Audit logs
- Not covered in this record
- Certifications
- Not covered in this record
- Data retention
- Not covered in this record
- Data residency
- Not covered in this record
- Training on customer data
- Not covered in this record
AI features
As the vendor or firm describes them. Notes in small print record what the sources leave open.
Forecasting and scenario agent
Runs what-if models to estimate potential outcomes.
- Kind
- Established machine learning and rules
- Suggests or acts
- Suggests
- Who controls it
- Not stated
- Availability
- Not stated
Control: Not stated. Finance users run scenarios; who controls configuration is not stated.
SourceGenpact Finance Actions HubInsight and diagnostics agent
Answers ad hoc natural-language queries and generates structured reports using generative AI.
- Kind
- General-purpose assistant
- Suggests or acts
- Suggests
- Who controls it
- Not stated
- Availability
- Not stated
Control: Not stated. Users query it; who controls configuration is not stated.
SourceGenpact Finance Actions HubAnomaly detection agent
Scans financial data to flag patterns and unusual values and trigger alerts.
- Kind
- Established machine learning and rules
- Suggests or acts
- Suggests
- Who controls it
- Professional services firm
- Availability
- Not stated
Control: Not stated.
SourceGenpact Finance Actions HubInsight narrative agent
Automates narratives and commentary for standard reporting packs.
- Kind
- General-purpose assistant
- Suggests or acts
- Suggests
- Who controls it
- Professional services firm
- Availability
- Not stated
Control: Human-in-the-loop review stated for every step.
SourceGenpact Finance Actions Hub
Questions to ask
For this offering
- What data is the proprietary model fine-tuned on, does it include other clients' data, and which base model does it build on?
- How are generated narratives and commentary reviewed and evidenced before they reach a reporting pack or board material?
- Is our data used to train, tune or improve any model, rule or shared knowledge base, including one used for other clients, and where is it processed and retained?
- Which model providers and sub-processors sit behind the AI features, and how will you notify us when they change?
- Can you show the population, baseline and period behind the outcome figures you publish, and will you commit to any of them in the contract as service levels?
From the diligence question set
- DQ-31 Who, by name and role, will design, build, test and run the AI in our finance process, and what AI has each of those people personally put into production?
- DQ-32 How does your firm decide that someone is qualified to build or govern AI for financial reporting, and who outside the delivery team reviews model risk before anything goes live?
- DQ-33 When the engagement ends, what will we own and be able to run ourselves, and what will work only with your firm? On what terms are your accelerators licensed afterward, and what does it cost to leave a run service?
- DQ-34 What will this cost in years two to five, including run fees, platform licenses, AI usage fees and change requests, and how does that compare with the savings case once the internal oversight roles we keep are counted?
- DQ-35 Do you receive fees, rebates, resale margins or other incentives from the software you are recommending, and do you, or will you, audit our financial statements?
- DQ-36 What results from clients like us can you show that someone other than your firm has verified, and by what method?
Offering · AI agent
Genpact Record-to-Report Suite
Genpact presents the Record-to-Report Suite as agentic AI for the close, in three modules: Journal Entry, Reconciliation and Intercompany. SourceGenpact Record-to-Report Suite
Its agents ingest data from ERPs, sub-ledgers and bank files, apply self-learning rules to match transactions and classify exceptions, and route anything needing a human decision. SourceGenpact Record-to-Report Suite
Genpact says the Reconciliation agents verify evidence, score confidence and classify open items, with a full audit trail of every action. SourceGenpact Record-to-Report Suite
It positions the suite as a finance domain layer over the client's existing ERPs and cloud AI platforms. SourceGenpact Record-to-Report Suite
Where it fits
CoE analysis A close-focused agent product from an outsourcing firm, sitting between the ERP and the reconciliation or close tools a team already runs.
CoE analysis Its value depends on how the auto-matching and journal thresholds are set and who signs off on them, which the page does not specify.
What it does
- Offering type
- AI agentSourcesGenpact Record-to-Report SuiteGenpact Limited, Annual Report on Form 10-K for the fiscal year ended December 31, 2025
- Workflow areas
- SourceGenpact Record-to-Report Suite
- Close
- Reconciliation
- Customer size
- Not stated
AI
- Kind of AI
- SourcesGenpact Record-to-Report SuiteGenpact Limited, Annual Report on Form 10-K for the fiscal year ended December 31, 2025
- Agents and automated actions
- Established machine learning and rules
- Suggests or acts
- Suggests and actsSourceGenpact Record-to-Report Suite
- Availability
- Not stated
The product page offers demos; no availability, edition or regional terms.
