What I have learned scoping, building, and handing over custom web applications — written
for the people paying for them as much as the people building them.
Nobody can quote a number from a one-line description — but the factors that move the number are
predictable. Here is what actually drives the price of a custom build, and how to make yours smaller.
What I write about
Mostly the parts of custom software that are decided before any code exists: how to tell whether a process is
worth automating, what makes a build expensive, why systems get abandoned six months after launch, and the
handful of technical decisions that turn out to matter far more than the stack you chose.
Two pieces I'm working on, both aimed at the decision rather than the code:
Build or buy a POS? A guide for Cambodian retailers on when an off-the-shelf system is
genuinely the better answer, and the three situations where it stops being one.
Moving off spreadsheets without losing your data. What actually goes wrong during a
migration, why the cleanup is bigger than anyone budgets for, and how to run the old and new systems side by
side so no single day decides whether it worked.
If there's something you'd rather read about — or a question you keep getting asked about custom software —
tell me and I'll write that instead.
Working on something similar?
If any of this sounds like a problem you are living with, I am happy to talk it through — whether or not it
turns into a project.