Hey friends, and a special welcome to new subscribers!
AI's moving fast, but most small businesses are still on the sidelines. It comes down to three things I see again and again.
"We're not a tech company"
All structure or all chaos
No collective learning system
Let’s explore:
Blindspot #1: "We're not a tech company"
Similar to how brick-and-mortar companies didn't understand why they should care about the internet. AI will be the core operating system for every business.
The question really is just whether you come to terms with that early or late.
Blindspot #2: All structure or all chaos
Most SMBs think they need enterprise-style rollouts OR they let people mess around with ChatGPT randomly. Both typically fail.
The non-deterministic nature of LLMs means you need to experiment to find what sticks. Considering how fast the technology is evolving, you can't experiment once. Implementing AI isn't a time-bound project but a mindset shift.
Blindspot #3: No collective learning system
This is the big one. Some employees discover AI tools and get excited. Others ignore them completely. Zero knowledge transfer. As a second-order effect of the point above...
The result is a very fragmented and ill-defined strategy.
Your company learns nothing as a unit. Every AI experiment stays isolated.
Solve this by creating shared repositories for prompts and context documents (see this week's tactic below, hint hint).
The window's closing
Most of your competitors are stuck in at least one of these blindspots. The ones who aren't will have built-in operational advantages that compound over time.
Pick the blindspot that's hitting your business hardest and fix it first. The math gets worse the longer you wait.
By the way, did you see the playbook we released last week? It’s for agencies (and other types of service businesses) that wants to find AI leverage. It’s based on the mental models, frameworks and tactics I’m using to implement AI in my own agency.
You can get it for free here.
Moving on to the tactic of the week:
The first AI document you need for your business:
Most teams jump into AI tools and wonder why everything sounds generic or off-brand.
The missing piece? You never told the AI what your business actually does.
It's like hiring someone and throwing them into client work without any onboarding. Of course the results are weird.
Your business context document
Create one master doc that every AI prompt gets. Include:
What you do (products, services, core value prop)
How you operate (your standards, decision-making style)
Key company info (team structure, current priorities)
That's it. Attach this to every single AI interaction and have everyone your team do the same.
Next level: Context stacking
Once your team's using the foundation doc consistently, you can layer on specific contexts:
ICP doc for marketing tasks
Brand guidelines for content
Process docs for support/ops
Marketing email = foundation doc + ICP doc + brand guidelines
Support response = foundation doc + process doc + recent feedback
…and so on.
Why this works
Without consistent context, every AI interaction is like hiring a new intern who knows nothing about your business. With proper context, it becomes more like a seasoned team member.
Most teams spend more time fixing generic AI outputs than it would take to build this foundation once.
If you have ideas you want me to write about, hit reply!
Until next week,
Martin
PS: feel free to connect with me on LinkedIn and say hi! It’s always fun to chat with new readers.
PS2: The Inside Track is all about highlighting the do-it-yourself way of adopting AI. That doesn’t work for everyone. We also do consulting if you’re more keen to explore the done-for-you path. Fill out the interest form here for a 30-min free call.