Ramp API Integration Research

Date: 2026-05-13 Product: Ramp (ramp.com) -- Spend Management Platform, New York City (US) Purpose: Evaluate Ramp's API capabilities for Orcha integration -- specifically for a German customer evaluating Ramp as a Pleo alternative for employee credit cards while continuing to use Orcha for AP invoices and DATEV for accounting.

Important Context: What Ramp Actually Is

Ramp is NOT an ERP or accounting system. Like Pleo, Payhawk, and Moss, it is a spend management platform that handles:

• Corporate credit cards (physical + virtual + AI-agent cards)
• Accounts payable / bill pay (Bill Pay module)
• Employee reimbursements
• Pre-accounting (AI-powered expense categorization)
• Procure-to-pay (POs, intake requests)
• Business travel

Ramp sits upstream of the ERP -- it captures, codes, and approves expenses, then exports pre-accounted data to the customer's actual accounting/ERP system (NetSuite, Sage Intacct, QuickBooks, Xero, Workday, Business Central, etc.).

Ramp is the highest-profile US fintech in this category (USD ~32B valuation as of 2026), known for API maturity, AI tooling, and aggressive "free tier funded by interchange" pricing.

Implication for Orcha: Ramp is a peer (spend management) layer, not a target system to push invoices into. For the customer scenario in scope (German company, Orcha = AP, DATEV = accounting, Ramp considered for cards), the integration question is narrower:

1. Can Orcha read card transaction data from Ramp to provide unified spend analytics alongside its invoice corpus?
2. Can Orcha write reference data (chart of accounts, cost centers, tax codes, vendors) into Ramp so the customer's card-coding stays consistent with the AP coding Orcha already drives?

However, the gating question is whether Ramp can service the customer at all. See section 1.

1. Summary -- Germany Availability First, Then Capability Matrix

Germany / EU Availability -- HEADLINE FINDING

Ramp is in the middle of an EU market entry that begins summer 2026 (i.e. right now / imminent). German-headquartered companies CANNOT directly sign up as Ramp customers today; the company is operating a waitlist with onboarding "beginning this summer". (prnewswire.com, ramp.com/blog/ramp-is-launching-in-europe)

The relevant facts:

Bottom line: Ramp's product is technically the most API-mature in this evaluation set, but the company is 6-12 weeks away from being a viable option for a German-HQ customer as of this writing (May 2026), and even then it will arrive without a DATEV integration, which is a hard requirement for any serious German SMB / Mittelstand customer using DATEV for accounting. Ramp's go-to-market in Germany will require either DATEV building support themselves, Ramp building it for v1 (no signal in public roadmap), or a third-party connector. This is a disqualifying gap for a customer who uses DATEV today and won't migrate.

The rest of this document continues for comparison reference and to inform Orcha's roadmap thinking once Ramp does land in Germany.

Capability Matrix

Overall assessment: Ramp has the most API-comprehensive and modern developer surface of any spend tool in this evaluation set (Pleo, Payhawk, Moss, Spendesk, Yokoy, Circula, Finway, Qonto). Webhooks are first-class with explicit transaction events. Reference data has full CRUD with batch endpoints (up to 500 items per upload). Bills CAN be created via API -- a meaningful differentiator vs Pleo, Payhawk, Moss. The Developer API is FREE on all plans (interchange-funded). However, two showstoppers: (1) German-HQ companies cannot directly sign up until summer 2026 onboarding rolls out, and (2) there is no DATEV integration. For a German company that uses DATEV and Orcha today, Ramp is not yet a real option.

2. API Landscape

API Layers

Authentication

Ramp does not support simple API-key static-token auth (unlike Payhawk's static Bearer model). Everything is OAuth, but client_credentials flow is essentially equivalent for server-to-server purposes.

Rate Limits

For comparison: Pleo = 10 req/s, Payhawk = 15 req/s, Ramp = ~20 req/s. Ramp is the highest in this peer set on raw throughput, and the per-IP application (rather than per-account) is unusual and slightly more permissive for multi-account orchestrators.

API Categories (Comprehensive)

Confirmed via the LLM-friendly plaintext API reference (llms-api.txt) and OpenAPI spec:

Recent (2025-2026) Additions

• Custom Records -- configurable native + custom tables, exposed in API
• Bills drafts -- separate draft endpoint set
• Deferred / async card lifecycle -- task-based pattern with polling
• MCP servers (3 of them) -- first-class AI-agent surface

Deprecation / Versioning

API is on v1 (/developer/v1). Two notable deprecations:

• GET /accounting/connection (singular) deprecated in favor of GET /accounting/all-connections + GET /accounting/connection/{id}
• Legacy monetary-value representation deprecated; canonical CurrencyAmount (integer in smallest denomination + ISO 4217 currency code) is the standard

3. Write Capability Verification

Confidence: HIGH. Ramp's write surface is the most comprehensive in the spend-management peer set evaluated by Orcha to date.

