Most agency owners aren’t behind on AI. They’re behind on structure. The software is sitting there. The VEs are hired. The conference badges are in a drawer. And the agency still feels maxed out, the owner is still the bottleneck, and the monthly tech bill keeps climbing while the work doesn’t get any better. I wrote The Independent Agency Operations Manual, 2026 Edition because I lived that exact cycle and got tired of waiting for a vendor or a consultant to fix it. It’s the whole system in one book: how to run an independent agency over the next five years. Routing, metrics, handlers, governance, the people work, the platform, all of it. This isn’t a hype book about AI. It’s a book about how the agency actually runs, and how to rebuild it so it runs on structure instead of on you. Here’s a taste of a few chapters so you know what you’re walking into. Ger your copy here The reason your tech stack isn’t working Walk through most agencies and you find the same setup. A workflow designed in 1995. Technology added in 2010. AI bolted on in 2024. Five tools that almost talk to each other. Three that overlap. Two nobody opens. And a software bill that has doubled in 18 months while the actual work of the agency hasn’t moved an inch. That’s not a tech problem. That’s a structure problem. Bolting AI onto a broken process just makes the broken process faster. Chapter 2 of the manual calls this the point solution trap, and it’s where most of the money in this industry is getting burned right now. Here’s how the trap works. You feel a pain. A vendor shows up with a tool that fits that pain exactly. The demo is clean. You buy it. Six months later it’s another line item nobody opens, solving a symptom while the actual disease sits untouched. The disease is the structure. Until you fix that, every tool you buy is just a faster version of the same mess. The manual is built to get you out of the trap and into a system that holds. [LINK] Four numbers that tell you if your agency is healthy Picture a Tuesday morning. The owner sits down and doesn’t open the AMS. She opens a portal. One screen. Four numbers across the top. AIES. RHL. SES. AIU. In under 30 seconds, she knows whether the agency is healthy. She takes a note about a 2:00 carrier meeting, and the day starts. Most agencies are nowhere near that. Ask an owner how the agency is actually doing and you get a feeling, not a number. They’re flying on a speedometer when they need a GPS. Chapter 3 fixes that. It gives you a small set of modern metrics that tell you what’s really happening, instead of the legacy KPIs that stopped meaning anything years ago. The book carries 35 metrics in total, but you don’t run your morning on 35 numbers. You run it on the right 4 at the top, with the rest underneath when you need to go deeper. Get the measurement right and every other decision gets easier. That’s the spine of the whole doctrine. Stop routing work by job title Here’s the shift that changes everything: you route work by task type, not by who has the title or who happens to have time. Every task that enters your agency goes to one of four handlers. A licensed person. A VE. A bot. An agentic bot. Each one runs in one of two modes: supervised, with a human signing off, or autonomous, where the system finishes the task end to end with nobody in the middle. The math is brutal once you see it. The same certificate of insurance that costs $25 to $40 on a licensed seat costs 10 to 50 cents on a bot. That’s not a 20% savings. That’s two orders of magnitude. A bank emails asking for a mortgagee change. Same request, every time. The information is right there in the email. It doesn’t touch coverage. A mistake is recoverable. A human in the middle adds zero value and adds delay. So the agentic bot reads the email, validates the request, makes the change, and confirms back to the bank. Nobody licensed ever saw it. That’s not automation for the sake of it. That’s putting the right work on the right handler. When you stop piling unlicensed work onto licensed seats, your licensed people go back to doing what you actually pay them for. Advising. Binding. Holding accounts together. The conversations that move business forward or keep it from falling apart. The four-handler model is Chapter 4, and it’s the part readers tell me changed how they look at their entire team. One model doesn’t fit a $400K agency and a $12M agency The book sorts agencies into five revenue bands. Practice, under $500K. Emerging, $500K to $1M. Core, $1M to $3M. Growth, $3M to $10M. Enterprise, $10M and up. The band matters because the move is different at each one. A Practice agency is mostly the owner and needs the bot layer before anything else. A Core agency at $2M might