HubSpot details no-code Agent Builder and practical marketing workflows
HubSpot published a guide explaining how teams can deploy custom AI agents using its no-code Agent Builder and Agent Hub workflow triggers. The article outlines four building blocks for agents: repeatable tasks, accessible CRM data, defined triggers, and measurable outcomes. HubSpot notes that its builder allows users to describe tasks in plain English, configure permissions, and connect…
What you can do with it
AI at Work →Build custom marketing agents inside your CRM
What you get You can track ad influence on sales.
Builds no-code task-oriented agents that take action using CRM records and external tools.
- Who for
- marketers, sales and operations
- Cost
- Price not stated
- Effort
- an afternoon
Use it for
- Analyze lost CRM deals
- Track ad influence on sales
- Group recurring support tickets
How to set it up · From the article
- Define a narrow goal, scope, and measurable deliverable for your agent.
- Write plain-text instructions specifying the role, approach, and how to handle missing data.
- Add CRM inputs, connected reference files, and tool permissions in Agent Builder.
- Run test scenarios on real records and refine instructions before live deployment.
Watch out Outputs vary across runs; sensitive actions require human approval routines
Key points from the news
- A 2026 Menlo Ventures survey of 5,000 U.S. adults found 24% of AI users report regular use of AI agents.
- HubSpot Agent Builder enables no-code agent creation using CRM data, external apps via MCP, and defined execution triggers.
- Unmuted and Driven demonstrated live marketing and sales agent workflows for tracking ad attribution and explaining compensation plans.
To demonstrate practical deployments, the company highlighted several business use cases. Filippo Irdi, demand marketing manager at Unmuted, uses an agent to cross-reference LinkedIn ad impressions with HubSpot closed-won deals to track offline ad influence. Driven co-founder Andres De Jonge outlined a sales compensation agent that explains payout discrepancies without performing math calculations directly, leaving computations to a dedicated formula engine. Additionally, the guide details connecting HubSpot to ChatGPT to analyze customer support ticket trends.
HubSpot emphasizes implementing guardrails and testing procedures prior to scaling. It advises against deploying agents when rules-based automation suffices and suggests setting human approval requirements for sensitive actions like data alterations.
How to build an AI agent: from goal to live deployment
HubSpot Marketing Blog · 22 October 2026
Loading the full article…
This text was published by HubSpot Marketing Blog and written by Jenny Romanchuk. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Marketing & Small Business
All →- Anthropic launches Claude Duet, a proactive AI agent app for US Pro and Max users · 22 src
- How to earn Google AI Professional Certificate for free via Coursera trial · 1 src
- Canva launches Canva Research lab to boost design AI for small businesses · 4 src
- Apple may add Gemini‑style Intelligent Assistant to iWork, update briefly posted · 1 src
- OpenAI adds slide creation to ChatGPT for Pro, Business, Enterprise users · 1 src
Comments
via GitHub Discussions