Customer support decision
Intercom
A competent engineer can build a narrow Intercom-like workflow (chat widget, inbox, KB retrieval, LLM assistant) in weeks, but reproducing Intercom’s proprietary Fin model, enterprise features, integrations catalogue, and polish at scale is not realistic for a single maintainer.
Visit website↗$19/mo
$228/yr
Per seat. Read off the official pricing page.
$100one-off80 h to build
$100/mo8 h/mo upkeep
On cash alone, building overtakes the subscription at 6 seats.
The code exists. It is not what you are paying for.
These 2 projects are real, published, and do the core job — and this page still says keep paying. What the subscription buys is proprietary models and brand trust, and none of that ships in a repository. Fork one anyway if you want to. Go in knowing what it does not carry. What stays hard ↓ · All Intercom alternatives, with the arithmetic →
What a replacement has to do
- Accept customer messages (web widget/email), surface context from prior conversations and KB, draft/reply messages (auto or human-in-the-loop), convert to tickets and route, and log interactions for reporting.
What it still won’t have
- Intercom’s proprietary Fin AI Agent and its claimed self-improving behavior
- Out-of-the-box 350+ integrations and their maintained connectors
- Enterprise-grade features (SSO, HIPAA, SLAs) and vendor support
- Polished UX, managed scaling, and platform reliability guarantees
- Built-in pricing/usage billing (per-outcome billing model and seat management)
What remains hard
- Proprietary models
Fin AI Agent Our AI Agent and AI architecture trained specifically for customer service.
- Brand trust
Trusted by 30,000 + leading brands
First-year cost
Keep paying
Paying is—cheaper in year one.
On cash alone, building overtakes the subscription at 6 seats.
Money you would actually spend
Time you would spend
—
What you would spend
What we assumed
The verdict above measures whether you could build it. This one is only about money.
Runnable build prompt
Build a minimal AI-enabled helpdesk using PostgreSQL, Node.js (Express), React for the agent UI, and a vector DB (e.g. Milvus or Pinecone) for retrieval. Include: 1) a web chat widget that creates conversations in Postgres; 2) a shared inbox React app showing conversation threads and simple ticket metadata; 3) ticket create/update, tag, and routing rules (rule eval in Node); 4) knowledge base ingestion (PDF/HTML importer -> embeddings) and retrieval endpoint; 5) AI reply generator calling an external LLM API with retrieved context and a human-approval handoff; 6) basic reporting endpoints (conversations/day, resolution outcomes). Out of scope: multi-channel gateways (WhatsApp/SMS/Phone), 350+ third-party integrations, enterprise SSO/HIPAA compliance, and per-outcome billing. Provide error handling, tests for core endpoints, and a Docker Compose dev setup.
How we checked
How the score was reached
- Pay verdict base20
- An open-source build was found+5
- 4 cited sources+3
- Price verified on pricing page+3
- Hard moats found in the evidence-3
- Evidence score28
The base comes from the verdict. Everything under it is a check that either happened or did not, and each one is a fact frozen in this record rather than a judgement made at render time - so the same evidence always produces the same number.
How scoring works →Cited sources · 4
Every page the run actually retrieved.
- official productIntercom — product page
- official pricingIntercom pricing
- open sourcepapercups-io/papercups
- open sourcechatwoot/chatwoot
Integrity checks
What held up, and what did not.
