Chatbase

· #431 most-used

Build, deploy, and continuously improve AI chatbots trained on your data

MarketingSupportAIAutomationLead GenerationWebsite & CMS

Chatbase is a platform for building and deploying AI chatbots trained on your own data — website content, help articles, product documentation, or any text you provide. Connect it to Actionist and your agents can route chatbot-captured leads to your CRM within about a minute of submission, retrain chatbots automatically when content changes, export conversation transcripts for quality review and compliance archiving, manage contacts across all your bots, and keep your account clean by decommissioning inactive chatbots on schedule — all without anyone logging into the Chatbase dashboard.

Try Chatbase with Actionist Visit Chatbase

Average time saved

10 hours

per person · per month

≈ 1 workdays back

Eliminates manual work. Agents eliminate manual lead exports, dashboard-based knowledge-base re-uploads, transcript reviews for content gaps, and GDPR contact deletion across multiple chatbots.

Schedule

What your Chatbase agent runs on autopilot

A week of scheduled jobs your Actionist agent will execute on your behalf.

28Scheduled jobs

7Agents at work

24/7Always on

Agents

Tue–Thu

Tue

Wed

Thu

10a

11a

12p

Multi-app workflows

Chatbase × every other app you use

End-to-end automations that span multiple apps — each one a real business outcome.

6Workflows

6Apps spanned

~15 hrsSaved / week

5Personas served

For sales

Featured4 apps

Chatbot lead routed to CRM and rep notified within a minute

When a visitor submits their contact information through the chatbot, Actionist retrieves the conversation context, creates or updates the contact in HubSpot with the chat transcript attached, posts a lead summary to the sales Slack channel, and creates a follow-up task for the assigned rep — the entire sequence completes within about a minute of the form submission.

~8 hrs

Time saved for your team — every week, on autopilot

The flow

Trigger·When a visitor submits their details through the website chatbot lead form

Trigger

Step 1

Chatbase

Lead submitted via website chatbot

read

Step 2

Chatbase

Get conversation messages for the lead's chat session

write

Step 3

HubSpot

Create or update contact with lead details and chat context

write

Step 4

Slack

Post lead summary with conversation highlights to #sales

write

Step 5

Google Calendar

Create follow-up task for the assigned sales rep

Result

Create or update contact with lead details and chat contextPost lead summary with conversation highlights to #salesCreate follow-up task for the assigned sales rep

The win

Saved per run

15 min

Runs / week

~30×

Every lead is in the CRM before the visitor leaves the page

Driven bySales Agent

ROI

Savings

What your team gets back — two angles: what you stop doing manually, and what that's worth.

Without Actionist

What you do manually today

With Actionist

What your agent runs for you

Sales
25 min / week
Manual lead export and CRM entry
Sales ops exports leads from the Chatbase dashboard each morning, copies names and emails into the CRM, and manually assigns follow-up tasks to reps — often hours after the lead came in.
Sales Agent
0 min
Agent routes chatbot leads to CRM within a minute
When a visitor submits contact details through the chatbot, the agent pushes the lead to HubSpot, attaches the conversation context, and notifies the assigned rep — before the visitor has even closed the tab.
Marketing
20 min / week
Manual knowledge base re-upload
The marketing team manually copies updated content into the Chatbase training interface each time a page changes — a step that gets skipped under deadline pressure, leaving the bot with stale answers.
Marketing Agent
0 min
Chatbot retrained automatically when content changes
When a blog post, landing page, or FAQ is updated, the agent sends the new content to Chatbase and the chatbot reflects the change within about a minute — no dashboard visit required.
Customer Support
90 min / week
Manual transcript review for content gaps
The support lead spends 2-3 hours each week reading through chat logs to identify questions the bot fumbled, copying them into a spreadsheet, and emailing the content team with a summary.
Customer Support Agent
0 min
Agent surfaces unanswered questions as content tasks weekly
Every Friday the agent reviews conversations, identifies recurring unanswered questions, and creates Notion tasks for the content team — content priorities come from real customer queries, not guesswork.
Human Resources
15 min / week
Manual new-hire chatbot setup
HR manually creates contact records in Chatbase for each cohort of new starters, copying details from the HRIS one by one — a task that often gets skipped, leaving the bot unable to personalise responses.
Human Resources Agent
0 min
Onboarding chatbot personalised for each new hire automatically
When a new hire is added to the tracker, the agent creates their Chatbase contact with role and department attributes — the bot greets them by name and serves relevant answers from day one.
Finance
10 min / week
Manual chatbot ROI calculation
Finance manually exports lead counts from Chatbase, matches them against ad spend in a separate spreadsheet, and calculates cost-per-lead — a process that takes 45 minutes and happens monthly at best.
Finance Agent
0 min
Chatbot lead-to-cost metrics calculated weekly without manual work
Every Tuesday the agent exports conversation and lead data from Chatbase and writes cost-per-lead metrics to the attribution spreadsheet — the ROI calculation is ready before the budget review.
Operations
20 min / week
Manual chatbot account housekeeping
Someone on the ops team periodically logs into Chatbase, identifies bots with no recent conversations, manually exports their history, and deletes them — a task that gets deprioritised until the plan limit is hit.
Operations Agent
0 min
Inactive chatbots decommissioned automatically on schedule
The operations agent reviews all chatbots weekly, exports history for inactive ones, deletes them, and posts a summary — the account stays within plan limits without manual housekeeping.
Legal
12 min / week
Manual data subject deletion across chatbots
Legal manually searches for the requester's email in each chatbot's contact list in the Chatbase dashboard, exports records separately, deletes them one by one, and updates the GDPR tracker — a process that typically takes 30-60 minutes per request.
Legal Agent
0 min
GDPR deletion requests fulfilled across all chatbots in minutes
When a deletion request arrives, the agent identifies all matching contacts across every chatbot, exports transcripts for the compliance archive, deletes the records, and logs completion — all within the same scheduled agent task.

