Customer support decision
Crisp
A competent developer can assemble a useful replacement (widget + shared inbox + KB + LLM) in about a week using existing open-source components and hosted LLMs; enterprise integrations, advanced workflow builders, and the vendor's ecosystem would be costly to replicate.
Visit website↗Open-source builds that already do this
Every project below is open source and already does this job today. Fork one, self-host it, or take the parts you need — the build prompt further down assumes an empty file, and this is the shortcut past that. Licences differ; check the one on each card before you ship.
Code Crisp publishes itself
Not a way out of the subscription — these are the vendor’s own repositories. Worth a look for how they build, and for anything you would have to integrate with.
- crisp-im/node-crisp-api★112 GitHub starsMIT:zap: Crisp API Node Wrapper↗
- crisp-im/php-crisp-api★59 GitHub starsMIT:elephant: Crisp API PHP Wrapper↗
- crisp-im/crisp-sdk-android★59 GitHub starslicence unclear:package: Crisp Android SDK, add a chat in any Android app and communicate with your users.↗
- crisp-im/crisp-sdk-ios★52 GitHub starslicence unclear:package: Crisp iOS SDK, add a chat in any iOS app and communicate with your users.↗
What a replacement has to do
- Provide a web chat widget that sends messages to a centralized inbox, ingest product docs/FAQs into a searchable knowledge store, call a hosted LLM API to generate agent replies (with context and KB grounding), persist conversations and metadata in a database, and expose a basic UI for agents to view/edit conversations and approve/edit AI replies.
What it still won’t have
- Built-in omnichannel integrations (WhatsApp, Instagram, Messenger, etc.)
- No-code visual workflow/automation builder
- Included AI credits and prebuilt AI Agent templates
- Enterprise features like SLAs, dedicated onboarding, enhanced rate limits, and white‑labeling
What remains hard
- Brand trust
Trusted by 10,000 companies of all sizes
First-year cost
Keep paying
Paying is—cheaper in year one.
On cash alone, building overtakes the subscription at 3 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 customer support workspace using Node.js (Express) + React, Postgres, and Redis: include an embeddable web chat widget, server endpoints to receive/send messages, Postgres schemas for conversations/users/messages, a KB ingestion script (crawl site + PDF import) that stores embeddings in a vector DB (e.g., Milvus or Postgres pgvector), an LLM integration module that retrieves top-K passages and calls a hosted LLM API for grounded replies, and an agent web UI showing a shared inbox with AI-suggested replies and the ability to edit/approve. Out of scope: multi-channel integrations (WhatsApp/Instagram), enterprise SLA features, white-labeling. Include error handling, rate-limit protection for LLM calls, basic unit tests for API endpoints, and deployment scripts (Dockerfiles + docker-compose).
How we checked
How the score was reached
- Build verdict base78
- An open-source build was found+5
- 4 cited sources+3
- Price verified on pricing page+3
- Evidence score89
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 productCrisp — official product page
- official pricingCrisp Pricing
- open sourcechatwoot/chatwoot
- open sourcepapercups-io/papercups
Integrity checks
What held up, and what did not.
