Make it real
We create a working prototype with a stated business use case and visible implementation.
Small Tool Guides Lab
We build the thing, then show what happened.
Working prototypes, realistic checks, measurable evidence, and honest limitations for small-business technology.
How the Lab works
A build is only the beginning
Every completed report separates observed evidence from interpretation so you can judge the result yourself.
Use defined scenarios
We check the experience at documented viewports and conditions instead of relying on a polished screenshot.
Record evidence
We collect only the performance, size, behavior, or cost data the setup can actually support.
Show the limits
We publish what worked, what did not, and which questions the experiment cannot answer.
Latest experiments
Built and documented
Completed work is searchable and indexable. Future experiments are labeled before any results exist.
I Built a 3D-Like Landing Page With AI — Here's What Happened
A working, scroll-driven product page built with CSS, SVG, and a small amount of JavaScript—then tested across seven viewport sizes.
What we measure
The metric follows the question
There is no universal Lab scorecard. A visual page calls for asset size, runtime behavior, responsive checks, and accessibility. An automation test would need different evidence.
- Build and route asset size
- Responsive behavior and overflow
- Keyboard and reduced-motion behavior
- Browser errors and interaction state
- Method, date, and known limitations