# Brainfish — full-text index for LLM ingestion

> Brainfish is AI product support agents for B2B SaaS. Resolve tickets across every channel, ground every answer in your real product knowledge, and never train on customer data.

## Customer outcomes

### ChangeEngine — 60% of ops time reclaimed and 1 in 4 customers self-serving
Source: https://www.brainfishai.com/customers/how-changeengine-reclaimed-60-of-ops-time-and-got-1-in-4-customers-self-serving
Industry: HR Tech
How ChangeEngine rebuilt its support motion around Brainfish, without hiring, and without rewriting a single workflow.
Metrics:
- 60% Ops team time reclaimed in first quarter
- 1 in 4 Customers self-serving every month
- 3 → 1 day Time to publish a new article
- 67% Reduction in time-to-resolve
Quote — Jess Chen: "The product can change from morning to night, day to day, or week to week. Brainfish gives the team a way to keep documentation moving at the same speed as the product."

### Coassemble — Doubled qualified leads and cleared the overnight queue
Source: https://www.brainfishai.com/customers/how-coassemble-doubled-qualified-leads-and-cleared-the-overnight-ticket-queue
Industry: L&D Platform
How Coassemble turned AI support into a self-serve revenue engine across customers, partners, leads, and internal teams.
Metrics:
- 2× Qualified leads from the website agent
- 22% Site visitors booked meetings with Brainfish
- ~80% Lead → opportunity conversion rate
- 0  Overnight tickets after launching
Quote — Eliza Clayworth: "All internal support questions now get answered by Brainfish directly in Slack, instead of clogging our platform support channel. We can discuss issues right there in the thread, and if something comes up that isn't already documented, we can immediately convert that conversation into a new knowledge article."

### Kloud Connect — 300+ hours saved with an in-app knowledge layer
Source: https://www.brainfishai.com/customers/kloud-connect-saves-300-hours-and-scales-support-with-an-in-app-knowledge-layer
Industry: Accounting Software
As Kloud Connect scaled rapidly across accounting firms of all sizes, their support team faced growing pressure to deliver fast, accurate guidance across a wide range of workflows and user roles. With administrators managing setup to accountants working daily in the platform,…
Metrics:
- 300+ Hours saved by the support team
- Multi User personas served (admins → accountants)
- 1 Source of truth, replacing Confluence
- Live Knowledge updates with no manual sync

### Vrio — 80% of developer questions resolved automatically
Source: https://www.brainfishai.com/customers/how-vrio-scaled-technical-support-without-slowing-down-engineering
Industry: Software / Developer Tools
How Vrio gave its API customers instant technical answers, and let its engineers go back to building.
Metrics:
- 80–85% Developer questions resolved automatically
- 30–50 → edge cases Daily engineer ticket volume
- Live Docs synced from CI/CD pipeline, no manual updates
- 2 → 1 Doc systems collapsed (Readme + Confluence → one knowledge layer)
Quote — Andrew Ronchetti: "I cannot understate the importance of Brainfish implementing into our workflows. Information remains current based upon the data that's deployed and we don't have to rely on manual updates."

### Gradeo — Scaled proactive exam support without adding headcount
Source: https://www.brainfishai.com/customers/how-gradeo-scaled-proactive-exam-support-without-increasing-headcount
Industry: EdTech
Gradeo is an end-to-end online assessment platform serving schools with exam preparation and high-stakes digital assessments. Their mission is to make secure, accessible, and stress-free online exams possible for every school.
Metrics:
- 3,600 Queries automatically handled during exam surge
- 89% AI answer accuracy
- 249 Manual support requests across two-month period
- 0 New headcount added to scale support
Quote — Alex Dore: "By creating a central, always-current and accessible knowledge base, Brainfish helped Gradeo to build user confidence and deliver scale. Schools, teachers, students and IT professionals used it intuitively to get instant answers during the most stressful part of their educational journey."

### Smokeball — Search-to-tickets cut by 74%, Support NPS up 17 points
Source: https://www.brainfishai.com/customers/how-smokeball-reduced-search-to-tickets-by-74-and-boosted-nps-by-37
Industry: Legal Software
How Smokeball shifted from reactive ticket handling to proactive in-product support, and got 15,000+ time-poor lawyers to self-serve.
Metrics:
- 30.8% → 15.3% Search-to-ticket rate (cut roughly in half)
- 67% → 83% Self-service rate
- 60 → 77 Support NPS (Training NPS 54 → 74)
- 849% ROI on the Brainfish investment
Quote — Bonny Wong: "Because of Brainfish, I think we have the best self-service hub in all of tech."

