Intercom vs Chatwoot

Intercom bills through a contract that your account team defines, combining seat-based charges for agents with usage-based charges for its Fin AI agent. Confirm in the written order form which seats, which Messenger or inbox features, which support tier and which Fin charges apply, and whether any of these renew automatically. Seat counts, outcome rates and tier boundaries change with the contract, so the figures in your own order form are the ones that matter. Ask for a year-by-year projection at your expected conversation volume, including what happens when volume rises, since the usage-based part of the bill moves with it.

The Community Edition has no license fee, but it is not zero cost to operate. You still pay for the servers or cloud instances, PostgreSQL, Redis, backups, email delivery, domains and TLS certificates, and the staff time to install and maintain it. The Community Edition license covers the code under the MIT Expat terms, with the enterprise/ directory licensed separately, so confirm which features sit in each edition before you assume a feature is free. Ask your team to list every component they will run and who owns each one.

Chatwoot's self-hosted plans page lists Premium Support at $19 per agent per month, which adds Captain AI, voice calls, custom branding, agent capacity, roles and permissions and priority support. The Enterprise Edition is listed at $99 per agent per month and adds SSO with SAML and SLA policies. These are recurring charges that scale with the number of agents on the plan, so the cost grows with headcount, not with conversation volume. Confirm current prices and billing terms directly with Chatwoot before you budget, since list prices change.

Hosting is recurring and depends on your chosen environment: a Linux virtual machine, Docker, Kubernetes with the Helm chart for larger high-availability setups, or one of the documented cloud guides. Chatwoot's requirements recommend at least 4 CPU cores and 4 GB of RAM for up to 10,000 conversations a day, and 8 cores and 8 GB for up to 20,000, plus at least 1 GB of swap. Costs rise with conversation volume, attachment storage, number of channels and high-availability replicas. Ask your provider or internal platform team to price the sized instances, storage, load balancing and data transfer for your expected peak.

Database and cache operations are recurring costs that you carry yourself. Chatwoot supports only PostgreSQL as its database and recommends Redis 7.0 or higher. Costs include managed or self-run database instances, storage growth, automated backups, point-in-time recovery, patching and monitoring. Redis adds memory and failover needs. Sidekiq can use more than 1 GB of memory on a very active server, so memory sizing drives cost. If you use a managed database service, ask for its price per instance size and backup retention in your region.

Captain AI has two cost layers: the Chatwoot plan that unlocks it, which is listed at $19 per agent per month on Premium Support, and the model usage behind it. Model usage is recurring and depends on your chosen provider, the number of automated conversations, message length and any hosting you do for models yourself. Confirm with Chatwoot whether Captain requires a specific model provider, and ask your provider for per-token or per-request pricing at your expected volume. Model costs can exceed the plan cost once automated volume grows, so measure a pilot before committing.

Staff time is the largest cost that teams often overlook, and it recurs. Your engineers handle Chatwoot upgrades, which ship as frequent releases, along with OS and dependency patches, database maintenance, backup restore tests, monitoring, and on-call response for outages. Ubuntu 20.04 is the only officially supported Linux distribution, so OS upgrades may also be needed. Ask for a clear split of who does what: internal staff, a managed hosting provider, or a partner under a retainer. A written quote should state the scope of upgrades, testing and support hours, and response expectations.

Migration is a one-time cost made of planning, export work, import tooling, testing and a parallel-run period. Intercom's CSV and S3 exports omit full transcripts, so moving full history usually means pulling content through the Conversations API, which requires engineering effort. You also rebuild the widget, inboxes, teams, automation rules and contacts. Effort depends on the number of conversations, attachments, custom attributes and integrations. Ask any implementation partner for a fixed-scope written proposal that lists the data objects, the export method, the validation steps and the cutover plan.

The Community Edition is free to download and run, so the software license is the only line that drops to zero. Your remaining costs are the servers and database you run, the people who install and maintain Chatwoot, and any paid tier you choose. Chatwoot's self-hosted plans page lists Premium Support at $19 per agent per month, which adds Captain AI, voice calls, custom branding, roles and permissions and priority support, and Enterprise Edition at $99 per agent per month, which adds SSO with SAML and SLA policies. Intercom's costs scale with seats and, for Fin, with resolved outcomes, so compare Intercom's current invoice against seat count, the tier you need and your hosting and staff time. Ask Intercom for a renewal quote and build your own model for the Chatwoot side.

Duration depends mostly on how much conversation history you need to keep, how many contacts, inboxes, teams and automation rules must be rebuilt, and how many channels you cut over. Pulling full transcripts is the slowest step, because Intercom's CSV and S3 exports omit message content and the Conversations API has to be used for full history. Running Intercom and Chatwoot side by side during cutover adds time but reduces risk. The scope of your widget, email and social channels and the number of agents who need training also shift the timeline. Ask any implementation partner for a written scope with deliverables and dates before work starts, and test the export on a sample of conversations first.

Plan for named owners in four areas: an administrator who can create accounts, inboxes and API tokens in both systems, an engineer who runs the server and database, a support lead who signs off on how inboxes, teams, canned responses and automation rules map across, and a security or compliance contact who reviews where customer data is stored. Someone must also decide which history to keep and how long, since transcripts often contain personal data. Your team will need time to test the new widget on your website and to retrain agents on the Chatwoot interface. Confirm these commitments before the project starts so that the schedule does not depend on availability you have not reserved.

A team with Linux, PostgreSQL and Redis experience can install Chatwoot using the documented options, which include a Linux VM installation script, Docker, and a Kubernetes Helm chart for larger high-availability deployments. The harder parts are the data migration, the cutover, and the hardening and backup routines that keep customer conversations safe. Chatwoot's requirements list Ruby 3.2 or later, Node.js 20.x, PostgreSQL only, Redis 7.0 or higher, and Ubuntu 20.04 as the only officially supported Linux distribution, so check that your base image matches before you start. Decide early whether you want in-house capacity for ongoing operation or a fixed-scope engagement with DeployOpen, which plans, implements, hardens, backs up and documents self-hosted platforms in your environment.

No. Captain AI is listed on Chatwoot's Premium Support plan at $19 per agent per month, not on the free Community Edition. The Community Edition covers the core inbox, website live chat, help center portal and reporting. Before you rely on Captain for automated answers, confirm the current feature list on Chatwoot's self-hosted plans page, and check which language model provider and hosting you would use, since the model cost is separate from the Chatwoot license. Also confirm how Captain handles the knowledge sources you plan to give it.

Single sign-on with SAML and SLA policies are listed on Chatwoot's Enterprise Edition plan at $99 per agent per month. Premium Support, at $19 per agent per month, adds Captain AI, voice calls, custom branding, agent capacity, roles and permissions and priority support, but the plans page places SAML and SLA policies on Enterprise. If your identity provider uses OIDC rather than SAML, confirm whether Chatwoot supports it on your plan before you commit. Ask Chatwoot for written confirmation of which edition includes each feature you need, because feature placement can change between releases.

You are responsible for applying upgrades to your own installation, so they become part of your operating routine. Chatwoot publishes releases on GitHub, such as v4.18.0 on September 18, 2026, and the release notes should be read before each update. Plan for a staging environment, a database backup before each upgrade, and a rollback path. Upgrade frequency and testing effort depend on how many custom integrations, automation rules and channels you run. Enterprise and Premium customers should confirm what support their plan provides for upgrade assistance. Upgrades can also be handled through a retainer with an implementation partner, which covers ongoing maintenance and testing in your environment.