Confirmed Write Endpoints / Operations

Key Capabilities vs Peers

• Batch-of-500 uploads for vendors, accounts, field options, tax rates -- significantly more efficient than peer-row-by-row APIs
• Idempotency keys required on sync endpoints -- prevents duplicate sync events
• Async / deferred task pattern for card operations with status polling -- handles real-world card-network latency
• Native POST /bills -- you can push an invoice into Ramp programmatically (Pleo, Payhawk, Moss, Spendesk all lack this)

Limitations

• No DATEV integration -- not via API, not via CSV templates, not via partner connector
• Static API keys (single-secret bearer tokens) are not offered; everything is OAuth -- slightly more setup than Payhawk's API-key model
• Card transactions themselves are not writeable (they originate from the card network); reimbursements and bills are
• Approval rules / spend programs are configurable via API but the approve/reject action on a specific in-flight request isn't a documented public endpoint

4. iPaaS & Middleware Findings

Workato is the most viable iPaaS path. Outside of that, the story is that Ramp's API is so straightforward and well-documented that the direct-API approach is the default -- there is essentially no Zapier or Make ecosystem around it. Combined with Ramp's three MCP servers, the official messaging is "use AI agents / MCP, or talk directly to the API; iPaaS isn't the primary channel."

5. Alternative Channels

DATEV Integration (Not Supported)

Ramp has no DATEV integration of any kind. Confirmed by:

• ramp.com/accounting-automation-software (no DATEV in 30+ supported ERP list)
• ramp.com/integrations (DATEV not listed)
• Ramp Help Center "Overview of Ramp Accounting" (not accessible from public web)
• No DATEV mention in Ramp's EU launch announcement materials
• No ZUGFeRD or XRechnung format support documented

This is the single most consequential gap for a German DATEV-using customer. The closest workaround would be:

1. Build a custom integration via Ramp's "Accounting Connections" API (POST /accounting/connection, then implement the two-way sync) -- effectively writing a DATEV bridge yourself
2. Use Workato or a similar iPaaS to bridge Ramp <-> DATEV via DATEV Cloud Services / Rechnungsdatenservice
3. Use Ramp's CSV export and manually load into DATEV
4. Wait for Ramp to ship a native DATEV integration (no public roadmap)

None of these match the out-of-the-box experience that Payhawk, Pleo, and most German-native tools (Circula, Yokoy, Moss, Finway) provide for DATEV today.

Native ERP/Accounting Integrations (Confirmed List)

HRIS Integrations

ADP Workforce Now (native connector), plus generic ERP/HR sync via API.

Webhooks

This is the strongest webhook posture in the peer set:

• Pleo: Svix-powered webhooks but expense data gated by Export Jobs
• Payhawk: real-time expense.* events, RSA-signed
• Ramp: real-time transactions.* events including pre-clearing authorized, HMAC-SHA256

CSV Exports

• Custom CSV export builder for any ERP / accounting system not covered by native integrations
• Supplier and CoA CSV import / export
• Manual rather than scheduled

6. Licensing & Access Requirements

API Access

Ramp Developer API is free on all plans. Confirmed via docs.ramp.com/llms-guides/getting-started.txt: "Sign up at ramp.com to provision an account, or request a sandbox to start building without one. ... Company -> Developer -> Create New App."

There is no API-gated tier and no per-call fee. The 200 req / 10s rate limit applies uniformly. Higher limits available on request via support ticket.

For comparison: Pleo gates Open API to Advanced ($109/mo+); Payhawk includes it on all paying plans; Ramp includes it on the free tier.

Plan Pricing

Ramp's commercial model is unique among the peer set:

For a German company with ~30 cardholders:

• Ramp Free or Plus = $0 - $450/mo
• Pleo Advanced = ~$340-540/mo
• Payhawk Cards & Expenses = ~$449/mo flat

Ramp is the cheapest by a wide margin -- but only if you don't need the multi-currency / international features (which gate to Plus or Enterprise).