### CareMaster — Self-service from 30% to 76% with no team growth
Source: https://www.brainfishai.com/customers/transforming-cx-with-ambient-ai-agents
Industry: Healthcare / NDIS
How CareMaster's NDIS platform absorbed a 23% user-base spike, without adding a single Customer Care headcount.
Metrics:
- 30% → 76% Self-service resolution rate
- +23% User-base growth absorbed without hiring
- 80%+ Service levels held steady through growth
- +8 pts Customer ease score improvement
Quote — Elisha Anderton: "Brainfish freed our Customer Care team from handling routine queries so they could focus on solving the complex challenges our community faces that truly required human expertise."

### Smokeball — Self-serve to human rate cut by more than 50%
Source: https://www.brainfishai.com/customers/smokeball-cut-its-self-serve-to-human-rate-by-over-50
Industry: Legal Software
Smokeball Australia is an award-winning provider of productivity software for law firms that increases billings, reduces time recording, and facilitates working from anywhere, anytime.
Metrics:
- −50% Click-through rate to human support (30.8% → 15.3%)
- 80%+ Deflection rate
- 800 Human support hours saved per month
- 750% ROI on the Brainfish investment
Quote — Omer Khan: "We're super impressed with the level of support you have shown us to date, and the product's efficacy after just a few months of working together."

### Circular — 20% of weekly support hours saved
Source: https://www.brainfishai.com/customers/circular-saved-20-of-team-support-hours-per-week
Industry: Tech-as-a-Service / Sustainability
Circular is a tech subscription service. The sustainable way to get the latest devices, from the best brands
Metrics:
- 20% Weekly support hours reclaimed
- 2,112 Annual hours saved
- 275% ROI on the Brainfish investment
- Near-zero Onboarding lift to go live
Quote — Yaniv Bernstein: "All the other options needed us to spend a lot of time onboarding—time we already didn’t have! Brainfish let us leverage our existing knowledge base to increase CX efficiency with near-zero upfront effort."

### Mad Paws — 636% ROI within 5 months, without 3 net-new agents
Source: https://www.brainfishai.com/customers/how-madpaws-saw-636-roi-within-5-months
Industry: Marketplace
How Mad Paws absorbed a doubling user base with the same support team, by replacing growth-by-headcount with AI self-service.
Metrics:
- 636% ROI within 5 months (≈2-month annualised payback)
- 30,000+ Customer queries resolved in the first 4 months
- −22% Inbound ticket volume
- 3–5 → 0 Net-new support agents avoided
Quote — Justus Hammer: "Brainfish is a no-brainer. I recommend it to anyone that wants to transform their customer support."

### Clipboard — Pinpointed customer pain at school-platform scale
Source: https://www.brainfishai.com/customers/understanding-customer-pain-points-at-clipboard
Industry: EdTech / School Operations
Clipboard is a platform for schools to manage all extracurricular programs, improving the experience for students, parents and staff.
Metrics:
- 100K+ Weekly active users on the platform
- 1.5M Events powered through the platform
- 140+ Schools served
- Live User-journey analytics surfacing UX gaps
Quote — Jenny Eggimann: "These analytics are amazing! We chose you because your analytics and user journey reporting is better than anyone else."

## Featured writing

### Introducing Brainfish Live Agent Handoff for Zendesk
Source: https://www.brainfishai.com/blog/live-agent-handoff-zendesk
Published: 2026-05-21T00:00:00.000Z
Brainfish Live Agent Handoff for Zendesk transfers the full AI conversation to a human agent in the same window. No re-explaining. No starting blind.

### The Complete Guide to Brainfish + Zendesk 2026
Source: https://www.brainfishai.com/blog/complete-guide-brainfish-zendesk-2026
Published: 2026-05-18T00:00:00.000Z
The 2026 guide to running Brainfish alongside Zendesk. Architecture, integration points, Live Agent Handoff, agent assist, content ops, customer proof.

### The Dead-End Handoff: Why Most AI Support Tools Still Drop the Customer
Source: https://www.brainfishai.com/blog/the-dead-end-handoff-why-most-ai-support-tools-still-drop-the-customer
Published: 2026-05-18T00:00:00.000Z
Most AI support tools dead-end the customer when they cannot resolve. The handoff is the most damaged moment in the conversation. It does not have to be.

### The AI Knowledge Layer Buyer's Guide for Heads of Support
Source: https://www.brainfishai.com/blog/ai-knowledge-layer-buyers-guide-heads-of-support
Published: 2026-05-13T22:25:38.292Z
An AI knowledge layer buyer's guide written for the CX leader. Eight evaluation criteria, the traps to avoid, and how to score vendors against deflection and CSAT.

