What can Operator do? It can run & optimize every feature across @fin_ai and the @intercom helpdesk. It finds trends & opportunities and suggests how to take advantage of them. It's your expert in the system.
Last night, Fin Labs New York hosted a conversation we've been wanting to have in public for a while.
Our VP of Customer Support, Declan Ivory, sat down with George Dilthey, Head of Support at Clay, for an honest, unscripted look at what it actually takes to use AI across the entire customer lifecycle.
A few ideas that resonated most with the room:
→ "Qualify everywhere" — Clay doesn't just qualify customers at sign-up. They've extended that logic to every touchpoint: support queue, growth strategist handoffs, even Fin conversations. Ask for a demo mid-support chat and the same qualification workflow routes you to the right person instantly.
→ Flip your metrics — George's team now treats high first contact resolution as a signal that Fin should be handling something. If it can be solved in one interaction, why is a human involved?
→ Path to 95 — Fin's own CX team is targeting 95% automation, currently at 84%. The goal isn't hitting 100%, but finding the threshold where AI handles everything it should, freeing humans for consultative, high-context customer work.
George put it well: "We're not thinking about what is this person's role versus this person's role. We're just able to help the customer in the best way possible."
The companies getting the most from AI aren't only using it to respond and resolve tickets. They're rethinking what a customer relationship looks like when an AI agent has full context and the ability to act.
Thanks to George and the Clay team for being so open about the real journey. You can find the full recording of their conversation below.
@adelbucetta Spot on. We'll be covering how to successfully launch and scale AI in our sessions throughout the week.
For those unable to attend in NY, our AI Agent Blueprint is available as a resource: https://t.co/qdQCuGFzp5
We’re off and running at Fin Labs New York.
Three days of practical, hands-on sessions on scaling AI in customer experience: automation, performance, team design, and building trust in AI Agents.
Join us and you'll leave with a clear plan for what to test, implement, or scale next.
This afternoon kicks off with our Founder Launch Studio before this evening’s meetup with Clay, exploring how to scale CX in the AI era.
We’re just getting started, see the full 3-day in-person event schedule at the link in the replies (and register if you’re in NYC this week!)
@attio replaced their contact form with a conversation, and it changed how they sell.
Instead of routing new visitors to a static form, Attio used Fin for Sales and built a real-time conversational entry point that qualifies prospects, answers complex product questions, and routes leads to the right next step. 24/7, without adding headcount.
The results after 3 months:
→ 9x increase in Messenger engagement from new leads
→ 50+ sales-qualified leads generated
→ 30+ startup program qualifications
→ One converted customer came in at 6x their average contract value
Matt Duffy, Head of New Business at Attio, saw where it was headed: "This is going to be the buying experience. More and more people are going to be doing this."
Fin isn't just a support agent. It's an AI Customer Agent – handling support, qualifying leads, and driving revenue across the entire customer journey.
Read the full story here: https://t.co/wcb9JnKRTJ
@glean, one of the fastest-growing enterprise AI companies, needed a support solution that could scale with them.
As Kat Crichton, Manager of Technical Support, put it: "We're an AI company, so we knew we needed an AI-first solution that could move fast, was easy to implement, and would scale with us."
Within three months of choosing Fin, they achieved:
- 83% chat automation rate
- 95% CSAT
- 100% Fin involvement across all chat surfaces
And the support model itself got smarter – Fin eliminated the bottleneck where everyday users had to go through administrators to get help, giving both audiences a direct path to resolution.
Read the full story from Glean at the link in the replies
The sales funnel is fundamentally different from service. You're not resolving a problem. You're engaging with a lead, learning about them, and guiding them to the next stage of a deal.
That required us to rethink the job Fin was doing from the ground up, and build Fin for Sales from scratch.
Watch Fedor Parfenov, Bethany Clark, and Ratidzo Zvirawa talk through the architecture and product decisions behind Fin's new sales role, and learn more about Fin for Sales in the replies.
At our Operator launch in San Francisco, we made something clear: the Fin platform isn't built just for AI agents; it's built for the humans on your team too.
→ Unified Insights across agents and humans.
→ Unified QA across agents and humans
(There's no other platform where you can do this.)
One platform to run your entire customer operation, exactly as it is today.
Learn more about Operator at the link in the replies.
@AvocadoMattress customers arrive with real questions: materials, certifications, firmness, what's right for their sleep. Kurt Dwiggins, Customer Experience Manager at Avocado, describes this as a spider web of questions that need to lead somewhere.
Fin for Ecommerce navigates all of it. It pulls live order statuses and delivery updates directly from Shopify, handles complex product questions in real time, and walks customers from early-stage research to a confident purchase automatically.
Fin’s results for Avocado:
→ 38% faster response times on chat
→ 67% faster response times on email
→ Fewer conversations that require human support
From "I have a question about your policies" to "I'd love to make a purchase."
Learn more about Fin for Ecommerce below.
@TrustVanta handles some of the most complex customer questions in tech: security, compliance, audit prep.
When their existing AI's resolution rate hit a ceiling, they ran a head-to-head test with 400 real conversations to find out if anything could do better.
Fin resolved 73%. Their incumbent: 49%.
They made the switch. Since going live, they've achieved:
- 71% resolution rate – above their 50% target, within months
- 96.7% CSAT – satisfaction rose as automation increased
- ~2,500 conversations resolved by Fin every month, no human required
One of their customers put it simply: "This is the first time ever that an AI chat resolved a technical issue."
But what Vanta's team is most focused on now isn't adoption. It's what comes next.
As Vanta's Director of Customer Support, Margarita Wilshire says, "We are shifting from AI adoption to AI orchestration."
Read the full story from Vanta at the link in the replies.
We recently hosted our first 2x webinar, sharing how we approach R&D acceleration at Fin.
Our hosts @darraghcurran, @brian_scanlan, @gregolsent, and special guest @clairevo committed to answering all 70+ questions in some way – so here they are, true to their word:
Last week in San Francisco, we introduced Operator: One AI agent for understanding, managing, and improving your entire customer experience.
Not a management dashboard, but a new way of working with AI at a strategic level.
Watch Eoghan set the scene below, and see the full announcement at the link in the replies.