Todd Dickerson

If you sell subscriptions, decide ahead of time what you'll say to a customer whose renewal payment declines. That's a moment they could go either way. At ClickFunnels, we sent failing payments to a different processor with custom code we built ourselves. With Payments.ai, we wanted that built in from day one, so you just choose what happens when a payment fails. That means the rule can cover what the customer hears too, not only where the payment goes next. For example: if a subscription payment declines, offer them 20% off. A reason to stay can be what tips it. Take a look at what happens today when one of your renewals fails. Then decide what you'd want to happen instead.

I build AI products for a living and I use AI for almost everything I do. So this is the warning I've been giving people: AI makes it very easy to stop developing deep thinking. Like any other muscle, your brain is a use-it-or-lose-it muscle. When you hand over jobs like research and your thinking to AI, you get the answer fast, but you skip the part where your brain does the work. As a parent, this is important to me. I want my daughters fluent with AI, and shutting it off isn't the answer. I also don't want them to forget how to work through a problem on their own and keep developing their mental and reasoning skills. One of the simplest fixes for this is about 2,400 years old. It's called the Socratic method. Socrates asked his students questions, one after another, until they reasoned their way to the answer themselves. The student did the thinking and Socrates kept pointing them in the right direction. You can set AI up to work the same way. Before you start, tell it something like: "Don't give me the answer. Ask me one question at a time to help me figure it out. If I get something wrong, ask a question that shows me where my thinking broke. Only explain it directly if I'm still stuck after a few tries." The AI finds the gaps in your reasoning and keeps you moving. But the answer comes out of your own head, so it stays there after your conversation with the AI.

My wife has caught more of my business blind spots than any advisor I've had. It's not just ClickFunnels, either. When something grabs me, I go deep, and I drop the ball somewhere else to do it. When that happens, she's usually the first one to notice. I’m extremely lucky that my outside view comes from someone who loves me and wants me to win no matter what. That part's a bonus. You don't need to be married to yours though 😅. A cofounder, a friend who runs another business, a customer, or someone on your team who'll be straight with you. More people than you think can give you an honest read on your business. A lot of entrepreneurs just forget to ask.

For years, becoming a manager meant getting further from the tools and the front-line work. With AI, that distance is shrinking. Everyone, managers and directors included, is an individual contributor (IC) now. I felt that distance in my own work. I coded the first version of @clickfunnels alone, but eventually my days became more about setting direction, reviewing work, and helping the team make decisions. Today, when I have an idea for a product or an internal tool, I can work through it with AI agents and bring the team something they can use. I still have to understand the problem, make decisions, and recognize when the result isn’t good enough. But the agents handle much of the execution now. That gives me a way to contribute directly without personally writing every line of code. And it applies well beyond engineering. A marketing director can build and test a new workflow and a support manager can build a tool around a problem their team sees every day. Leadership still means setting direction and helping people do great work. But now more leaders can get close enough to the work to build alongside them, too.

My EA built an app that lets us assign him more apps to build. I’m starting to think he didn’t think this through 😂 Roberson is the first person to tell you he’s not a developer. He’s opened the terminal twice in his life, both times by accident. But he’s also been lending a hand with @overskillapp's marketing, which has somehow led to him building an “Idea Board” for the office. Naturally, I made him test it. This is quickly becoming one of my favorite parts of building @overskillapp. Watching people around the office realize they can turn their weird ideas into real software. It means more people can contribute their ideas in a form the rest of the team can actually see, use, and improve. You get fewer ideas trapped in conversations and more things people can react to. That’s the culture I want @overskillapp to create, inside our office and everywhere else. Now we just have to see whether Roberson can build the dog-groomer portal his own app assigned him 😂

My daughters know I build companies. Lately, I’ve been watching them realize they can build things too. I’ve also been watching what they’re learning from it. A few months ago, my youngest was getting frustrated with Block Blast. It was a little too complicated for her, so I said, “Let’s make your own version. You can just click buttons and cool things happen.” We told AI what she wanted. Now she has a game made just for her that she loves. I’ve definitely mentioned this before, but my oldest is also building a bathing suit line for girls who love to dress modestly. Both Ashley and I are very involved in @ClickFunnels, @overskillapp, and our other businesses. So my girls have seen the launches and the things that work. I hope they also notice how often building starts with just one small problem someone cares enough to solve. I want my girls to see me ask questions when I don’t know the answer. I want them to see me try an idea before I know if it’ll work, and change course when I learn something new. In third grade, I sold a photocopied cheat-code magazine at school because I had a stapler and a friend who could draw. Today, my daughters have tools I couldn’t even have imagined back then. If watching me build gives them anything, I hope it’s the confidence to notice something they wish existed, trust their idea enough to tell me about it, and take the first step toward making it real. What do you hope your kids remember about the way you worked?

Our support team's calendar is full of 1-1 customer calls now. A year ago it was full of tickets. Here's what changed ⬇️ For years our support ran on Fin, Intercom's AI agent. It handled the easy tickets and routed the rest to a human. Last year I decided to replace it. My chief-of-staff agent, the same one that manages memory and context across everything we build, trained a new support bot on Grok. We fed it our documentation, past tickets, our tone, and the specific edge cases Fin used to punt on. A ticket comes in exactly like before. The bot reads it, checks logs, and resolves what it can on its own. It closes tickets faster, including some that Fin had already flagged as unresolved and passed along. Whatever it can't solve still goes to a person. That's a small slice now, and it's usually something that actually needs a human. So the team got their time back, and they're spending it on the work a bot can't replicate. Getting on video with a customer, understanding what they're really struggling with, and helping them work through it. Productivity is up 3x and customers get more direct access to our people when it matters. That's the AI outcome I really care about. Not simply doing the same work faster, but creating enough capacity for people to do more valuable work.