### Inside the 5%: How Three Companies Made Their AI Pilots Actually Deliver ROI
Source: https://www.brainfishai.com/blog/inside-the-5-how-three-companies-made-their-ai-pilots-actually-deliver-roi
Published: 2026-05-07T00:00:00.000Z
Why do 95% of AI pilots fail to show ROI? Three case studies show the 5% that win: unified knowledge, context by cohort, and measurable support automation.

### April Launch Round-Up: Brainfish Works Where Your Team Works
Source: https://www.brainfishai.com/blog/april-launch-round-up-brainfish-works-where-your-team-works
Published: 2026-05-05T00:00:00.000Z

### 7 things support, CS, and product teams are doing with Brainfish MCP and Claude
Source: https://www.brainfishai.com/blog/7-things-support-cs-and-product-teams-are-doing-with-brainfish-mcp-and-claude
Published: 2026-04-29T00:00:00.000Z

### AI Knowledge Base for Customer Support
Source: https://www.brainfishai.com/blog/ai-knowledge-base-for-customer-support
Published: 2026-04-15T00:00:00.000Z
Learn what an AI knowledge base for customer support is, why traditional help centres fail at scale, and how to implement a self-updating knowledge layer that improves resolution rates across every channel.

### How to Build an AI Knowledge Base
Source: https://www.brainfishai.com/blog/how-to-build-an-ai-knowledge-base
Published: 2026-04-14T00:00:00.000Z
Most teams build an AI knowledge base backwards. They connect an existing help centre to an AI tool, watch the demo perform well, ship it, and then watch it quietly fail over the following months as the product evolves and the knowledge doesn't. Tickets rise. Confidence scores drop. The team blames the AI model. The model isn't the problem. The knowledge infrastructure is. Building an AI knowledge base that actually works in production, not just in demos, requires a different approach than building a traditional help centre. This guide walks through each step.

### Why Your AI Support Is Only as Good as Your Knowledge Layer
Source: https://www.brainfishai.com/blog/why-your-ai-support-is-only-as-good-as-your-knowledge-layer
Published: 2026-04-09T00:00:00.000Z
Most AI support deployments fail for knowledge reasons, not model reasons. This post lays out why AI support quality is a function of the knowledge layer behind it, what a working knowledge layer looks like, a concrete diagnostic to run on your own AI, and the 2026 moves support leaders are making as a result.

### What Is an AI Agent Knowledge Base?
Source: https://www.brainfishai.com/blog/what-is-an-ai-agent-knowledge-base
Published: 2026-04-03T00:00:00.000Z
What Is an AI Agent Knowledge Base? Quick answer: An AI agent knowledge base is a structured, machine-readable repository that an AI agent queries in real time to answer questions and execute tasks. It differs from a traditional knowledge base in that it is designed for programmatic retrieval — not human browsing — and must stay automatically current as the underlying product changes.AI agents are only as good as the knowledge they can access. Build a flawed knowledge base and your AI agent hallucinates, misroutes, or confidently answers the wrong question. Build a strong one, and your agent resolves issues end-to-end without human intervention.

### AI Knowledge Base vs Traditional Knowledge Base
Source: https://www.brainfishai.com/blog/ai-knowledge-base-vs-traditional-knowledge-base
Published: 2026-04-01T00:00:00.000Z
Traditional knowledge bases are human-written and quickly go stale as products change. AI knowledge bases ingest information from multiple sources, keep answers fresh, and power semantic retrieval for AI agents and chatbots. This guide breaks down the differences, when each approach fits, and what to consider if you’re migrating.

### Retrieval Observability: Seeing the Full Chain Behind Every AI Answer
Source: https://www.brainfishai.com/blog/retrieval-observability-langchain
Published: 2026-03-31T00:00:00.000Z
TL;DR: Retrieval observability means seeing the entire chain of how an AI system retrieved and ranked information before generating an answer not just the final output. Without it, debugging a wrong AI answer is like debugging a SQL query by only looking at the UI result. The solution: trace every retrieval decision, make answers deterministically reproducible, and expose confidence scores so you're asking the right question, what did the model retrieve? not just what did it say?

