# From Zaps to Answers: A Chat Interface for Your Agents | Chat4Agent Blog > A plumber built 36 Next-Gen Zaps to run a $3.8M business. Here's how Chat4Agent turns that kind of automation into a chat interface your whole team can use. Source: https://chat4agent.com/from-zaps-to-answers-a-chat-interface-for-your-agents --- # From Zaps to Answers: A Chat Interface for Your Agents *By Zoé, Backend engineer · 8 min read* A Master Plumber who calls himself "not a computer person" now runs a $3.8 million company on Zapier. His story shows that automation is no longer the hard part. Delivering answers to the people who need them is. ## The question behind every automation project Ross Hunt is an Army veteran and a Massachusetts Master Plumber. In 2022 he started 3R Plumbing & Mechanical out of his garage with one pickup truck and about $300,000 a year in revenue. This year the company is on track for nearly $3.8 million, with 15 employees, a shop in Norwell, and customers in about 21 South Shore towns. This story comes from a customer case study by Rob Ayre on the Zapier blog, *How 3R Plumbing & Mechanical runs on Zapier and Next-Gen Zaps*. What stands out is not the size of the setup. It is the question that drove it: > "How do I get this information without having to go look it up?" — Ross Hunt, founder and owner, 3R Plumbing & Mechanical If you build AI agents and automations, whether you are a no-coder, a developer or an agency, that question should sound familiar. It is the question your users, your team and your clients ask every day. And it is exactly where most automation projects stall: the workflows run, but people still can't easily get what they need from them. This article looks at what 3R built, where the friction remains, and how [Chat4Agent](https://chat4agent.com) gives your AI agents the interface they need to deliver value to real people. ## What 3R built: an impressive back end Ross started with classic Zaps in late 2022: new hires flowing from a Google Form into DocuSign offer paperwork, lead and form data moving between apps. As the company grew, so did the automations, until 3R was regularly exceeding its monthly task limit. Then he joined Zapier's Next-Gen Zaps early access program. His method is simple: he describes what he needs to Claude, connected to Zapier through Zapier MCP, and Claude writes and publishes the workflow. Ross reviews the logic, tests it and checks the data lands in the right place. Today, **36 Next-Gen Zaps are live, running about 1,400 times a week**, across **30 connected apps** led by ServiceTitan. His architecture rule is clear: > "ServiceTitan is the database. Zapier is the nervous system." On top of that foundation, scheduled Zaps feed a portal for each role: - **CSR portal**: a call-back queue of abandoned calls, unbooked leads and estimates needing follow-up, refreshed every 15 minutes. - **Sales portal**: every open estimate with its line items, with follow-ups pushed back into ServiceTitan. - **Warehouse and purchasing portal**: POs, receipts and returns synced with ServiceTitan. - **Marketing portal**: ad spend, calls, booked jobs, reviews and memberships by channel. - **Owner dashboard**: monthly revenue updated hourly, with a refresh button that fires a Zap on demand. Add a 6:30 a.m. weekday morning brief, a Friday report with tech grades and revenue by tech, a data-watchdog Zap, alerts when a booking stays pending for 20 minutes, and Google Ads fed with real revenue data. That is a serious operation. ## Where the friction still lives Now look closer at how people actually reach that intelligence. When Ross needs a new item in the ServiceTitan price book, he types a request into a Claude project that already knows 3R's GL accounts and labor times. In other words, **the most powerful way to use the system is conversational**. But that conversational access sits with the owner, inside his own AI tools. Everyone else gets portals, emails and alerts. Those are useful, but they are push channels and fixed views. A salesperson who wants to know one specific thing about one specific estimate still has to open a portal and dig. A technician in the field posts a PO in Google Chat because chat is the natural place to work, not a dashboard. Ross himself names the end goal in his roadmap: **one operations app** where everyone at 3R gets their answers and does their work without switching programs or waiting on the office. And he points out the obstacle most small businesses face: > "They think it's going to cost a fortune, take forever, and break the minute they touch it." That is the gap. The back end is solved. The front end, the place where a non-technical person asks a question and gets an answer, is still being stitched together. ## The cost of leaving agents without an interface If you build automations for yourself or for clients, here is what happens when the interface is an afterthought: - **Agents nobody uses.** A workflow that only the builder can trigger is a personal tool, not a business asset. Its value stays locked with one person. - **The owner becomes the bottleneck.** If only the founder can ask the AI for things, every request routes back through them. That is precisely what Ross wants to avoid by not growing the office at the same rate as revenue. - **Endless dashboard sprawl.** Each new need becomes a new portal, a new email, a new view to maintain. Five portals today, more tomorrow. - **Expensive custom front ends.** Building a proper web app for an agent means design, authentication, hosting and maintenance. For a developer, that is weeks not spent on the agent logic. For an agency, it is a cost that kills the margin on smaller clients. - **Lost clients.