Customer Setup Checklist

1. Today (May 2026): German customer cannot directly sign up. Must join the ramp.com/europe waitlist.
2. Once direct EU onboarding ships (summer 2026):
   • Sign up via standard onboarding
   • Verify KYB / KYC for EU/UK entity
   • Choose plan (Free works for cards + Developer API)
3. Admin navigates to Company -> Developer in Ramp dashboard
4. Click Create New App, name it
5. Under Grant types, select Client Credentials
6. Under Scopes, enable: transactions:read, accounting:read, accounting:write, users:read, bills:read, bills:write (start narrow)
7. Store Client ID + Client Secret securely
8. Generate webhook secret + subscribe to transactions.cleared, transactions.authorized, etc.
9. Implement HMAC-SHA256 signature verification using webhook secret
10. Note: no DATEV path -- if accounting reference data is needed, Orcha would push via the Accounting / Vendors / Tax / Field APIs directly

7. Orcha-Specific Deep Dive

Critical Question Comparison: Ramp vs Pleo on the Customer's Pain Points

Critical Question: What Role Does Ramp Play Relative to Orcha?

Ramp and Orcha have overlapping functionality in the Bill Pay module of Ramp -- but the customer scenario in scope explicitly excludes Ramp's Bill Pay module (the customer keeps Orcha for AP).

So with Bill Pay off, Ramp's responsibility is cards + reimbursements; Orcha's is AP invoices. But the two streams cannot converge in DATEV via Ramp -- Ramp will not export there. The customer would have to:

• export Ramp data via CSV manually, or
• export Ramp data via the Accounting Connections API to a custom DATEV bridge, or
• use Workato to bridge Ramp -> DATEV Cloud Services

This is fundamentally worse than the Pleo or Payhawk options for a DATEV customer.

Integration Scenarios for the Customer

Scenario A: Orcha reads card transaction data FROM Ramp for unified spend analytics (PRIMARY USE CASE)

• Feasibility: HIGH (technically), if customer is on Ramp
• Subscribe webhook to transactions.authorized (for pre-clearing visibility) and transactions.cleared (settled) -> Orcha endpoint
• On event, GET /developer/v1/transactions/{id} for full detail
• Optional: GET /transactions?sync_status=SYNC_READY for incremental polling sweep against missed webhooks
• Pull receipt files via Receipts API (GET /receipts/{id}) -- TTL verification required before deciding whether to mirror to Orcha S3
• HMAC-SHA256 signature verification on webhooks

Scenario B: Orcha writes reference data INTO Ramp for consistent coding (SECONDARY USE CASE)

• Feasibility: HIGHEST in the peer set (full CRUD + batch endpoints), if customer is on Ramp
• Push chart of accounts via POST /accounting/accounts (up to 500 in one call)
• Push cost centers / projects via POST /accounting/fields + POST /accounting/field-options
• Push tax codes via POST /accounting/tax/code/options
• Push vendors via POST /accounting/vendors (up to 500 in one call)
• Use POST /accounting/syncs with idempotency keys to mark export complete
• Note: this would not be a substitute for the missing DATEV integration -- it's just keeping Ramp's card-coding aligned with Orcha's AP-coding. The customer still needs to manually bridge Ramp -> DATEV.

Scenario C: Customer migrates AP from Orcha to Ramp's Bill Pay module

• NOT relevant for this customer scenario. Customer keeps Orcha for AP.

Scenario D: Orcha pushes invoices into Ramp

• POSSIBLE via API -- POST /bills exists. Unique vs Pleo/Payhawk/Moss.
• But irrelevant for this customer scenario.

Scenario E: Bridge Ramp transactions to DATEV via Orcha

• Feasibility: MEDIUM, would require new build
• Orcha reads Ramp transactions via webhook + GET, transforms them to DATEV booking format, and pushes via Orcha's existing DATEV integration
• Effectively turning Orcha into Ramp's missing DATEV connector
• Requires careful design: how to handle Ramp's pre-accounting vs DATEV's account/cost-center semantics, FX, BU tax codes, KOST1/KOST2, etc.
• Real engineering effort, but feasible

Key API Differences vs Peers

On pure API capability, Ramp wins on almost every axis. On real-world availability and German operational requirements, Ramp loses on the only two axes that matter for this customer.

Receipt URL Longevity -- Action Required

As with Payhawk, the single most important unverified question is the receipt URL TTL policy. Ramp has a first-class Receipts API (GET /receipts, GET /receipts/{id}) but the URL TTL policy is not stated in public documentation (the Receipts guide page is JS-rendered SPA, inaccessible via WebFetch).

Recommendation: same probe pattern as Payhawk -- before committing engineering effort, run an authenticated probe against a Ramp sandbox. Fetch a receipt, capture the URL, re-fetch at T+1h, T+6h, T+24h, T+72h, T+7d. Industry pattern for fintech is signed S3/CloudFront URLs with 1h-7d TTL; if TTL is short, mirror to Orcha storage at webhook time.

