Merge
One API for hundreds of customer-facing integrations, plus agent tools, LLM routing and AI governance
By Merge (Merge API, Inc.) · HQ San Francisco, US · 4.5/5 Value-Position score (estimate)
Positioning guardrails
Best for
- B2B SaaS product teams that need hundreds of customer-facing integrations without building and maintaining connectors in-house
- AI agent and voice-AI builders that need authenticated, scoped access to third-party tools and systems
- Platform engineering teams consolidating several point integration vendors into one Unified API
- Enterprises that must govern employee AI access, MCP distribution, model spend and DLP in one place
Ideal size: 50–500 people · Scale-up or enterprise with a product engineering team shipping integrations
Not for
- Teams that need only one or two integrations and can build them directly against vendor APIs
- Products with no engineering capacity to adopt and maintain an API integration layer
- Buyers who want a self-serve, fully transparent price list beyond the entry tier, since Professional and Enterprise are contract-only
- Companies looking for an end-user SaaS application rather than embedded integration and AI-connectivity infrastructure
Value metrics scorecard
Time-to-Value
1–2 weeks to a first integration
~14 days to first production value
Total Cost of Ownership
$7,800/yr
Starts at $7,800 · Subscription by production Linked Accounts: first 3 free, then $650/month for up to 10, $65 per account after. Professional and Enterprise are contract-based.
Implementation Friction
2/5
Engineering + admin effort required
Value-Position score
out of 5 · model estimate
Full cost breakdown
Mandatory implementation fee
None
Seat tiers
Priced on production Linked Accounts, not seats: 3 free on Launch, 10 included at $650/month, volume discounts on higher tiers
Add-on costs
- $65 per additional Linked Account beyond 10 on the Launch plan
- Professional adds custom fields, field-level scopes, custom sync frequencies and 60-day sandboxes
- Enterprise adds audit trail, go-live support, unlimited sandboxes, dedicated account manager and uptime SLAs
- Single-tenant or self-hosted deployment available for purchase on higher tiers
Company & support
Who is behind Merge, and how your team gets help once it is live.
Company
- Founded
- 2020 · 6 yrs in business
- Headquarters
- San Francisco, US
How you get support
- PhoneNot listed
- EmailAll plans
- Live chatPaid plans
- Support portal / ticketsNot listed
- Community forumNot listed
- Help centre / docsAll plans
- Dedicated account managerEnterprise only
- In person / on-siteNot listed
- Hours
- Not recorded
- Response time
- Support and uptime SLAs offered on the Enterprise plan
Launch includes email support; Professional and Enterprise add chat. Enterprise adds a dedicated account manager, shared Slack channel and premium go-live support.
“Not listed” means the vendor’s public pages don’t mention that channel, not that it is unavailable. Ask about it during evaluation.
Market position
Where Merge sits against its closest alternatives. Pick any two of cost, speed, friction and buyer score, and up to 9 companies to compare.
Quadrant view
Typical annual cost × Time-to-value
The lines cross at the median of the solutions shown, so about half sit on each side of each line.
Companies on the chart 6 / 10
- Merge
- CircleCI
- Superblocks
- Appsmith
- Atomicwork
- mindzie
Add or change companies
Up to 10 companies including Merge. Listed closest first.
Stack fit signal
Compatibility with standard B2B ecosystems.
Ships an official MCP server. Connects to Claude Code, Claude Desktop, ChatGPT connectors and Cursor out of the box.
AI & MCP readiness
What Merge ships in AI, and what it asks of your ecosystem.
AI features shipped
Agent Handler gives AI agents authenticated access to thousands of third-party tools with scoped permissions and audit logs. Gateway routes across model providers with failover, per-project routing policies and spend visibility. Workforce governs employee AI with MCP distribution, model allow/block lists, budgets, DLP on responses and prompt-injection guardrails.
Your data & models
- Trains on your data
- Never trains on your data
- Runs on
- OpenAI, Anthropic, Google, Open-weight models
- AI pricing
- Billed by usage or credits
In your ecosystem
- AI connection
- Official MCP server
- Model key
- Either
- AI usage audit
- Basic visibility
Compliance attestations
* IAPP AIGP certifies individuals, not products. It means named staff hold the credential — not that the platform does.
Bottom line
Merge is the connectivity layer for production AI: one API for hundreds of customer-facing integrations (HRIS, ATS, accounting, CRM, ticketing) plus agent tooling, LLM routing and AI governance. B2B SaaS and enterprise teams use it to ship integrations in days rather than quarters, give agents authenticated access to thousands of third-party tools, route across model providers, and govern employee AI access with MCP controls, DLP and audit. The entry tier is free for three production linked accounts, then $650/month.
Frequently asked questions
How quickly can we ship our first integration with Merge?
Vendor-published customer evidence points to days, not months. Telnyx integrated voice AI agents with dozens of Merge MCP servers in under a week, and one CTO reports integrating, testing and releasing against the Unified API in less than a sprint. Bonusly reports building HRIS integrations roughly 20x faster than doing it in-house, and Merge positions the platform as unlocking hundreds of integrations in days rather than years. Plan for one to two weeks to a first production integration, then add catalogue coverage incrementally, and note that 60-day developer sandboxes ship with the Professional plan.
What does Merge cost?
The Launch plan starts free with three production Linked Accounts. Beyond that, Launch is $650 per month for up to 10 production Linked Accounts, then $65 per Linked Account, with volume discounts available. Professional and Enterprise are contract-based: Professional adds custom fields, field-level scopes, custom sync frequencies, 60-day sandboxes and go-live packages, while Enterprise adds audit trail, premium go-live support, unlimited sandboxes, a dedicated account manager, a shared Slack channel and support and uptime SLAs. No mandatory implementation fee is published.
Is Merge secure and compliant enough for enterprise reviews?
Merge states it is certified in SOC 2 Type II, ISO 27001, HIPAA and GDPR, and also references CCPA. It describes encryption in transit and at rest with an additional application encryption layer for PII, hosting in AWS data centres across US, EU and APAC regions, and a choice of multi-tenant, single-tenant or bring-your-own-storage deployment. Enterprise controls include SAML SSO, role-based access control, scoped permissions, selective sync, data deletion endpoints, audit trail and annual penetration testing.
Does Merge support MCP and AI agents?
Yes. Merge Agent Handler gives agents authenticated access to thousands of third-party tools with scoped permissions and audit logs, and Merge publishes an MCP connector list alongside its Unified APIs. Merge Workforce distributes approved MCPs and skills to teams, links Okta or Entra identities so permissions update automatically, restricts each agent's reach down to individual endpoints, blocks or allow-lists models per team, and applies DLP that redacts sensitive data before it reaches a model.
Can we bring our own LLM keys, or does Merge bundle models?
Both options are documented. Merge Gateway routes to every major model provider through one API with automatic failover when providers go down, and the changelog describes models offered as managed or bring-your-own key. Gateway supports prebuilt or custom routing policies, per-customer and per-project guardrail overrides, model and region restrictions, zero-data-retention flags, per-team budgets, and usage and spend reporting, which is how Merge positions control over model cost without lock-in.
Does Merge train AI models on our data?
Merge's privacy policy states that Merge does not develop, improve, or train generalized AI and/or ML models using Personal Information, so customer data is not used to train generalised models. Buyers routing through Gateway should still confirm terms for the upstream model providers they select, noting that individual models and vendors in Merge's catalogue are documented with zero-data-retention options, and that request and response payload logging is enabled by default on newly created organisations and can be managed in the dashboard.