Project Journal
I’m funding college the unglamorous way: by turning a family collection into cash and lessons learned. Instead of real estate, stocks, or crypto, my dad spent years picking up collectibles, mostly Batman, with a mix of other franchises. As his interest in storing a big collection faded, I took on the job of selling it and treating it like a data project: scan barcodes on my phone, fetch UPC details, list items, and track every sale.
What I looked for
I wanted to understand: which franchises move best, how shipping & tax affect margins, and whether release era matters (vintage vs. 90s vs. modern). To keep the data honest, I wrote a cleaner that fixes “bundle” listings: header rows with blank titles and a quantity allocate their total price, shipping, tax, and profit across the next N component items.
What the data says (so far)
- Franchise differences: Batman sells steadily with lots of mid-price items; other lines spike more but with fewer transactions. Brand strength shows up as consistency as much as peak price.
- Era matters: Older pieces tend to command higher prices and stronger margins; the late-1990s show a noticeable dip (oversupply + mass production), while some newer releases rebound with niche demand and better condition. The price pattern by release year looks more U-shaped than a bell curve.
- Price vs. profit isn’t linear: Higher sold price does not guarantee higher net profit—shipping, fees, and tax can flatten margins. Filtering outliers makes the typical relationship much clearer.
- Operational takeaway: For quick cash flow, list reliable mid-tier Batman items first; for margin, prioritize genuinely older pieces or modern niche runs in great condition. Avoid over-weight, low-margin items unless they’re bundled to reduce per-item shipping overhead.
Conclusion
Treating the collection like a small business beats guessing. A few practical rules emerged: (1) list steady sellers early to fund the rest, (2) favor older/vintage where margin persists, (3) watch shipping weight and fee drag, and (4) use bundles intelligently—but split the accounting so the data stays truthful.
Future of the Project
With over 70 more active listings being ready to be sold online, I am still stalwart in continuing my eCommerce(ish) endeavors. I hope to continue this project and update the masses with my fortune that I've acquired selling more batman collectibles then any collector, heck, any manufacturer or distributor in the world!...alas I am but a mere CS(ish) major trying to find his footing in any place that'll take me in. Until Next Time!
Stack & tools
- Data entry & source: eBay exports, Google Sheets, phone barcode scans + UPC lookup
- Cleaning: Python (
pandas) with a bundle allocator (header → components) - Formats: JSON (for charts), CSV (for Sheets/download)
- Frontend: Plotly.js for interactive charts (line, bar, box, histogram, scatter)
- Site: GitHub Pages + Jekyll, optional GitHub Actions to auto-clean data
- Versioning: Git / GitHub
Last updated: September 9, 2025