# Dildo Depreciation
## How a Divorce Valuation Triggered My Most Absurdly Useful Automation Build
In January, after four years of marriage and months of long-distance negotiation, I told my wife I wanted to separate. It was a bittersweet decision. We had shared something expansive—transformative even—but certain challenges just couldn’t be resolved while still tethered together.
There was grief, yes, but also a sense of relief. I’d stopped trying to fix what couldn’t be mended. What remained was logistics—legal, tedious, often absurd. If getting married takes two pieces of paper, getting divorced takes hundreds. And California, ever thorough, insists on itemizing it all: assets, debts, income, expenses, furniture, tech, the occasional stray kitchen spoon.
This is a story about one item on that list. A single, rubberized spark that lit up my grief with laughter—and compelled me to engineer a system so precise, it could price everything from an Apple HomePod to a used dildo.
---
## The Complication
One afternoon, I finally carved out the time and emotional bandwidth to review her disclosure—the legally required financial spreadsheet where divorcing spouses list their belongings and assign each a Fair Market Value. Most of her document read like she’d clicked “Submit” before finishing the tutorial: IKEA furniture held its original retail value, there were line items for mugs and cords, and a few things seemed to defy the laws of math.
I wasn’t mad. I was...charmed, if anything. She’d never been particularly good at forms. Her inner child seemed to grab the mouse every time she opened a spreadsheet.
Still, it had to be redone. Fair Market Value is not a vibe. It’s a legal requirement.
I was deep into the sea of mismatched numbers when I found it.
---
## The Punchline Pivot
There it was: “Vibrator (joint property) — Purchase Price: $100 — Current Value: $75.”
I blinked.
I read it again.
Then, like a wine glass fracturing from too much pressure, I cracked. Full belly laugh. Alone in my office, stunned by the specificity and the misplaced optimism. Not only had she assigned a used sex toy a 75% retained value, she’d done so with the straight face of a court document.
It was magnificent. Ridiculous. And now—legally—my problem.
Because, of course, I had to respond. Which meant researching how to fairly appraise a two-year-old dildo in the State of California.
I asked Perplexity AI the question directly: “What is the proper depreciation schedule for a used dildo in a divorce filing?”
It informed me, with robotic composure, that novelty items typically depreciate 50% per year and that “there is not a robust market for secondhand sex toys.” Translation: worthless.
But this moment wasn’t just comic gold. It was catalytic.
---
## The Engineering Response
That dildo—unwitting, unblinking—became my North Star.
If I was going to respond to this absurdity with clarity, I needed a system. One that could handle absurdity *and* accuracy. Something scalable. Something sovereign.
So I built one.
**Step 1: Capture Reality**
I started with **Gemini Live**, letting it stream video from my phone’s camera as I walked through the house, narrating everything I saw. Ownership, context, sentiment. I didn’t just want data—I wanted a record of meaning. Gemini parsed each object into rows of structured information: what it was, where it lived, who it belonged to.
**Step 2: Rebuild Purchase History**
For anything I’d bought on **Amazon** (and I buy a lot on Amazon), I used a Chrome extension to pull 12 years of order data. With Amazon no longer offering exports, this hack gave me the source material for price and date of purchase.
**Step 3: Enrich with Clay**
I loaded both datasets—Gemini’s room-by-room inventory and my Amazon order history—into **Clay**, a no-code platform I’ve used since its Product Hunt debut in 2021. Clay does three things brilliantly: it *scrapes*, *enriches*, and *acts*. I set it to match observed items with historical purchases using fuzzy logic and lookup tables. Most matches were quick. The mismatches? I handled manually—still faster than a spreadsheet war.
**Step 4: Catch the Outliers**
Not everything came from Amazon. For those, I used **Make** to forward receipts from email into Gemini, which parsed the HTML and transformed it into JSON. From there, it joined the master table.
**Step 5: Appraise Everything**
Now came the hard part: assigning fair market value. I needed a prompt that would produce consistent and legally defensible outputs across wildly different items. The test object? Of course, the dildo.
If the system could gracefully evaluate a HomePod *and* a latex vibrator, I knew I’d struck gold. It passed—repeatedly. And in every run, the dildo came back: value = $0.
Perfect.
---
## Resolution and Reframing
What began as a moment of petty comedy became a triumph of emotional clarity and technical precision. I didn’t just build a spreadsheet—I built a system that could metabolize chaos, legal ambiguity, and post-marital absurdity into structured truth.
That dildo forced me to stop handwaving and start building. It demanded I bring my full self—emotional, logical, and creative—to a bureaucratic task most people simply dread.
And in the end, it gave me back my time, my sanity, and a truly impressive Airtable base.
I even bought the domain: **[DildoDepreciation.com](http://DildoDepreciation.com)**. It’s now proudly listed as a separate post-marital asset.
---
## Coda
A laughable object made sacred by the system it inspired.
—
Written in collaboration with a creative AI assistant.
Contexts