### The Hidden Cost of RAG Maintenance: When Knowledge Pipeline Work Consumes Your Sprint
Source: https://www.brainfishai.com/blog/the-hidden-cost-of-rag-maintenance-when-knowledge-pipeline-work-consumes-your-sprint
Published: 2026-03-30T17:23:37.546Z
TL;DR: Most teams building RAG systems spend 20–30% of sprint capacity on knowledge pipeline maintenance — connector updates, doc syncing, accuracy regression testing — instead of building their actual product. For a senior engineer at $200–250k/yr fully loaded, that's $40–75k annually in pure maintenance overhead before you count the accuracy failures downstream. Auto-updating knowledge layers eliminate this tax entirely, freeing your team to focus on what they should be building.

### How to Automate Support Deflection Using AI Inside Your Product
Source: https://www.brainfishai.com/blog/how-to-automate-support-deflection-using-ai-inside-your-product
Published: 2026-03-30T00:00:00.000Z
Learn how to deflect support tickets with in-product AI , from auditing and structuring your knowledge base to choosing the right in-app touch points, setting smart escalation paths, and measuring what actually moves deflection rate.

### AI Knowledge Base: The Ultimate Guide for 2026
Source: https://www.brainfishai.com/blog/ai-knowledge-base-the-ultimate-guide-for-2026
Published: 2026-03-29T00:00:00.000Z
Brainfish’s 2026 guide explains what an AI knowledge base is, how it differs from a traditional knowledge base, and how to build one that reduces hallucinations and improves self‑serve support. An AI knowledge base is the knowledge layer under chatbots and agents: it ingests information from help centers, tickets, Slack, and release notes; structures it into AI‑ready semantic chunks with metadata and embeddings; and serves grounded answers via retrieval‑augmented generation (RAG) with feedback loops and analytics.

### Why 'We Need More Training' Is Killing Your Sales Performance — And How AI Enablement Fixes the Real Problem
Source: https://www.brainfishai.com/blog/sales-enablement-and-ai
Published: 2026-03-26T00:00:00.000Z
*The modern enablement engine isn't about producing more content. It's about delivering trusted knowledge at the exact moment it matters.* Sales leaders say it constantly: *"We need more training."* It's the go-to response when pipeline misses, demos bomb, reps struggle with basic objections. It feels controllable. You can schedule it, measure it, point to it in a QBR.

### Cut Support Back-and-Forth in Half: Brainfish Now Supports Image Attachments
Source: https://www.brainfishai.com/blog/cut-support-back-and-forth-in-half-brainfish-now-supports-image-attachments
Published: 2026-03-25T00:00:00.000Z

###  Knowledge Infrastructure for AI Agents: Why the Knowledge Layer Is the Most Important Part of Your Stack
Source: https://www.brainfishai.com/blog/knowledge-infrastructure-ai-agents
Published: 2026-03-21T00:00:00.000Z

### The Knowledge Layer API: How to Point Your RAG Pipeline at Clean, Current Knowledge
Source: https://www.brainfishai.com/blog/knowledge-layer-api-langchain
Published: 2026-03-19T21:33:49.070Z
TL;DR: Your RAG pipeline's accuracy problem isn't retrieval — it's knowledge quality. A knowledge layer API sits between your retriever and source documents, auto-syncing fragmented sources (Confluence, Notion, Slack, Drive), detecting conflicts before they reach the model, and eliminating the custom connector maintenance that kills most teams. Point your LangChain retriever at a clean endpoint instead of managing five broken pipelines.

## Webinars

### What AI Support Actually Looks Like When It Works
Source: https://www.brainfishai.com/webinars/what-ai-support-actually-looks-like-when-it-works
Most AI support deployments fail at the knowledge layer, not the model. This session shows the architecture that fixes it, from three teams that built it.
Hosts: Daniel Kimber (CEO and Co-founder @ Brainfish), Danielle Wilson (GTM @ Brainfish)
95% of enterprise AI pilots fail to deliver measurable ROI, according to MIT’s2025 State of AI in Business Report.
‍
But the teams getting real results, with lower ticket volume, higher deflection, and agents that actually hold up under pressure, aren't running better models. They built a better knowledge layer.
‍
In this session, Brainfish CEO Daniel Kimber breaks down how three organizations made AI support actually work. See what they changed, how they structured the system behind it, and why they landed in the 5% that delivered measurable results.
‍
WHAT WE'LL COVER:
The MCP era: how Claude, Cursor, and VS Code Copilot can now search, draft, audit, and update your knowledge layer, not just cite from it
The knowledge layer: what it looked like before, what they built, and why it changed everything the agent could do
The connections that mattered: scattered docs, support history, and product context, unified into something an agent could actually use
What they kept, what they cut, and what had to be rebuilt from scratch
The results: deflection rates, accuracy benchmarks, and what the team can now handle that was impossible before
Open Q&A: architecture, tool selection, build vs. buy, what to fix first, plus a live MCP setup walkthrough
‍
WHO THIS SESSION IS FOR:Heads of Support. CX leaders. Anyone who said yes to the AI investment and now has to make it work. If your agent is live and underperforming, or you're deciding what to build before you deploy, this session is where to start.