Orcha Integration Capability Summary

8. Recommended Path Forward

Assessment: Ramp is the Best Spend-Tool API in the Market, But Not a Real Option for This Customer

Ramp's developer surface is materially ahead of Pleo, Payhawk, Moss, Spendesk, Yokoy, and every other tool in the spend-management peer set on:

• Webhook richness (pre-clearing authorization events, multi-stage transaction lifecycle)
• Reference-data write API (batch-500 uploads, full CRUD on vendors / accounts / fields / tax)
• API access pricing (free on all plans)
• AI / MCP integration surface (3 dedicated MCP servers)
• Bill push capability (POST /bills -- unique in the set)

But for this specific customer scenario -- German company, DATEV for accounting, Orcha for AP, evaluating Pleo replacement for cards -- Ramp fails on the two non-API attributes that matter most:

1. Cannot onboard a German-HQ customer today (May 2026), only via summer-2026 EU waitlist
2. No DATEV integration of any kind, and no public roadmap signal that one is coming

These two gaps are larger and more concrete than any API advantage Ramp offers.

Recommended Architecture (If Ramp Becomes Viable in Q3-Q4 2026)

Assuming Ramp ships German direct onboarding in summer 2026 AND the customer is willing to either (a) accept manual / Workato-bridge DATEV export, or (b) wait for native DATEV, the integration would look identical to the Payhawk pattern:

[Card transactions]            [Vendor invoices]
       |                              |
       v                              v
   Ramp  --- real-time webhook --->  Orcha (read-side: unified spend analytics)
       |                              |
   *** GAP ***                        |
   No DATEV path                      v
   (manual / iPaaS bridge needed) -> DATEV
                                      ^
                                      |
                                  Orcha (AP write directly to DATEV)

Read side (Ramp -> Orcha):

1. Subscribe webhooks to transactions.authorized + transactions.cleared
2. On event, GET /developer/v1/transactions/{id} for full detail
3. (Optional) periodic GET /transactions?sync_status=SYNC_READY reconciliation sweep
4. Pull receipts via GET /receipts/{id} (verify TTL first; mirror to Orcha storage if short)
5. HMAC-SHA256 signature verify on webhooks

Write side (Orcha -> Ramp):

1. Push chart of accounts via POST /accounting/accounts (batch up to 500)
2. Push cost centers / projects via Accounting Fields API
3. Push tax codes via Tax Code Options API
4. Push vendors via POST /accounting/vendors (batch up to 500)
5. Mark exports complete via POST /accounting/syncs with idempotency keys

Missing piece: Ramp -> DATEV bridge. Either:

• Orcha builds it (reads Ramp transactions, transforms to DATEV booking format, pushes via existing Orcha DATEV integration) -- meaningful engineering effort
• Customer manually exports Ramp CSV -> DATEV
• Workato or similar iPaaS handles the Ramp <-> DATEV mapping (no off-the-shelf flow exists today)

When Ramp Integration Would Make Sense

1. Multi-country group with US HQ and German subsidiary: Already viable today. US parent is Ramp customer, German entity issues EUR cards via Stripe-Ireland. Orcha reads Ramp data for the German entity, pushes invoices to DATEV directly.
2. German customer post-summer-2026 launch, willing to absorb DATEV gap: Wait for direct onboarding, accept that Ramp -> DATEV bridge will be manual or Orcha-built.
3. German customer who doesn't use DATEV: Theoretically possible but rare -- DATEV's 90%+ market share among German tax advisors makes a non-DATEV German SMB unusual.

When Ramp Integration Does NOT Make Sense

1. Right-now decision (May 2026) for a German-HQ Pleo replacement: Ramp is not yet eligible for direct onboarding. Payhawk is the correct comparison target (German market, DATEV native, free API, real-time webhooks).
2. DATEV-using customer with no appetite for custom bridging: Ramp's missing DATEV integration is fatal.

What NOT to Build

• Don't build DATEV-via-Ramp bridge speculatively. Wait for either Ramp to ship it natively or for a real customer to require it.
• Don't migrate AP from Orcha to Ramp's Bill Pay module for this customer scenario -- Orcha is the AP authority and Ramp Bill Pay doesn't export to DATEV either.

