Created from a single voice note with Agent Craft
98% of the top law firms on earth bought a product a customer swore…
98% of the top law firms on earth bought a product a customer swore nobody would buy. Here's how that happened. At a previous company I walked into a major law firm to demo my product to the IT director. He liked it. Genuinely liked it. Then he told me the truth as he saw it: nobody was going to buy this. The only way I'd sell a single license was to price it at the cost of their maintenance contract, because every firm was already running a competitor's product they'd bought licenses for. I said thank you. And I thought, I think you're wrong. So I priced it aggressively. Enterprise licensing or nothing. No discount to squeeze into the gap he described. That product went on to sell to 98% of the top law firms in the world and drove well over $100 million in revenue while I was there. He was dead wrong. Now, I'm not telling you to stop listening to customers. You should listen. I learn something every single time I sit across from one, and I've been talking to a lot of them lately. Customer feedback is incredibly useful. But it can also mislead you, and here's why. Your customers have their own perspective, their own constraints, their own reasons for telling you what they tell you. That IT director wasn't lying to me. From where he sat, his read was completely reasonable. His firm had already paid for something else. Why would anyone buy again? He just wasn't the one who had to decide whether the business was possible. I was. That's the part people get backwards. They treat customer feedback as instructions when it's really just input. You take it in, you weigh it, and then you make your own call. The IT director gave me real information about how firms were buying and what they'd already committed to. That was useful. What he got wrong was the conclusion he drew from it, because drawing conclusions about whether a business works was never his job. If I'd priced that product at maintenance cost like he told me to, I'd have proven him right. Instead I trusted my own read of the value and let the market settle the argument. It did. Building: AgentCraft dot ai Grow your personal brand in minutes per day with voice notes
More content from Agent Craft
- LinkedInFor years I posted the same LinkedIn garbage as everyone else. "Thrilled to announce." "Please welcome Sarah." Safe, polished, forgettable. Here's a thought that won't leave me alone. What would LinkedIn be like if every CEO posted what they were actually thinking? Not what their marketing team approved. What was really on their mind: the challenges they're facing, the strategies they're betting on, the difficulties they're overcoming. The good, the bad, and the ugly. That version of LinkedIn would be full of real stories worth reading. You'd learn something. You'd want to do business with those people, because you'd know how their mind works before you ever spoke to them. There's a name for it. Building in public. Instead we get a feed of press releases and picnic photos. "So happy to be at this picnic." Everyone playing it safe and sounding the same. I want the opposite. People being outspoken about what they're doing at work. Real insight from humans facing real problems and figuring them out, sometimes in public, sometimes together. That's what makes life interesting. It's what would make LinkedIn interesting too. So here's my honest question. What's the one thing you're actually wrestling with at work right now that you'd never post? That's the post I want to read. Building: AgentCraft dot ai Grow your personal brand in minutes per day with voice notes
- FacebookThere are two kinds of AI tools in the market right now. The first kind started as something else and had AI features added later. The second kind was built from scratch with AI as the foundation. The difference matters more than most people realize. Bolted-on AI has to work around decisions made before AI existed. Native AI tools don't carry that baggage. Agent Craft was designed this way from day one, which is why it can draw on more than 58 different tools to actually execute marketing programs, not just suggest ideas. Barrie Hadfield put it simply: "it just becomes a further extension to our work." That's the practical test for any AI tool. Does it feel like an add-on you have to manage, or does it slot into how your team already operates? If it's the former, the adoption rate will stall. People will stop using it when things get busy. The teams getting real results from AI marketing aren't the ones using the most tools. They're using tools built to work the way they work.
- BlueskyDoes trial support actually kill product-led growth? If someone from sales contacts you during a free trial, does it make you more likely to convert, or does it make you trust the product less?
- ThreadsMost "AI-first" tools are just ChatGPT with a prettier interface. Agent Craft was built in the agentic era from scratch. That's not a marketing claim. It's a different architecture entirely.
- LinkedInVolume is where most SMBs quietly lose. Not on quality. Not on ideas. On volume. Showing up consistently, across the channels where buyers actually spend time, compounds in ways that sporadic brilliance never does. That's just how distribution works. The problem is that consistent, multi-channel content used to require a team. A writer, a scheduler, someone to adapt a LinkedIn post for an email newsletter, someone else to cut it for a short-form video caption. For a small business, that's either a significant payroll line or a backlog that never gets cleared. Agent Craft was built to close that gap. The core mechanic is simple: speak a thought, and the system handles the rest, formatting, adapting, and distributing finished content across every major channel without manual reformatting in between. No prompt engineering. No marketing background required. A voice note goes in, and coordinated distribution comes out. This matters for a specific reason. The constraint for most SMB executives isn't having things to say. It's the gap between having a thought and that thought reaching an audience. That gap is where content dies. Agent Craft removes it. The operational shift is real. When content production no longer requires a dedicated workflow, the hours that once went into drafting, adapting, and scheduling get redirected toward the work that actually moves the business. For a company without a marketing department, that reallocation is the difference between having a content presence and not having one. Tens of millions of small and medium businesses are in this position right now. Every one of them wants good marketing done. Most of them don't have the headcount to do it manually at any meaningful scale. That's the market Agent Craft exists in. And the product is live.
- LinkedInThe biggest campaign flops I've seen weren't the ones that blew up. They were the ones that never needed to happen in the first place. And I bet you're guilty of it too... Here's the pattern: a campaign is running well. Consistent results, predictable performance, solid returns. Then the new quarterly plan rolls around and there's pressure to show something new. Not better optimisation. Something new. So the working campaign gets killed to fund the next one. That's not a good decision, you know it - but you do it anyway, perhaps it's even a habit now. The average campaign length right now is around four months. The campaigns people still remember, the ones that actually built brands, ran for decades. That gap isn't a coincidence. Marketers are too close to their own work. You're living inside a campaign almost every day. Of course you get tired of it. But your customers aren't seeing it the same way you are, and your boredom is not a business reason to pull the budget. It's not always the marketers fault - pressure can come from your founder or CEO. I don't know a single founder who like losing money. Not one. So perhaps the issue is that we are interpreting a request to look for more alpha, as a request to put all your eggs into one basket and shoot for the moon. Communication is a funny thing. The smarter move is to run tighter secondary pods. Smaller experimental campaigns running in parallel, testing the new idea while the control keeps running. You only scale the new one if it outperforms what's already working. T he problem is most teams don't have the infrastructure to run experimentation properly, so they default to replacing rather than testing. They take the budget from a campaign that's delivering and hand it to something completely unproven. You lose predictable results. What you gain is entirely uncertain. If a campaign is working, your job is to protect it while you test around it. Not replace it because the calendar flipped.