### The Modern Sales Enablement Engine: How AI Is Replacing the Playbook. 
Source: https://www.brainfishai.com/webinars/the-modern-enablement-engine-trusted-knowledge-at-the-moment-of-need
AI adoption in sales has doubled in two years. The teams pulling ahead aren't just buying tools, they're building the knowledge layer that makes AI actually work.
Hosts: Allie Kaiser (Head of Business Development, Brainfish ), Roz Greenfield (Co-founder and Chief Enablement Officer)
Everyone's Buying AI for Enablement. Almost No One's Getting the Results.
Most AI projects fail and it's usually not the models, but it's what's underneath them. Your product changes weekly. Your docs change quarterly. Critical knowledge lives in Slack threads and people's heads. And every time someone leaves, months of tribal expertise walk out with them.
AI can't fix a broken knowledge layer. But the teams building that layer first? They're seeing 35% higher win rates and reps ramping in half the time.
‍
What We'll Cover in 35 Minutes
The AI enablement landscape in 2026: what's working, what's noise, and where the market is actually heading
Why most AI enablement tools underperform: and the knowledge infrastructure gap that explains it
The capabilities driving real ROI right now: real-time coaching, AI role-play, predictive analytics, and automated content delivery
How to make what you already have AI-ready: turning docs, call recordings, and tribal knowledge into a trusted foundation AI can actually use
A practical framework for leading AI tool selection: because enablement is increasingly being asked to own this decision
‍
This session is for Enablement, RevOps, Sales Ops, and CS leaders evaluating or implementing AI.
‍
‍
‍

### Customer Innovation Day: Brainfish for Internal Productivity
Source: https://www.brainfishai.com/webinars/q3-customer-innovation
What’s new, what’s improved, and how teams are using Brainfish to work smarter together.
Hosts: Amelia Dunton (Head of Customer Success, Brainfish), Ash Ma (Head of Product, Brainfish)
Join us for an exclusive customer-only webinar, where we are walking through the latest features we’ve shipped this quarter and showing exactly how teams are using them to answer questions faster, cut context switching, and keep internal knowledge consistent, without extra manual work.
‍
According to Atlassian’sState of Teams 2025report, teams lose around25% of their workweeksimply searching for information internally — a full day each week spent tracking down knowledge instead of helping customers, closing deals, or building great software.
‍
During this session you’ll see:
What we are hearing from customers about internal productivity challenges and metrics you can track
How to use the Brainfish API to build an internal chatbot inside your ticketing system that delivers accurate product answers
How to use Brainfish Slack and the Brainfish Assist Chrome Extension to support internal teams in their daily workflows
Live Q&A
Whether you’re looking for ways to streamline internal workflows, improve cross-team alignment, or get more value out of Brainfish today, this webinar will give you actionable takeaways you can use immediately.
‍

### From Hallucinations to Certainty: The Missing Knowledge Layer for AI
Source: https://www.brainfishai.com/webinars/from-hallucinations-to-certainty
Learn how to ground AI in Real Product Knowledge and eliminate hallucinations.
Hosts: Daniel Kimber (CEO and Co-founder @ Brainfish), Danielle Wilson (GTM @ Brainfish)
AI doesn’t fail because the models are bad. It fails because the knowledge underneath it is scattered, outdated, and unowned.
As products evolve, critical knowledge lives everywhere: Slack threads, walkthrough videos, tickets, internal workflows, and in the heads of a few experts. When AI is built on this messy reality, it confidently gives wrong answers, creates inconsistent experiences, and erodes trust in your tools.
‍
In this30-minute session, see how to turn fragmented knowledge into a single, living layer that powers AI, teams, and self-service.
You’ll walk away knowing how to:
• Deliver AI experiences that are accurate, personalized, and grounded in real product context
• Automatically capture high-value knowledge from Slack, docs, videos, and workflows
• Avoid the pitfalls of traditional KBs, RAG pipelines, and AI tools that break at scale
• Build a knowledge foundation that grows as products, teams, and AI use cases expand.
‍
If you’re responsible for AI, CX, or enabling teams with AI-powered tools, this session will show you how to move from experimental AI to dependable, value-driving experiences.