Most struggling software projects can be rescued, but not all of them. A practical framework for deciding between rescue and rebuild, written for non-technical leaders who have to make the call.
A backend-for-frontend earns its keep when the UI should not know which systems supply the data. On a recent consolidation project, the upstream pipeline pivoted two weeks before contract end and the front end never noticed. That is the test.
The software-as-a-service model has dominated enterprise technology for twenty years. It will not collapse. But the structural advantage it has always relied on is quietly being eroded, and most valuations have not caught up with that reality.
Buying a technology business? What to evaluate in technical due diligence, what the results mean, and how to factor technical debt into your acquisition valuation.
On a junior-heavy team, code review is the highest-bandwidth way to transfer architectural intent. The trick is slowing it down, writing the reasoning rather than the verdict, and asking questions instead of issuing corrections. Here is how I ran it on a recent engagement.
Technology governance fails not because non-technical leaders lack information, but because the information is framed in the wrong language. A practical framework for holding a technology team accountable without needing to read the code.
Greenfield discovery used to mean weeks of read-only research before anything ran. With Claude Code feeding off the kickoff transcript and a deliberately stupid mock layer, I had a runnable front end and back end in front of the team by the end of week one.
For decades, businesses have agonised over whether to build custom software or buy an off-the-shelf product. That decision is becoming irrelevant. Here is why the distinction is collapsing, and what it means for how you commission technology going forward.
Every AI output is a draft. Every AI session needs a brief and a verifier, defined before you start. If you cannot describe the verifier, do not start.
A practical decision guide for founders wondering whether to hire a fractional CTO. The triggers that signal you need one, the situations where you don't, and how to tell the difference.
There is now software that listens to interview questions, generates answers using AI, and displays them invisibly on the candidate's screen. Your screen sharing software cannot detect it. Here is what to do about it.
Gartner reports that only 22% of organisations are getting significant value from AI. After working inside organisations on both sides of that number, the gap is not technical. It is structural, and the evidence is hiding in plain sight.