have 14 people today and a real path to 5 or 7 over the next 36 months, with the freed-up labor redeployed into growth instead of walked to the door. Chapter 5 runs the deep dive on each band, including the today reality and the 2029 direction. You read your band, you get your playbook. Governance is the part nobody is teaching Once you start running agentic bots, a new question shows up. How much do you let them do on their own? That’s governance, and almost nobody in the industry is teaching it cleanly. The vendors won’t, because their job is to get you deployed. The carriers are still figuring it out themselves. The state departments are a year or two behind the technology. So it falls to you, the owner. The good news is it’s not complicated, just unfamiliar. You pick the right environment. You set the right autonomy mode for each task. Coverage-affecting work stays supervised. Narrow, recoverable, structured work like that mortgagee change can run autonomous. And you track the hours your team spends watching the bots, because that supervision time is a real cost most agencies never put on the books. Chapter 6 lays out how to govern this with confidence instead of fear. The agencies that learn it early run agentic clean. The ones that don’t end up either too scared to use the tech or using it with no controls at all. Both lose. Your AMS just got demoted This one makes people uncomfortable, so let me use my own numbers. My agency has 11 people. In a typical week, 3 of them open the AMS. I’m not one of the three. I haven’t opened my AMS in over two years. My two producers haven’t either. We write business, manage renewals, and run the whole agency without ever touching it. That’s 27% of my team using the system I’m paying full freight for. We pay roughly $4,000 a month for our AMS across 10 seats. Three of those seats belong to people who actually open it. By moving to a leaner, open system, we’re cutting that cost 40 to 60%. Dollar for dollar, that’s enough to fund a VE, or half a producer’s marketing budget, or the routing layer itself. The AMS isn’t dying. It’s sliding from the operating system to the system of record. The daily work moves to the routing layer, and the AMS quietly stores the result in the background. Chapter 9 shows you how to figure out where your agency sits on that curve and what to do about a bill you’re almost certainly overpaying. Most owners look at their AMS invoice differently the day after they read it. The honest chapter about what didn’t work Most business books only show you the wins. This one has a chapter on the failures. I named the tools I tried that didn’t pan out. The workflows I built that fell apart. The predictions from the prior edition I got wrong, including where I misjudged how fast certain pieces would move. If a manual is going to tell you to bet your agency on a new model, it owes you the honest version of where that model has already cracked. That’s Chapter 8. It’ll save you from paying for the same mistakes I already paid for. Where this is all heading The through-line of the whole book is a platform called NexOne, the orchestration system my team has spent the last two years building alongside my own agency, for agency owners. I put the build in the book on purpose. Not as a sales pitch, but because abstract ideas don’t help you if you can’t see what they look like in a real system. NexOne is the concrete version of the doctrine: the routing engine that sits in front of the AMS and does what the CRM used to do, and more. The last chapter looks five years out. What I think happens in the next 12 months, the next 36, and the next 60. The agencies that build structure now compound the lead. The ones that wait spend the same money later and end up behind. What to do about it None of this is complicated. It’s just disciplined. Start with an honest audit. Not a consultant. A notepad and a week of attention. Count how many people on your team actually open the AMS. Write down every recurring task that crosses the agency. For your top buckets, list who handles each one today: licensed, VE, or bot. Then you’ll see it. You don’t have a headcount problem. You have a routing problem. From there it’s a sequence, not a shopping spree. Measure first. Route by task. Move unlicensed work off licensed seats. Govern your bots. Reinvest the time you save into sales and growth instead of more of the same. The full sequence, the 35 metrics, the 5 revenue bands, the governance, the people work, and the platform that runs all of it are in The Independent Agency Operations Manual, 2026 Edition. No vendor pitch. No sponsored content. Just the operating manual I wish someone had handed me five years ago. It’s here. The 2026 Operations Manual.