Supply chain / procurement software · Deep dive

Zip

Procurement orchestration for the enterprise — a single intake front door that routes every purchase request across finance, legal, IT, security and the ERP stack underneath — now valued at $2.2B and racing to convert an approvals wedge into an agentic AI suite before the suites it sits on bundle it away.

emerging

The question that decides it: Zip's control point is the intake layer — the form every employee touches before money moves — sitting on top of ERPs and P2P suites (SAP Ariba, Coupa, Oracle, NetSuite, Workday) that own the purchase orders, invoices, master data and the CFO's existing contract. Those owners are now bundling their own AI agents into suites already paid for, while Zip pushes downward into payments, sourcing and a 50-agent platform of its own. Can Zip convert the intake beachhead into actual suite displacement — customers ripping out Coupa/Ariba modules and paying Zip suite-level ACVs — faster than SAP and Thoma Bravo-owned Coupa ship good-enough agentic intake as a free checkbox on renewals?

My take ★★ Very interesting ★ Interesting − Maybe ✗ Not interesting

Snapshot

Zip sells the front door to enterprise spending: a single intake point where any employee requests any purchase, which then orchestrates the approval chain across finance, legal, security, IT and procurement, and pushes the result into whatever ERP or procure-to-pay system already runs underneath. Founded in 2020 by two ex-Airbnb product and engineering leaders, it raised roughly $371M — capped by a $190M Series D at $2.2B led by BOND (Oct 2024), billed as the largest procurement-tech financing in over two decades — and claims OpenAI, Snowflake, Discover, Sephora, Anthropic and Dollar Tree as customers. In 2025 it says it processed $355B in spend across 7M+ suppliers and crossed $6B in cumulative customer savings (company, Dec 2025). It invented the “intake-to-procure” category; the question now is whether a category built on sitting politely on top of Coupa and SAP survives those vendors bolting agents onto the systems they already own.

Founding story

Rujul Zaparde is a repeat founder with a scar. He left Harvard at 17 for FlightCar (YC W13), a marketplace letting air travelers rent out their parked cars, raised about $40M, scaled to 100+ employees across 12 airport locations — and shut it down in 2016, selling the technology to Mercedes-Benz R&D. He then did the deliberately unglamorous thing: product management at Airbnb. There he met Lu Cheng, a six-year Airbnb engineer who rose to head of engineering for the Experiences business. The Zip insight came from inside the machine: at a company of Airbnb’s size, buying a $10K software tool meant a maze of Slack messages, hidden approvers in security, legal, privacy and finance, and no visibility into where a request stood. Procurement software — Coupa, Ariba — was built for the procurement department, not for the 99% of employees who touch it twice a year. They founded Zip in 2020, went through YC’s Summer 2020 batch as a three-person company, and won CRV’s first institutional seed check by summer 2021 (Forbes, May 2023). Second-time-founder discipline shows in the sequencing: nail the wedge (intake), then expand down the stack.

How it works

Mechanically, Zip is a workflow graph compiler for corporate purchasing. An employee opens Zip (web or Slack) and describes what they want to buy — a SaaS renewal, an agency contract, a component order. Zip’s engine reads the request’s attributes — spend amount, department, vendor status, data-access profile, geography — and dynamically generates the approval chain that request needs: budget owner, then FP&A above a threshold, security review if the vendor touches customer data, legal for paper, IT for integration, privacy for GDPR scope. Steps that can run in parallel do; every stakeholder sees status in one place instead of a forwarded email chain.

The critical design choice is that Zip does not replace the systems of record — it writes into them. Pre-built connectors sync approved requests into ERPs (NetSuite, SAP, Oracle, Workday, Microsoft Dynamics), generate the PO in Coupa or Ariba where those remain, push contracts to Ironclad, tickets to Jira/ServiceNow, and vendor records into the ERP vendor master. Supplier onboarding — tax forms, banking details, security questionnaires, sanctions screening — runs inside the same flow. Since 2023 Zip also executes the money side itself where customers want it: PO generation, invoice capture and matching, and virtual vendor cards issued per-approved-purchase with spend controls, closing the classic gap where a card gets swiped before any approval exists. The 2025 agentic layer threads AI through each node: agents that validate purchase requests, pre-fill vendor risk assessments, extract contract terms, flag duplicate spend and benchmark price — 50+ of them, unveiled at Zip’s AI Summit (Jun 2025), with OpenAI, Canva, Wiz and Webflow as early testers (VentureBeat, Jun 2025).

