Guide

Self-hosted customer support software: what it includes, what it costs, and how to roll it out

Self-hosted customer support software runs on a server you control. Conversations, contacts and reports are stored in your own database, not in a vendor's cloud. This guide covers the parts of a support stack, the cost for three team sizes, the work you take on, and a plan for the first week. It uses DChat as the worked example.

The five parts of a support stack

Most support teams need the same five things. When you compare products, check each one separately, because vendors bundle and price them differently.

Part What it does In DChat
Live chat A widget on your site and a console for agents Free, unlimited agents and domains
Help center Articles customers can search before they ask Free, with widget search
Contacts and reports Who you talked to, and how the team is doing Free, with CSV export
AI agent Answers from your content and hands over to a person Premium, on OpenAI, Anthropic or local Ollama
Automation and roles Routing rules, custom roles, extra channels, voice Premium

DChat is chat-first. Its ticket features are simpler than a dedicated ticketing system, so a team that lives in long email threads with many custom ticket fields should test that part carefully. The helpdesk buyer's checklist lists the questions to ask.

What it costs for 5, 10 and 25 agents

Hosted support tools usually charge for every seat, and some add a fee for each AI answer or resolution. Self-hosted pricing moves part of the cost to your own server. Here is DChat's licence cost at list price:

Team Community (live chat, help center, contacts, reports) Premium for every agent
5 agents $0 $95 a month
10 agents $0 $190 a month
25 agents $0 $475 a month

Premium is $19 per covered agent per month, and you choose which agents it covers. Enterprise, at $49, adds single sign-on and advanced SLA policies. If paid coverage lapses, the paid features switch off and live chat keeps running. Current terms are on the pricing page.

Add your own costs to this: the server, the SQL Server licence or edition you use, backups, and the time of the person who looks after them. If you use a cloud AI model, you also pay that provider for usage.

What you run yourself

Self-hosting gives you control, and it gives you work. Be honest about who will do each job before you start.

  • The server. DChat needs the ASP.NET Core 8 runtime and at least 1 GB of RAM, with 2 GB or more recommended. It runs on Windows Server 2016 or later with IIS, on Linux behind Nginx, or in Docker.
  • The database. SQL Server 2014 or later; SQL Server 2022 is recommended. Every transcript, contact and report is stored there.
  • Backups and restores. Back up the database on a schedule, and test a restore before you rely on it. See how to back up and restore.
  • Updates. You decide when to install a new release. Read the release notes first.
  • HTTPS and access. You issue the certificate and decide who can reach the dashboard.

The full list is on the system requirements page.

Where the data goes

With self-hosted support software, the default answer is "nowhere": conversations stay in your database. Traffic leaves your network only for what you set up. In DChat that means the AI provider you choose, the email you send, and any channel or notification you connect, such as Slack, Microsoft Teams or Discord.

AI is the part to check most carefully. With OpenAI or Anthropic keys, the conversation text the AI works with is sent to that provider. With a local Ollama model, prompts and answers stay on your own infrastructure; the trade-off is hardware and model quality. See where live chat data is stored and the self-hosted AI chatbot page.

A first-week rollout plan

  1. Day 1: install on a test server. Download the package, create the database and sign in. Follow the guide for Windows and IIS or Linux and Docker.
  2. Day 2: add the widget to a staging page. Send yourself a chat from a phone and from a desktop browser. See how to add live chat to your website.
  3. Day 3: set up agents and departments. Create agent accounts and departments, and decide what visitors see when no one is online. The offline form collects their message for later.
  4. Day 4: write the first ten help articles. Start with the questions your team answers most often. Published articles are searchable from the widget.
  5. Day 5: set up backups and HTTPS. Schedule the database backup, restore it once to a scratch database, and put the production certificate in place.
  6. Days 6 and 7: go live on one page, then all pages. Watch the first real conversations, then roll the widget out to the rest of the site.

Add the AI agent after the team is comfortable with live chat. It answers better once the help center has real content to draw on.

When hosted support software is the better choice

  • No one on the team can run a server, a database and backups.
  • Agents need a native mobile app on day one. DChat has no native iOS or Android app; its web dashboard installs on a phone as a web app and can send push notifications.
  • You want the vendor to carry uptime and security patching under a managed commitment.
  • You need a large catalogue of ready-made integrations more than you need control of the data.

Common questions

What is self-hosted customer support software?

It is support software you install on a server you control, so conversations, contacts and reports are stored in your own database instead of a vendor's cloud. You run the server, the database and the backups.

How much does self-hosted customer support software cost?

With DChat, live chat, the help center, contacts and reports are free for any number of agents. Premium, which adds the AI agent, custom roles, automation, extra channels and voice, is $19 per covered agent per month: $95 for 5 agents, $190 for 10, $475 for 25. You also pay for your own server and database.

Can self-hosted support software use AI without sending data to a cloud provider?

Yes, with a local model. DChat's AI agent can use a local Ollama server, so prompts and answers stay on your own infrastructure. With OpenAI or Anthropic keys, the conversation text the AI works with is sent to that provider.

Who should not self-host customer support software?

Teams with no one to run a server, a database and backups, and teams that need a native mobile app or a vendor-managed uptime commitment on day one. A hosted tool is the better fit for them.

Self-hosted customer support software

Deploy live chat on your own terms, not on someone else's pricing model.

DChat gives you the installable server, web dashboard, website widget, and desktop agent tools in one self-hosted product, with live chat free to run, and an AI agent that answers, acts with the approvals you set and hands off to your team when you want one. Run it on infrastructure you trust, on the model you choose.

Deployment

Install on Windows or Linux, behind IIS or Nginx, in a VM, or in Docker if that fits your stack.

Commercial model

Live chat is free with unlimited agents. Premium, with the AI agent, is billed per agent per month, never per conversation.

AI Flexibility

Run the AI agent on a local Ollama model, or connect OpenAI and Anthropic with your own provider accounts.