FitCalcs: Testing an Idea for Online Income
How AI-assisted opportunity research and my own interest in working out led to FitCalcs, one of my first experiments in earning money online.
This article was drafted with AI assistance and reviewed by Haobin Li.
FitCalcs started with a business question: could I build something useful on the internet that might eventually earn money? It is one of my first attempts to explore that question. As of November 15, 2026, it has not earned any revenue.
That is part of the story I want to preserve. A working website is a concrete result, but it is different from a working business. FitCalcs is an experiment in moving from an opportunity suggested by research to a product people can actually use.
Finding a topic with AI-assisted research
I began by asking an AI agent to research topics that might have commercial potential. It produced several possible directions. I selected some of them and did further research before deciding what to build.
I treated those recommendations as starting points. A convincing explanation of why a topic might work does not establish that people will find a new website, use it, or generate enough revenue to sustain it. Those questions remain open after the research is finished.
Fitness had another advantage for me: I work out too. That gave me a personal connection to the subject when choosing among the topics. I could imagine using a simple calculator myself, even while the commercial opportunity was still uncertain.
That connection does not make me a medical or nutrition expert. It gives the project a reason to hold my attention while I work on the product, its explanations, and its limitations.
The business hypothesis
The intended revenue model is display advertising. The calculators are publicly accessible without a subscription or account. The hypothesis is that useful tools and supporting explanations could attract readers, and that advertising could eventually help fund the site.
There are several assumptions inside that sentence. Publishing a calculator does not guarantee that anyone will discover it. A visit does not necessarily mean the tool helped. An advertising-supported plan does not mean advertising income has begun.
For now, the honest description is that I have built a product to test those assumptions. I have not demonstrated an online income stream.
Turning the idea into a useful website
FitCalcs now brings together fitness and nutrition calculators with reference tables, guides, and methodology pages. The tools include calorie and macro estimates, running pace, and strength calculations.
The engineering challenge is consistency. The same calculation can appear in an interactive tool and in a supporting table. If those surfaces implement their math independently, a later correction can leave them disagreeing. The project keeps calculation functions in shared modules so different pages can reuse the same logic.
The explanations matter alongside the result. A formula produces an estimate under particular assumptions. Showing a precise-looking number should not hide those assumptions or turn a general calculator into a personalized recommendation. Methodology and source context are part of the product rather than an optional appendix.
The site also has English, Spanish, French, and Simplified Chinese interfaces. Shared calculation logic remains separate from translated copy and locale routes. A translated page should express the same computation clearly, rather than becoming a separate implementation of it.
What has been achieved, and what has not
The concrete result is a published website with working tools and a structure for maintaining related explanations. That is useful progress in learning how to turn a researched idea into something operational.
The commercial result is still unresolved: there has been no revenue. I should not use the existence of calculators, additional languages, or search visibility as a substitute for that outcome.
FitCalcs connects two motivations for me: trying to earn money online and building in a subject I personally use. AI-assisted research helped me choose a direction. The next part of the experiment is learning what the published product actually earns the right to become.