# ShowTell case study: building the thing I needed after shipping

URL: https://wonderful.so/blog/showtell-case-study
Markdown URL: https://wonderful.so/blog/showtell-case-study.md
Content type: Case study
Published: 2026-03-26
Updated: 2026-03-26
Author: Rulian at Wonderful
Asked by: Rulian
Tags: Case Study, AI, Products

## Question

ShowTell case study: building the thing I needed after shipping

## Answer

ShowTell came from a problem I kept creating for myself.

I would finish a product, and then I would immediately have a second job: make the marketing assets. Videos. Screenshots. Showcase pages. All the material required to explain what I had just built.

So I built the tool I wanted.

## The Build

I built ShowTell in March 2026 over a few days, with about twenty focused hours of work using Claude Code and Codex.

Without AI, this would have been roughly three months of full-time effort, or around $36,000 of work at my normal benchmark. That is exactly the kind of idea I probably would not have green-lit before. AI changed the math.

## Where AI Helped

AI helped most with implementation. It made it practical to move quickly through the coding work once I had the product shape clear.

The important decisions were still mine: what the product should do, what market it was for, how the workflows should fit together, and which constraints actually mattered.

## The Hard Part

The hardest engineering constraint was keeping the system deployable headlessly on Linux while still pushing the feature set far enough to be useful.

That is the kind of tradeoff I care about in client work too. Not just "can we build it," but "can we build it in a way that will survive contact with the real world?"

ShowTell is live now and getting ready for broader distribution. It is a good example of what AI changes for me: not judgment, but the number of worthwhile experiments I can afford to run. It fits closely with how I think about [AI-assisted development](/ai) and [fixed-price phases](/services/fixed-price).