Skip to content

Zip

Overview: AI-powered procurement orchestration platform (zip.com, formerly ziphq.com) that pioneered the “intake-to-procure” category and, since May 2023, markets a unified intake-to-pay suite. Raised a $190M Series D led by BOND at a $2.2B valuation (October 2024). Reports $6B+ in customer savings across hundreds of billions in processed spend and 7M+ suppliers (December 2025). Customers include OpenAI, Snowflake, Discover, Reddit, Sephora, Dollar Tree and Lyft. At Zip Forward (September 16, 2026) it announced Sourcing and PO Superagents, expanded AI Risk Orchestration (TPRM), Accrual Automation and an integrations Marketplace.


1. Deployment models Cloud-only multi-tenant SaaS. No on-premise or hybrid options. Implementations typically take weeks, per reviews.

2. Data model — unified or modular/siloed Unified platform with a single request record carried from intake through approval, PO, invoice and payment. Modules (Intake-to-Procure, Procure-to-Pay, Sourcing, Contract Orchestration, Vendor Management, Supplier Onboarding, Risk Orchestration, Global Payments) are sold separately but share one data model. Often deployed as a layer over an ERP or existing S2P suite.

3. Technology stack No-code workflow engine with dynamic intake forms. AI layer consists of 50+ live AI agents and, since June 2026, “Superagents” (initially covering procurement, contracts, AP, configuration and intake) governed by the same roles, permissions and audit controls as employees. Underlying LLM providers and architecture details are not publicly disclosed.

4. Uptime/SLA guarantees SOC 2 Type II audited and ISO/IEC 27001:2022 certified; TLS 1.2+ in transit and AES-256 at rest. Specific uptime SLA figures are not publicly disclosed and are contract-defined.


5. AI/ML capabilities — native vs bolted on Native. AI is embedded directly in the orchestration workflows (intake, contract review, invoice coding, vendor research, approvals) rather than offered as a separate product.

6. Generative AI support Yes — AI Procurement Concierge for conversational intake, AI contract review against playbooks, invoice data extraction and coding, and vendor research. Zip MCP (beta June 2026, GA summer 2026) exposes Zip data and actions to ChatGPT, Claude and other AI assistants under Zip’s permission model.

7. Agentic/autonomous workflow layer Yes — core strategy. Superagents reason and act inside workflows with deterministic logic for high-impact steps. The PO Superagent drafts requests, recommends PO closures and suggests change orders; the Sourcing Superagent coordinates and analyzes sourcing events while the buyer keeps award decisions.

8. AI in supplier/vendor evaluation and scoring AI Risk Orchestration (expanded to full TPRM in September 2026) triages requests, decides whether an assessment is needed, pre-fills questionnaires, scores suppliers against custom risk frameworks and automates periodic reassessments with third-party monitoring data.

9. Predictive analytics Spend insights and savings-opportunity detection, plus price benchmarking via data partners (NPI, SpendHound, Tropic, Beroe, Brightfield, TEEM). Accrual Automation generates ERP-ready journal entries from PO data. Supply-disruption prediction is not publicly documented.

10. AI feature update frequency and public roadmap Very high cadence, with two major annual launch events (AI Summit in June, Zip Forward in September) plus continuous releases. No detailed public roadmap is published beyond event announcements.


11. ERP and financial system integrations Pre-built bi-directional connectors to NetSuite (Built for NetSuite certified), SAP S/4HANA, Oracle Fusion, Workday Financials, Microsoft Business Central and Sage Intacct, plus SAP Ariba and Coupa. 100+ integrations overall.

12. Open API Yes — REST APIs and flat-file imports to create, read and update vendors, requests, POs and invoices. Zip MCP adds an AI-assistant interface. The new Marketplace launched with 20+ integration templates (ERP, P2P, CLM, GRC) and an AI-guided setup wizard.

13. SSO and identity/auth standards SSO with any SAML-based IdP, SCIM user provisioning/deprovisioning, and granular role-based permissions by role, department and group.

14. Pre-built connectors to supplier networks or third-party data providers No proprietary supplier network comparable to Coupa or Ariba; suppliers interact via a vendor portal. Connectors to CLM (e.g., Ironclad), GRC/risk intelligence, and benchmarking data partners. Global Payments supports 140+ countries and 40+ currencies; the underlying payments partner is not publicly detailed.


Tender & Procurement-Specific Functionality

Section titled “Tender & Procurement-Specific Functionality”

15. RFx types supported Zip Sourcing launches RFx events (RFI/RFP/RFQ) directly from intake requests, centralizes supplier responses, supports vendor scoring and simplifies award decisions. eAuctions, reverse auctions and advanced optimization scenarios are not publicly documented, so depth appears lighter than dedicated sourcing suites.

16. Compliance and audit trail Full audit trail of requests, approvals, AI agent actions and RFx events. Policy-driven approval routing across procurement, legal, security, IT and finance. SOC 2 Type II and ISO 27001:2022. No public-sector tendering compliance features are publicly documented.

17. Configurability — custom workflows, scoring criteria, approval hierarchies High — no-code workflow builder and dynamic intake allow procurement teams to configure approval chains and forms without IT. Reviewers note some limits on admin settings and dashboard configurability.


18. Known gaps and limitations

  • Reporting and analytics are the most frequently cited weakness on G2 and Capterra (limited custom reports, reliance on CSV exports).
  • Integrations are sometimes described as not fully “out of the box,” requiring internal technical effort; periodic disruptions reported with third-party tools (e.g., Ironclad).
  • P2P/AP and sourcing are younger than the core intake product — some reviewers find invoicing, change-order visibility and CLM depth lacking vs. mature S2P suites.
  • Sourcing depth (eAuctions, multi-round/multi-attribute optimization) is not publicly demonstrated.
  • Not designed for direct materials, BoM management or engineering-led/capital procurement; strongest in indirect spend.
  • Many headline AI capabilities (Superagents, MCP, Sourcing/PO Superagents, TPRM expansion) are 2026 releases — maturity at scale is unproven.
  • Limited support coverage outside Western time zones reported by APAC users.
  • Pricing is not public; module-based packaging can raise total cost.