Architecture Decision

Pre-Implementation Verification Needed

If Ramp becomes a real option later:

1. Probe receipt URL TTL via authenticated sandbox
2. Confirm full webhook event catalog -- the public docs list ~12 event types but the OpenAPI spec is the source of truth
3. Confirm transaction field schema -- specifically MCC, raw descriptor, approver chain access pattern, FX detail, card virtual/physical flag location
4. Test multi-currency settlement in EUR for German entity -- behavior post-Billhop integration
5. Verify if DATEV makes the integration list in 2026 EU rollout -- this is the make-or-break feature

Recommendation Summary

For the customer scenario (German company, DATEV + Orcha + Pleo replacement for cards): NO, do not pursue Ramp. Recommend Payhawk instead. Payhawk has the same realtime + free-API advantages that drove the Pleo evaluation, plus a best-in-class DATEV native integration, and is available to German customers today.

Ramp is materially better than Pleo, Payhawk, and the rest of the peer set on every pure API axis, BUT loses on the two non-negotiable customer requirements:

1. German direct onboarding: not until summer 2026 (waitlist only today)
2. DATEV integration: does not exist, no public roadmap

Revisit Ramp in Q4 2026 for a different German customer who either does not use DATEV or is willing to accept manual / iPaaS bridging for DATEV export. Until then, this research is filed as a reference baseline for the strongest possible spend-tool API in the market.

9. Sources

Official Documentation

• Ramp Developer API home
• Ramp API LLM index (full list of guide slugs)
• Ramp API reference (plain text) -- comprehensive endpoint catalogue
• Ramp API guides combined (plain text)
• Ramp OpenAPI specification
• Quickstart / Getting Started
• Authorization -- OAuth 2.0 flows, scopes, 10-day token TTL
• Webhooks -- event types, HMAC-SHA256 signature, retry policy
• Rate Limiting -- 200 req / 10s per IP
• Cards
• Monetary Values -- ISO 4217 + canonical integer-cents representation
• ERP Integrations -- DATEV NOT mentioned
• Accessing the Developer API (Support) -- 403 from public web; gated to logged-in Ramp users
• Downloading receipts (Support) -- admin-only receipt download in UI

EU / Germany Availability

• Ramp Europe waitlist page -- "coming this summer", waitlist signup, current EU operations via Stripe Payments UK / Stripe Technology Europe
• Ramp is launching in Europe (blog)
• Ramp acquires Billhop (press release, March 2026) -- regulatory licenses in UK + Sweden, EEA + UK expansion
• Behind the scenes: How Ramp built a global corporate card program (Stripe) -- confirms EUR card issuing for entities in Ireland, Netherlands, Spain, Germany, France, Italy via Stripe
• Ramp Global page -- "available in 190+ countries", multi-currency
• EU / UK Entity KYB / KYC Checks (Support) -- 403 from public web
• International overview (Support) -- 403 from public web
• Will my Ramp card work internationally? (Support) -- 403 from public web
• Europe Payment Card Supplement (Legal) -- minimal public surface
• Ramp Is Coming to Europe (neobanque.ch)

Integrations & ERP

• Ramp Integrations -- 200+ listed; DATEV NOT among them
• Ramp Accounting Automation Software -- 30+ ERPs listed; NetSuite, Sage Intacct, Workday, Dynamics, Oracle Fusion, Acumatica, QuickBooks, Xero, Zoho Books, Trimble; DATEV NOT listed
• Overview of Ramp Accounting (Support) -- 403 from public web

iPaaS

• Ramp on Workato -- 4.9/5 rating, generic CRUD action set + HTTP connector
• Ramp on Unified.to -- unified API wrapper
• Zapier: no dedicated Ramp Business connector found (multiple unrelated apps with similar names: WorkRamp, Ramper Marketing, etc.)
• Make.com: no native Ramp module found
• n8n: no native or community Ramp node found
• Celigo: not in marketplace

Pricing & Plans

• Ramp Pricing -- Free, Plus ($15/user/mo), Enterprise (custom)
• Ramp Pricing & Plans (Vendr) -- median ACV $21,600, platform fees $0-$15k/mo
• Ramp Plus overview (Support) -- 403 from public web
• Ramp Pricing (CheckThat.ai)
• Ramp Pricing 2026 (Capterra)

Reviews & Background

• Ramp Card: 2026 Review (NerdWallet)
• Ramp Review 2026 (Headwest Guide) -- confirms Plus tier includes "International and multi-currency support" for CAD/UK/EU
• Ramp launches enhanced corporate cards (PYMNTS)
• Ramp Reviews 2026 (G2)

Apidog Reference (Third-Party)

• How to use Ramp API (apidog) -- third-party summary, useful cross-check for auth + webhook patterns

DATEV Context (for the gap analysis)

• DATEV Accounting Germany explained (Facturwise)
• How DATEV became unshakable in German accounting (project-b)
• DATEV Integration in E-Commerce (XICTRON)
• How to succeed with your DATEV API integration (Chift)