Product and business overview

The platform now spans four named layers. Intake-to-procure is the original wedge: the universal request front door plus the no-code workflow builder and orchestration library of pre-built approval blueprints. Procure-to-pay (launched May 2023) adds POs, invoicing, matching and payments, letting mid-market customers run Zip as their only procurement system. Vendor cards (Jun 2023) are virtual cards bound to approved requests. Zip AI / agentic orchestration (2024-25) is the monetizable AI tier — agents for intake triage, vendor risk, contract review, and from 2026, accounting and contract orchestration. Zip Premier packages enterprise-grade orchestration, and a 2026 Enterprise Transformation Office signals the push upmarket into global rollouts. Customers skew tech-forward (OpenAI, Snowflake, Coinbase, Canva, Databricks, Reddit, Instacart, Toast) but increasingly include boring giants — Discover, Prudential, Northwestern Mutual, Dollar Tree, Sprouts, Mars — which is where the durable ACV lives.

Business model and pricing

Enterprise SaaS with platform pricing, not per-seat: fees scale with company size or annual spend volume, deliberately so every employee can touch intake without a seat-count tax. Sacra pegs the average contract around $82K a year; no price list is published, and third-party buyers’ guides describe quotes starting in the low five figures for mid-market and multiples of that for enterprise (2025-26). Vendor cards add an interchange economic, and the AI agent tier is the upsell vector. Gross margins are standard high-80s SaaS (Sacra estimate). The ROI pitch is cycle time and leakage: IDC’s Zip-commissioned study claims 25% procurement-productivity gains, and the company claims $6B cumulative customer savings and 10M days of cycle time eliminated (Dec 2025) — company-fed numbers, but the kind CFOs repeat in renewal meetings.

Traction over time

Revenue is the soft spot in the public record: GetLatka’s $193M ARR (2025) estimate against $18.2M (2023) implies a 10x in two years — directionally plausible for the logo velocity but unaudited and probably generous; Sacra’s $82K average contract times “hundreds of enterprises” suggests a lower base with fast expansion. Customer count more than doubled between May 2022 and May 2023 (TechCrunch), and by 2025 the roster read like a who’s-who of AI-era tech plus Fortune 500 ballast. No 2025-26 equity round has been disclosed as of August 2026 — at this burn profile, either efficiency is real or a Series E is coming.

Market analysis

Procurement software proper is roughly a $9-10B market in 2025 growing ~10% a year (GM Insights, Polaris, 2025) — but that undersells the motion. Gartner forecasts supply-chain software spend with agentic AI growing from under $2B in 2025 to $53B by 2030, with orchestration platforms taking a large slice (Gartner, Apr 2026), and Gartner now maintains a dedicated procurement-orchestration category in Peer Insights. The structural forces: SaaS sprawl made every employee a buyer; security/privacy reviews made every purchase cross-functional; and agentic AI is collapsing the labor cost of the workflows suites never automated well. The counterforce is consolidation gravity — CFOs in a cost-cutting cycle prefer fewer vendors, and the biggest procurement budgets already sit inside SAP and Oracle contracts. Zip’s own Gartner recognition cuts both ways: being named a Visionary in the 2026 Source-to-Pay Suites Magic Quadrant (Jan 2026) validates the company — by reclassifying it as a suite vendor competing with Coupa and SAP directly, orchestration-layer neutrality be damned.

