The strategic read
One article, written for founders and CTOs. Every piece opens with three punchlines so you can read it in twenty seconds and decide if the rest is worth your time.
How to choose a software development outsourcing partner
- Ask for the detail behind the logos: what was built, results with numbers, and someone you can speak to
- Sector knowledge shows up later as fewer surprises, and how a supplier uses AI tells you whether it backs up expertise or replaces it
- Before signing, find out who manages the work, which certifications they hold and who owns the code at the end
Legacy system modernisation services: what’s actually involved
- A system is legacy if it is expensive to change, poorly documented or depends on one person to keep it running, whatever its age
- Replacing everything in one cutover concentrates the risk; a staged migration keeps the business running throughout
- Every step needs a way back, and go-live has to fit around your operation rather than the developer’s diary
How we use AI, and why engineers still make the calls
- We use AI every day to write code, tests and documentation faster, so clients get more done for their budget
- Supply chain runs on exceptions that live in experienced engineers’ heads, not in training data
- Everything AI produces is reviewed by an engineer who knows the domain before it ships
When your developer leaves, what goes with them?
- When a key developer hands in their notice, the code stays, but years of knowledge about why it works the way it does often goes with them
- Developer tenure in the industry averages two to three years; in-house teams have less continuity than they appear to
- A well-run long-term development partnership holds knowledge at team level, not in individuals. When someone leaves, the work continues without a gap
The system that runs the business (and nobody wants to touch)
- Legacy systems rarely fail outright, they accumulate workarounds until making any change becomes too risky and too expensive to attempt
- AI tools can read and document a legacy codebase in a fraction of the time, which makes a lower-risk approach possible
- Every year you wait makes it harder: technical debt compounds, and the people who understand the old system gradually move on