Side projects

I saw an opportunity to make dirty soda recipes more accessible through a physical product. Inspired by cocktail recipe decks, I designed a collection of cards that made discovering and recreating drink combinations simple and fun.

I designed the complete recipe deck and sourced manufacturers for both the cards and their holders. Finding the right supplier meant comparing pricing, production options, and minimum order quantities.

I ultimately found a manufacturer willing to produce an initial run of just 50 units. I calculated unit economics across manufacturing, shipping, packaging, and selling price to understand the costs and viability of the product.

The finished product launched at approximately $25 per deck, available through Etsy and my own website.

I created 52 short-form videos demonstrating recipes from the deck, filming and publishing content across TikTok and YouTube.

Rather than relying on a single format, I experimented with hooks, on-screen text, background music, and video styles. I found that strong opening hooks and seamless loops tended to perform better, while energy-drink-related content often attracted more views on TikTok.

Kazzi Soda is still available for purchase, and the business remains a work in progress.

Building Kazzi Soda taught me how to evaluate a product opportunity, manage manufacturing trade-offs, calculate unit economics, and test marketing strategies with real customers.

The next challenge is improving conversion: turning content engagement into purchases and finding a repeatable way to grow the business.

I saw an opportunity in buying and reselling LEGO sets. It was my first test of whether I could consistently find products priced below what buyers would pay. I sold 23 sets, averaging about $92 each.

I expanded into Pokémon cards by buying underpriced ones, sending the best to PSA for grading, and selling them. I had researched the market and knew that highly graded cards can command much higher prices, and I checked my cards were in excellent condition before paying to grade them.

To price each card, I looked at what the same card had recently sold for and priced mine in line with those sales. The grading bet paid off: my 16 PSA-graded cards averaged about $126 per sale, compared with about $16 for ungraded cards. The top sale was a PSA 9 Charizard for $549.99.

I then expanded to items I bought secondhand in Japan, looking for things that would sell for more in the United States than they cost in Japan. This phase was my busiest: 2024 was my highest-volume year, with 50 sales across Japanese video games, anime figures, collectibles, and more.

Tokyo Treasures taught me to validate demand with real sales data before setting a price, to decide what to buy based on what has already worked, and to keep customers happy while the product mix changes. Each new category started with a small test, and I kept the ones that worked.