Competitive intel

See the competitor table. Three fronts. Above, the suites: Coupa (Thoma Bravo, ~$8B take-private, MQ Leader) and SAP Ariba own the transactional record and the renewal leverage; their agent pushes (Navi, Joule) aim to make an external intake layer redundant, and every Zip deployment that merely front-ends them is a standing invitation to bundle. Beside, the orchestration pack: Oro Labs (ex-Ariba founders, ~$60M), Levelpath ($100M, Battery-led B in Jun 2025, founders who already sold Scout RFP to Workday), Omnea (Accel-backed, EMEA), Tonkean (agentic no-code) — all fighting for the same category Zip named, mostly at lower price points. Zip outguns all of them on capital, logos and category ownership. Below, spend-management platforms (Ramp, Brex, Airbase) creep up from cards and expenses into procurement for the mid-market. Zip’s edge is the compounding workflow graph and its integration surface; its exposure is that it owns neither the ERP record below nor — yet — enough of the payment flow to be unremovable.

History and evolution

• 2020 — Founded by Zaparde and Cheng; YC Summer 2020 batch; three people.
• Summer 2021 — First institutional seed led by CRV; $25M Series A follows (CRV).
• May 2022 — $43M Series B at $1.2B led by YC Continuity with Tiger Global; 130+ employees; “concierge for procurement” positioning (TechCrunch).
• May 2023 — $100M Series C at $1.5B (YC, CRV, Tiger); launches procure-to-pay, becoming “intake-to-pay.”
• Jun 2023 — Vendor Cards launch; first user conference Zip Forward; early GenAI features ship through 2023.
• Oct 2024 — $190M Series D at $2.2B led by BOND with DST, Adams Street, Alkeon; $4.4B cumulative customer savings claimed.
• Jun 2025 — Zip AI Summit, Brooklyn: 50+ purpose-built agents; “agentic procurement orchestration” category declared; OpenAI, Canva, Wiz, Webflow in beta.
• Dec 2025 — Claims $6B cumulative savings, $355B spend processed in 2025, 26M approvals, 10M AI insights.
• Jan-mid 2026 — Named Visionary in Gartner’s Source-to-Pay Suites MQ; rebrand (and move to zip.com); London office; new Chief Sales Officer; agents for accounting and contract orchestration; Enterprise Transformation Office. No stumble on the scale of a lawsuit or layoff is in the public record — the notable absence is any disclosed revenue figure, ever.

What people say

The case for. G2 reviewers rate Zip 4.6-4.8 stars (2025-26), and the recurring theme is that requesters actually use it voluntarily — a consumer-grade interface and configurable workflows that ended the where-is-my-request email chain. Procurement leaders describe deployment in weeks on top of stacks that took years to install. Glassdoor (Zip HQ, ~111 reviews) skews positive: product-and-engineering-driven culture, sales reps clearing 300-400% of target, a sticky product “beloved by users.” Investors’ revealed preference — BOND, DST and Alkeon paying a 46% markup in a brutal 2024 for growth rounds — is its own review.

The complaints. The most consistent customer gripe is reporting and analytics depth: fine for tracking requests, thin for the spend-cube analysis CPOs expect from a suite (G2 themes; ProcureDesk, 2026). Pricing opacity irritates mid-market buyers — every tier is “contact sales.” Rivals argue Zip adds a layer, and a bill, on top of tools you already own; where a customer runs full Coupa, the double-pay question surfaces in every renewal. Glassdoor’s negative tail cites favoritism, thin planning, and the grind of selling consultative procurement deals with young AEs. And every headline traction number — savings, spend processed, hours saved — is company-supplied and unaudited, with third-party ARR estimates diverging wildly. The uncomfortable version: Zip has named a category, but not yet proven publicly that the category’s economics belong to it rather than to the suites beneath it.