** For agencies especially, a client judges your work by what they can see and touch. A brilliant n8n workflow behind a raw webhook does not look like a product. In home services, as the source puts it, a missed call is a missed job. In automation, an agent without an interface is a missed opportunity. ## Chat4Agent: the front end your agents are missing [Chat4Agent](https://chat4agent.com) is built for exactly this gap. It gives AI agents built on Make.com, n8n, Zapier and similar platforms a clean, intuitive chat interface, with no technical skills required. The principle mirrors Ross's architecture. Your database stays your system of record. Your automation platform stays the nervous system. Chat4Agent becomes the face: the place where people talk to your agent in plain language. Why chat is the right interface for agents: - **It matches how people already ask questions.** Ross's own best workflow is typing a request to an AI. A chat interface extends that experience to everyone. - **One window instead of many portals.** Instead of building a new view for every role, you expose an agent that can answer many questions through a single conversation. - **Pull, not just push.** Morning briefs are great, but people also need answers at 2 p.m. on a Tuesday. Chat lets them ask when they need to. - **Professional out of the box.** Your agent gets an interface that looks like a product, not a prototype, which matters when you put it in front of a team or a client. ## What this looks like in practice Using the 3R story as inspiration, here is how a chat front end could change daily work. These are illustrations of the approach, built on the kinds of workflows described in the case study. **Before**: a CSR checks the call-back queue in a portal, then switches apps to look up the details of a specific lead. **After**: the CSR asks the agent which leads are still unbooked and gets the answer in the conversation. **Before**: a salesperson waits for the morning email to see their follow-ups, then digs through the sales portal for line items. **After**: they ask the agent about a specific estimate whenever they need it. **Before**: the owner is the only person who can talk to the AI that knows the business. **After**: each role gets access to an agent connected to the same automations, without the owner relaying requests. The underlying Zaps or scenarios do not change. What changes is who can use them, and how easily. ## Why it fits every type of builder ### For no-coders Ross taught himself everything on late nights and says plainly: "I'm not a computer person." If you recognize yourself in that, you already know how to build the workflow on Zapier, Make or n8n. With [Chat4Agent](https://chat4agent.com), you don't need to learn front-end development to put it in front of users. You stay focused on the logic that makes your business run. ### For developers Your time is best spent on agent logic, data quality and reliability, the kind of guardrails Ross built, like the cleanup Zap designed so it could never touch a sold estimate. Building and maintaining a custom chat UI is undifferentiated work. Chat4Agent lets you ship the interface quickly and get back to what makes the agent valuable. ### For agencies Stories like 3R's prove there is real demand among small businesses, including trades companies, for AI-powered operations. Ross is convinced: "If a plumber can build this, any trades owner can." But many of those owners will want someone to build it for them. Chat4Agent lets you deliver agents as polished, client-ready products, so you can offer AI agent creation as a repeatable service rather than a custom development project every time. ## Answering the usual objections **"My team already has dashboards."** Keep them. Dashboards are excellent for at-a-glance numbers. A chat interface covers everything else: the specific, unplanned questions no dashboard anticipated. **"I'd rather not touch my working automations."** You don't have to rebuild anything. Chat4Agent sits in front of agents you build on the platforms you already use. **"It sounds technical."** Chat4Agent is designed for people without technical skills. If you can build a Zap, you can give it an interface. **"Is it worth it for a small business?"** 3R's goal is to grow to $5 million and beyond without growing the office at the same rate, and to keep technicians in customers' homes instead of on data entry. Getting answers to people faster, without adding headcount, is exactly the kind of leverage small businesses need. ## The lesson from 3R Ross's rule is simple: nobody at 3R should be doing work a computer can do. His automations prove that the back end of an AI-powered business is now within reach of anyone willing to learn. The next frontier, the one he is already heading toward with his single operations app, is making all that intelligence accessible to everyone, in one place, in plain language. That is the job of the interface. And it should not take weeks of development to get it right. ## Give your agents a face today You've already done the hard part: building agents and automations that work. Now let your team, your clients and your users actually talk to them. **[Try Chat4Agent for free](https://chat4agent.com)**, connect the agent you built on Make, n8n or Zapier, and get a professional chat interface up and running in minutes. **Your automations deserve to be used. Chat4Agent turns them into an interface anyone can use, with no code required.** *Source: Rob Ayre, "How 3R Plumbing & Mechanical runs on Zapier and Next-Gen Zaps", Zapier blog.* --- *Source: [How 3R Plumbing & Mechanical runs on Zapier and Next-Gen Zaps | Zapier](https://zapier.com/blog/3r-plumbing-next-gen-zaps)* --- **Your AI agent, finally has a face.** Chat4Agent turns an agent built on Make.com, n8n or Zapier into a branded chat widget, live in minutes. Free forever, no credit card: [start free](https://chat4agent.com/signup).