Two firms, same starting point
This is the week that matters most in the series, because it forces the real question. Same firm, two futures. Which one are you building.
I know two firms that were basically identical. Twelve staff each. Eight years in market. Both serving the same kind of NZ and AU service business clients. Both with a solid reputation. Both profitable. Both run by an owner who was the bottleneck in their own firm and knew it.
That last part matters. Neither of them was in denial. Both could have told you exactly which parts of the week should not have needed them. Knowing was never the difference.
Eighteen months later, the two firms look nothing like each other.
Firm A, eighteen months on
The owner is still the bottleneck. Still on the tools on a Sunday night. Still approving every supplier invoice. The team is bigger, but the work has not changed. Competitors are using AI to undercut on price, so fees are under pressure. The owner is tired, the team is tired, and the business has not compounded.
Worst part, the owner knows AI was supposed to fix this. They bought some tools. They sat through some demos. Nothing stuck. The tools are mostly idle, and the ones that are running are running without governance, so when something breaks in six months there is no documentation to fix it.
The firm is not lazy and it is not stupid. It is busy. Every quarter the system rebuild gets pushed because there is delivery to get out the door, and the delivery is what pays. That is how eighteen months goes past.
Firm B, eighteen months on
The work that never needed the owner is automated. Supplier invoices approve themselves inside a documented, governed workflow. The team is trained. AI is in production across the delivery pipeline, with the governance gates that mean it still works when it breaks. The owner got the weeks back. They are in the room only for the calls that need them. Fees held, because the value delivered held. The business compounded.
The owner is not exceptional. They are not a genius. They just put the work in, in the right order, with a system.
What Firm B actually did first
This is the part people expect to be complicated, and it is not.
They picked one process. Not a platform, not a strategy, one process that was costing real money every month. They wrote down how it actually ran, including the ugly parts nobody documents. They decided who owned it and what happened when it failed. Then, and only then, they put AI on it.
Boring. Repeatable. It is also why it held. When the tool changed underneath them nine months later, and it did, the process survived the tool. Firm A had it the other way around, so when their tool changed there was nothing underneath to hold the work up.
The single variable
How they put AI in.
Both firms had the same budget for AI tools. Both had the same talent on the team. The only difference was structure. Firm B built a system before they deployed. Firm A bought tools and tried to retrofit the system later. The system won. The tools alone lost.
Why this is a divide, not a spectrum
A spectrum lets you be in the middle and feel okay about it. A divide means you are on one side or the other.
The reason it is a divide, not a spectrum, is compounding. AI capability compounds. Reputation compounds with it. A small head start eighteen months ago is a wide gap today. There is no safe middle. Either your firm is compounding on the right side, or it is drifting toward the wrong side, and the cost of catching up grows every quarter.
That is also why “we will look at it next year” is not the neutral option it sounds like. Next year you are not starting from here. You are starting from further back, against competitors who are further ahead.
Which firm are you
If you are reading this and a quiet part of you recognised Firm A, you are not stuck. You just have not built the system yet. That is a very different problem from being behind on talent or capital, and it is a much more fixable one.
The window to choose your side is open now and will not stay open. Your competitors are making the same choice this quarter.
What to do next
Book a 15 minute call with me to work out which side you are on and what it would take to cross. If it is not a fit I will say so. The newsletter is the lighter option, and the AI Readiness Audit sits in between.
Week 3 of The AI Divide lands next week. Reality is the moat.
