What was the result?
self-reportedHjalmar Gislason says his small startup GRID "cancelled $32,000/year in SaaS subscriptions," Contentful and Twilio Segment, and replaced them with tools built with Claude Code in "about two days for each replacement."
Hjalmar Gislason is the founder of GRID, a small software startup. His team was paying for two big platforms but using only part of them. They built their own simpler replacements with Claude Code that do exactly what they need. He says the homegrown tools are "obviously not as feature-rich as what they replaced. They just do exactly what we need." Before, buying made sense even if you used 20% of the features; now, he says, custom tools are "a weekend project."
Who did this?
How do I try it?
- List your biggest software subscriptions and the features you actually use.
- Pick one where you use a small part of a big tool.
- Write down exactly what the replacement must do, and nothing more.
- Build it with Claude Code in a test setup and move a copy of your data over.
- Switch only after it runs cleanly for a few weeks.
Editorial prompt, not GRID's build. We pay [$] a year for [tool] but only use these features: [list]. Write a plan for a simple replacement: what it must do, what data it stores, how we move our data over, and what we lose by not having the full product. Estimate the work in days.
Make it run every week
A play works once. Turn it into a routine in three small steps:
- Run it once and check the result yourself.
- Give your AI a schedule. Tell it when to run this play (every week, or every month) and to message you with what it found before you act.
- Keep a log. One line per run: what was checked, anything flagged, what you did.
What should I watch out for?
Self-reported. GRID is a software company with developers, so two days per tool will not be typical for every business. You take on upkeep and security. Safer version: keep the old tool until the new one has run cleanly for a few weeks.
Did it work for you?
PageMotor keeps this site free.