Zendesk to acquire Forethought to expand AI agents for customer support
Updated 19:23 EDT / March 11, 2026 - Zendesk announced plans to acquire Forethought, an AI startup focused on automating customer service. The price wasn't disclosed.
For support leaders, this points to a faster move toward AI agents that don't just answer FAQs, but take action across systems and channels. Expect tighter workflows inside Zendesk and more opportunities to deflect repetitive tickets without sacrificing quality.
What Forethought brings
Founded in 2017, Forethought builds AI that interprets customer requests using natural language processing and machine learning. It ingests support data-historical tickets, help center content, and internal docs-then uses large language models to generate responses, extract details, and trigger actions in connected systems.
The platform covers email, chat, messaging, and voice. Common automated tasks include order tracking, password resets, appointment scheduling, and basic troubleshooting-exactly the high-volume work that clogs queues.
How Zendesk plans to use it
Zendesk plans to fold Forethought's technology into its Resolution Platform, which the company says improves by learning from every customer conversation without manual retraining. With this acquisition, Zendesk aims to push into self-learning AI agents that can generate, adapt, and execute more complex workflows across any channel or platform.
Planned capabilities include support for additional channels and a wider range of service environments, with deeper workflow automation and smarter handoffs to human agents when needed.
Quote to note: "Forethought was founded on the belief that AI will transform customer experience for every business," said co-founder and Chief Executive Sami Ghoche. "With Zendesk's platform, resources and global reach, we will bring our technology to many more organizations around the world, move faster on innovation and continue pushing the boundaries of what AI can do in customer experience."
Why this matters for support leaders
AI agents are moving from static Q&A to end-to-end task resolution. That means higher containment on routine flows, faster time to first response, and fewer context switches for agents.
The catch: value depends on data quality, clean workflows, and clear escalation rules. If your knowledge base is messy or systems aren't integrated, results will stall.
Practical steps to get ready
- Data readiness: centralize help content, tag intents, and clean historical tickets; remove outdated or conflicting articles.
- Workflow selection: start with 3-5 high-volume, low-risk intents (e.g., order status, password resets, plan changes).
- System mapping: confirm API access for CRM, billing, order management, and scheduling tools; document auth and permissions.
- Guardrails: define escalation triggers, tone rules, and fallback messages; set up mandatory disclosures where needed.
- Measurement: track containment rate, FCR, CSAT, AHT, and handoff quality; review transcripts weekly at launch.
- Change management: train agents on new handoff patterns and how to correct AI outputs; reward coaching behaviors.
- Compliance: validate data retention, PII handling, and access controls across every integrated system.
- Pilot first: run a staged rollout by channel and intent; expand only when KPIs stabilize.
Where this could show up in your stack
- Email: automated replies and routing with intent detection and suggested macros.
- Chat and messaging: guided workflows that collect details, verify identity, and complete simple actions.
- Voice: call summarization, next-best-action prompts, and faster escalations with context preserved.
- Agent assist: suggested responses, knowledge surfacing, and one-click actions during live conversations.
Funding context
Before the acquisition announcement, Forethought raised $117 million over five rounds. Backers include 8VC Management, Collaborative Fund Management, Frontline Ventures, New Enterprise Associates, Samsung Next Ventures, Sound Ventures, South Park Commons, Village Global VC, Cleo Capital, K9 Ventures, Industry Ventures, Operator Collective, and Blue Cloud Ventures.
Key takeaways for your roadmap
- Prioritize a single source of truth for knowledge; keep it versioned and auditable.
- Start with intents that have clear outcomes and existing APIs; avoid edge cases early.
- Codify bot-to-human SLAs and make the handoff obvious to customers.
- Instrument everything; a weekly KPI review beats big quarterly reworks.
- Pair AI agents with agent assist to improve accuracy and shorten training time.
If you're building skills for this shift, explore AI for Customer Support and the AI Learning Path for User Support Specialists.
Learn more about the companies involved: Zendesk and Forethought.
Your membership also unlocks:
