An IT company gave Claude everyone. Nothing changed.
Six months, 30 developers, zero measurable improvement. The reason wasn't the AI.
Last month we looked at CALMER, which is the framework I use to transition a business from founder-dependent chaos to a self-running system.
The Architect phase of that framework relies heavily on a specific worksheet. This issue breaks down that exact worksheet, starting with a fundamental question most founders completely get backward.
The Magic bullet trap
Scroll through LinkedIn for five minutes and you’d think AI is a light switch. Build a few AI agents, hand them to your team, and watch your operational bottlenecks vanish. Most founders I talk to genuinely believe they are ready for exactly this.
The reality is a lot less glamorous.
Take a 30-person IT company that recently approached us to get “AI-ready.” They are smart, technical, and exactly the kind of team you would expect to nail this. Their strategy was the classic magic bullet plan: give every developer access to Claude, tell them to experiment, and wait for the code to get faster and cheaper.
Six months later? They had a few isolated wins, but zero consistency across the board. Some developers used it daily; others barely touched it. A few shipped code faster, but most showed nothing measurable. Every single gain lived in a silo, and none of it translated into a company-wide improvement.
When we ran them through our readiness test, the bottleneck was glaring. Even as a tech company, they didn’t have clean data or mapped processes. There was nothing reliable for an AI to act on. The fix wasn’t more AI; it was fixing their operations first.
4 stages of AI readiness
AI readiness isn’t a binary setting. It is a four-story building where each floor serves as the foundation for the next, and you cannot skip a level.
1. Standardization: Is your core process documented, or does it only exist in someone’s head? Until it is standardized, you have nothing consistent to build on. AI applied to a messy process just produces chaos at a faster rate.
2. Digitization: Is your data clean and centralized, or is it scattered across Slack DMs, spreadsheets, and emails? This is exactly where the IT company got stuck. If your data is messy and unreachable, automation has nothing to grab onto.
3. Automation: Do your repetitive tasks run on automated triggers, or are you still relying on human memory and manual copy-pasting? AI needs a structured workflow to live inside.
4. Artificial Intelligence: Only now does AI enter the picture. Think of it as an AI Sandwich where automation prepares and cleans the data (the bottom piece of bread), AI handles the nuance and judgment calls (the filling), and automation routes the final output to the right tool or person (the top piece of bread). The AI is useless without the bread.
If you are trying to jump straight into AI without the first three layers, you are trying to build a roof with no walls.
A quick 3-question audit
You don’t need a massive operational review to figure out where you are stuck. Just answer these three questions honestly:
Could a brand-new hire run your core process using only a written checklist, without asking for help? – If the answer is no because it requires “tribal knowledge,” you are at the Standardization stage.
Is there a single, definitive source of truth for your business data? – If your team has to check five different tools to find one clear answer, you are at the Digitization stage.
If someone forgets a step in your pipeline, does a system catch it, or does it just silently fail? – If a dropped ball goes unnoticed by software, you are at the Automation stage.
My rule of thumb: The first question you answer “no” to marks the floor you are currently standing on. That is where your actual work lives, and it is almost never the AI layer.
🔧 Tools & Resources
Make.com (automation): This is our go-to for connecting tools and moving clean data between them. If a workflow gets incredibly complex with heavy branching logic, we graduate the system to n8n.
Pipedrive (digitization): This is where we house the single source of truth for sales data. It works because the interface is simple enough that teams actually use it. If you need deep custom reporting, we pair it with a dedicated dashboard tool.
OpenRouter (AI): sits in the AI layer, inside the automation, not bolted on top. It does the judgment step on data that automation has already cleaned. We use it to choose the right LLM for the right tasks, and we can switch between providers.
Stop guessing, start building
If you want a clearer diagnosis, I built a free AI Readiness Test. It scores all four categories, shows you exactly which floor you are currently standing on, and gives you the single next step to move forward. It takes about eight minutes to complete.
For shorter tech updates and early previews of what I’m working on, you can also follow me over on LinkedIn.
You don’t need to have AI completely figured out today. You just need to know which floor you’re standing on.
Once you see the actual layer you’re dealing with, the next step stops being a guessing game and becomes completely obvious.
And clarity is where confidence comes from.
Build a calmer business,
– Yuri
Yuri Vonchitzki
LinkedIn · YouTube · My services
P.S. Whenever you’re ready, there are 2 ways I can help you:
1. Join FutureBrief Plus, stop building alone. Get instant access to our private group chat for troubleshooting, Q&As with me, and a living library of proven automations, SOPs, and mental models to build a business that runs without you.
2. Let us build the machine for you. If you are an operator scaling past $500k and want to remove yourself from the daily grind, Ninjabot can audit your ops and build this entire ecosystem for you.
Reach 21,000+ business founders and operators implementing AI: sponsorship details are here.