- Models named
- Not stated
Control
- Who controls it
- SourceGenpact Record-to-Report Suite
- Professional services firm
- Customer finance team
Commercial
- Pricing model
- Not stated
- Public price
- Not stated
Evidence
- Independent outcomes
- None found in the sources read.
- Sources read
- Vendor-stated 5
- Independent market evidence 4
- Independently verified outcomes None read
- Pricing 1
- Known limitations None read
- Coverage
- Directory record
- Last verified
Delivery and lock-in
- Delivery model
- Not stated
- Run service offered
- Not stated
- Team qualifications published
- Not stated
Not stated in the sources read.
- IP and post-engagement license
- Genpact's 10-K says its solutions often include proprietary methodologies, software and reusable knowledge capital, and that "our agreements with our clients govern the ownership of such intellectual property." No terms for this product, or for use of its agents after an engagement ends, are published in the sources read. [C-5108]SourceGenpact Limited, Annual Report on Form 10-K for the fiscal year ended December 31, 2025
- Savings claims (firm-stated)
- SourceGenpact Record-to-Report Suite
Stated by the firm, not verified
up to 40% lower workload during close cycles
Marked on the page as a projected estimate. No baseline, population, period, or treatment of service fees is stated.
- Alliance partners
- SourceGenpact and Google Cloud Expand Alliance to Bring Agentic Solutions to the Office of the CFO (May 7, 2026)
- Google Cloud
Security and control
- Single sign-on
- Not covered in this record
- User provisioning
- Not covered in this record
- Role-based access
- Not covered in this record
- Administrator controls for AI
- Not covered in this record
- Audit logs
- Not covered in this record
- Certifications
- Not covered in this record
- Data retention
- Not covered in this record
- Data residency
- Not covered in this record
- Training on customer data
- Not covered in this record
AI features
As the vendor or firm describes them. Notes in small print record what the sources leave open.
Journal Entry module
Turns almost any input into a standardized journal template, learns from every entry and catches anomalies.
- Kind
- Agents and automated actions, Established machine learning and rules
- Suggests or acts
- Not stated
- Who controls it
- Professional services firm
- Availability
- Not stated
Control: Not stated.
SourceGenpact Record-to-Report SuiteReconciliation module
Self-learning agents verify evidence, score confidence, classify open items and route exceptions to people, with a full audit trail.
- Kind
- Agents and automated actions, Established machine learning and rules
- Suggests or acts
- Suggests and acts
- Who controls it
- Professional services firm
- Availability
- Not stated
Control: Not stated; routing rules for human decisions are not described.
SourceGenpact Record-to-Report SuiteIntercompany Assistant
Agents align multi-entity transactions, reconcile goods receipts against financial records and answer balance questions.
- Kind
- Agents and automated actions, General-purpose assistant
- Suggests or acts
- Suggests and acts
- Who controls it
- Professional services firm
- Availability
- Not stated
Control: Not stated.
SourceGenpact Record-to-Report Suite
Questions to ask
For this offering
- Which steps do agents or automation complete without a person approving them (for example posting entries, clearing reconciliations or releasing payments), and how are those thresholds set, changed and evidenced?
- Is the suite delivered as software we operate, as part of a Genpact-run service, or both, and who configures matching rules and auto-posting thresholds?
- Will you provide a SOC 1 Type 2 report for the outsourced finance processes, and does its scope include the AI platform and agents used to deliver them, including changes to models and prompts?
- Is our data used to train, tune or improve any model, rule or shared knowledge base, including one used for other clients, and where is it processed and retained?
- Which model providers and sub-processors sit behind the AI features, and how will you notify us when they change?
- Can you show the population, baseline and period behind the outcome figures you publish, and will you commit to any of them in the contract as service levels?
From the diligence question set
- DQ-31 Who, by name and role, will design, build, test and run the AI in our finance process, and what AI has each of those people personally put into production?
- DQ-32 How does your firm decide that someone is qualified to build or govern AI for financial reporting, and who outside the delivery team reviews model risk before anything goes live?
- DQ-33 When the engagement ends, what will we own and be able to run ourselves, and what will work only with your firm? On what terms are your accelerators licensed afterward, and what does it cost to leave a run service?
- DQ-34 What will this cost in years two to five, including run fees, platform licenses, AI usage fees and change requests, and how does that compare with the savings case once the internal oversight roles we keep are counted?
- DQ-35 Do you receive fees, rebates, resale margins or other incentives from the software you are recommending, and do you, or will you, audit our financial statements?
- DQ-36 What results from clients like us can you show that someone other than your firm has verified, and by what method?