+ 100s of other Chatbase automations

Average time saved

19 hrs / person / month

Calculator

Calculate what your team saves

Team size

5 people

Hourly rate

$75 / hr

Hours saved / week

Hours saved / year

625

Annual ROI

$46,875

Based on Chatbase's typical team usage — the visible tasks plus a few other automations the agent runs: ~2.5 hrs / person / week of admin work automated.

Connect

How to plug Chatbase into Actionist

Pick the connection method that suits your environment.

Connect Chatbase using your account's Secret Key. The key gives Actionist full read and write access to chatbots, conversations, leads, and contacts in your Chatbase workspace.

Open Chatbase Settings

Generate a secret key

Navigate to the API Keys section and click Generate new secret key. Copy the key immediately — it is only shown once.

Paste into Actionist

Paste the key into the Secret Key field in Actionist and click Test connection. Actionist will confirm the handshake with a read-only health check.

Credentials you'll need

Secret Key*

Chatbase dashboard → Settings → API Keys → Generate new secret key

Actions

15 actions your agent can call

Read and write operations available to your Actionist agent.

Triggers

1 event your agent can react to

Events your agent watches for, and the actions it kicks off in response.

FAQs

Questions about Chatbase + Actionist

How does Actionist connect to Chatbase?

Go to the Apps tab, find Chatbase, and click Connect. The standard path is an API key — navigate to your Chatbase dashboard, open Settings, and generate a Secret Key. Paste it into Actionist and click Test connection. Actionist will run a read-only call against the health endpoint to confirm the handshake before any agent actions run.

What permissions does the agent need on my Chatbase account?

Your Chatbase Secret Key carries the full permissions of your account. The key grants read and write access to all chatbots (agents) in the account — create, update, delete, and read conversations, leads, and contacts. Store it as a credential in Actionist rather than pasting it in plain text. If you need to scope access, create a dedicated Chatbase workspace account and generate a key there.

Can I connect Chatbase to other apps in the same workflow?

Yes — Chatbase is most useful alongside the apps where conversations produce value. Common combinations: push new leads collected by your chatbot into HubSpot or a Google Sheet; trigger a Slack alert when a lead submits their contact information; export conversation transcripts to Notion for quality review; update chatbot training data when a new document is added in Google Drive. Any of Actionist's connected apps can send or receive data alongside Chatbase in the same agent task.

What are the most common things agents do with Chatbase?

The four patterns that come up most: (1) lead routing — when a chatbot visitor submits their contact information the agent pushes the lead to your CRM and notifies the sales team; (2) conversation review — exporting transcripts on a schedule so the team can spot unanswered questions and gaps in training data; (3) knowledge base updates — retraining the chatbot automatically when new content is published; (4) contact sync — keeping Chatbase contacts aligned with records in your CRM so the chatbot has accurate context about returning customers.

What is the difference between Chatbase API v1 and v2?

Chatbase exposes API v1 endpoints for chatbot creation, settings updates, retraining, and data export. API v2 (available on the Standard plan and above) adds structured error codes, cursor-based pagination, SSE streaming for chat, and richer conversation and contacts endpoints. Actionist uses whichever version the operation requires — the agent handles the API versioning transparently.

What webhook events does Chatbase support?

Chatbase supports one webhook event: `leads.submit`, which fires when a visitor fills in the lead-capture form on your chatbot (name, email, phone). There is no polling required — the webhook delivers the payload to Actionist within about a minute of the submission. For other events — new conversations, message volume changes — use a scheduled agent task that calls Get conversations or Get leads on a cadence you define.

Can I retrain my Chatbase chatbot automatically through Actionist?

Yes. The Update chatbot training data action sends new text content to the Chatbase API, which retrains the underlying model on that content. This is available on the Standard plan and above. You can trigger it from Actionist whenever a new document is added to Google Drive, a page is updated in Notion, or a new help article is published — the chatbot stays current without anyone manually re-uploading files in the Chatbase dashboard.

How do custom contact attributes work with Actionist?

Custom attributes let you store structured data on Chatbase contacts beyond the default name, email, and phone fields — for example, plan tier, account ID, or last purchase date. Actionist can create or update custom attributes on contacts when events happen in other systems: when a deal closes in HubSpot, the agent updates the contact's plan tier in Chatbase so the chatbot can personalise responses for that customer immediately.