Outlook: the open question

The whole case turns on whether intake is a control point or a feature. The bull case requires: enterprises keep treating the employee-facing layer as the strategic one — the place where policy, AI agents and spend data converge before any transaction exists; Zip’s procure-to-pay and payments volume grows until removing Zip means re-implementing procurement, not deleting a form; agent adoption converts into a second ACV line; and displacement deals (Zip replacing Coupa/Ariba modules outright, as its Visionary MQ slot implies) show up as reference customers in 2026-27. If GetLatka’s growth curve is even half right, a Series E or IPO file confirms it. The bear case requires: SAP and Coupa ship credible agentic intake bundled at renewal, procurement consolidates back into the ERP contract during a spending squeeze, the orchestration pack commoditizes intake pricing from below, and Zip is left owning a UX layer with $82K contracts against suite vendors monetizing the same workflows at 10x the ACV — the Slack-vs-Teams shape, where naming the category doesn’t mean keeping it. Watch three tells: whether Zip ever discloses audited revenue or files, whether announced wins start saying “replaced” rather than “integrated with,” and whether Coupa’s 2026-27 releases make an external front door look like a tax.

How a challenger would attack it

Zip’s exploitable seam is that it charges suite-level money for a layer it doesn’t own on either side. It sits above ERPs it must write into and below employees it delights, with $82K average contracts, “contact sales” pricing at every tier, and a customer base that still pays Coupa or Ariba underneath — the double-pay question already surfaces at every renewal. A challenger attacks with transparent, self-serve pricing and an agent-native intake that undercuts the orchestration premium: Zip’s own 50-agent pivot concedes that the workflow graph — its original moat — is now cheap to generate, and an AI-from-scratch entrant (the Levelpath pitch) carries no pre-agentic architecture to defend. The second vector is the analytics gap: G2’s most consistent complaint is thin reporting and no real spend-cube, so a rival that leads with CPO-grade analytics inverts the wedge — win the spend-data layer first, give intake away. Third, come up from payments the way Ramp and Brex already are: Zip’s vendor-card interchange is nascent, and a spend platform that already moves the money can bundle intake for free, making Zip the removable form on top. The counter Zip lacks is lock-in: until removing it means re-implementing procurement, it is a UX layer with enterprise pricing.

Same playbook, new buyer

Intake-to-approval orchestration is a pattern, not a procurement feature — any cross-functional request that dies in a Slack-and-email maze is the same product. The nearest ports: headcount requisitions, IT access grants, marketing and legal work intake — Tonkean already gestures at this with its “AI Front Door,” but no one has Zip’s execution at it. Within procurement, the sharper shifts are down-market and offshore. Zip prices for enterprises and staffs an Enterprise Transformation Office; the mid-market that finds every tier “contact sales” is open to a transparent, self-serve orchestrator at a tenth the ACV — and Zip can’t chase it without repricing the base that supports a $2.2B valuation. Geographically, Omnea’s EMEA wedge shows the shape: European buyers want local data posture and SAP fluency, and Zip’s London office (2026) is a late arrival, not a beachhead. The most durable variant is vertical: regulated industries — the Discover/Prudential/Northwestern Mutual cohort Zip is only starting to win — need compliance-native workflows (vendor risk, sanctions, model governance) that a horizontal platform retrofits; a vertical orchestrator ships them as the product. Zip’s roadmap is consumed by suite displacement upward; every one of these lateral moves is off its critical path.

Sources and further reading

• BusinessWire — Zip secures $190M Series D at $2.2B, led by BOND (Oct 21, 2024)
• TechCrunch — “Procurement platform Zip raises $100M at a $1.5 billion valuation” (May 16, 2023)
• Forbes — “Procurement Startup Zip Announces $100 Million In New Funding” (May 15, 2023)
• TechCrunch — “Zip nabs $43M for its growing ‘concierge for procurement’” (May 25, 2022)
• VentureBeat — “Zip debuts 50 AI agents to kill procurement inefficiencies — OpenAI is already on board” (Jun 2025)
• BusinessWire — “Zip Surpasses $6 Billion in Customer Savings as Agentic Procurement Orchestration Transforms Enterprise Purchasing” (Dec 2, 2025)
• Contrary Research — Zip business breakdown and founding story (updated 2024-25)
• Sacra — Zip valuation, funding and business model (2024-25)
• BusinessWire — Zip launches Vendor Cards (Jun 29, 2023)
• PR Newswire — Coupa named a Leader in the 2026 Gartner Magic Quadrant for Source-to-Pay Suites (Zip named among gainers) (Jan 23, 2026)