Back

How We Price

Our Pricing Methodology

No Crystal Balls.
Just Data.

Every valuation FlipLens gives you is backed by real sold data from real marketplaces. We don't guess. We don't use "vibes." We use math, machine learning, and an unreasonable amount of marketplace data.

Why Most Pricing Is Garbage

Here's the dirty secret of resale: most "price guides" are based on asking prices, not sold prices. Someone lists a vintage lunch box for $500 on eBay. A price guide scrapes that listing and tells you it's worth $500. Six months later, it sells for $45. The price guide still says $500.

FlipLens only uses completed sales data. If it didn't actually sell at that price, it doesn't count. This single decision makes our valuations dramatically more accurate than anything based on listing prices.

Where Our Data Comes From

A valuation is only as good as its data. Here's what feeds our pricing engine:

Real Transaction Data

We analyze actual completed sales from multiple marketplaces — not wishful-thinking asking prices. If someone listed a Beanie Baby for $50,000 but it sold for $3.50, we know the truth.

Marketplace Aggregation

Prices from Mercari, Poshmark, Facebook Marketplace, and specialty platforms are cross-referenced to build a complete picture of real-world demand.

Trend Analysis

Our AI tracks price momentum over time. That vintage Pyrex pattern that's been climbing 15% per quarter? We see it. That fad item that peaked last month? We see that too.

Condition Modeling

A mint-in-box Star Wars figure and a played-with one live in different price universes. Our AI assesses condition from your photos and adjusts valuations accordingly.

Category Expertise

Coins, electronics, vintage clothing, collectibles, books — each category has its own pricing dynamics. Our models are trained on category-specific patterns, not generic averages.

Real-Time Updates

Prices aren't static. A celebrity mention, a TikTok trend, or a supply shortage can move markets overnight. Our data refreshes continuously so you're never working with stale numbers.

How a Scan Works (Step by Step)

From photo to profit estimate in under 10 seconds. Here's what happens behind the scenes:

1

Visual Recognition

You snap a photo. Our AI identifies the item — brand, model, edition, era, and distinguishing features. It's like having an antiques expert, a sneakerhead, and a coin dealer all looking at your photo simultaneously.

2

Market Data Retrieval

We pull recent sold data across multiple platforms. Not asking prices (those are fiction). Not one marketplace (that's a sample size of one). Actual completed transactions from everywhere that matters.

3

Statistical Analysis

We calculate low, median, and high price points using statistical modeling that accounts for outliers, condition variations, and seasonal fluctuations. That one auction where two collectors went to war? We know that's not the normal price.

4

Profit Calculation

We factor in platform fees (eBay's 13.25%, Mercari's 10%, etc.), estimated shipping costs, and your purchase price to give you actual take-home profit — not revenue that disappears into fees.

5

Demand Scoring

Every item gets a demand score from 1-10 based on sell-through rate, search volume, and listing competition. High value means nothing if nobody's buying. We tell you both.

6

Buy/Pass/Maybe Verdict

All of the above distills into a clear recommendation. BUY means the math works in your favor. PASS means walk away. MAYBE means it depends on your risk tolerance and how good you are at negotiating the purchase price down.

On Accuracy (An Honest Take)

No pricing tool is 100% accurate. Anyone who claims otherwise is selling something (probably a pricing tool). Here's what we can tell you:

±15%

Typical accuracy range for common items with strong sales data

3-Range

Low, median, and high estimates so you can plan for best and worst case

Confidence

Each scan includes a confidence level — we tell you when we're less sure

Rare or unique items with limited sales history will naturally have wider price ranges. A 1955 doubled-die penny with strong comps will get a tight estimate. Your grandma's hand-painted ceramic cat? Wider range. We're honest about uncertainty because your money is on the line.

What We Don't Do

We don't inflate prices to make you feel good about a purchase. If that vintage jacket is worth $12, we'll tell you it's worth $12.

We don't use listing prices as evidence of value. Someone can list a used sock for $10,000. That doesn't make it worth $10,000.

We don't provide certified appraisals. For insurance, estate, or legal purposes, you need a certified appraiser. We're a field tool for quick decisions, not a replacement for professional appraisal services.

We don't guarantee profits. Markets change. Condition matters. Shipping costs vary. We give you the best data available — the buying decision is yours.

Think We Got It Wrong?

Every scan result includes a "Report Inaccurate Pricing" button. If you know the market better than our AI (and for niche items, you might), tell us. Every report is reviewed and used to improve our models. The more feedback we get, the smarter the system becomes.

Known Limitations

We believe in transparency. Here's where our AI pricing can be less accurate:

Hyper-Niche & Regional Items

Items with very limited sales history (e.g., local pottery, regional memorabilia, one-of-a-kind art) may have wider price ranges. Our AI has fewer data points to work with for these categories.

Condition-Sensitive Items

Items where small condition differences create large price swings (e.g., graded coins, sealed vs. opened collectibles, book first editions) may need manual verification. Photos can't always capture every flaw.

Rapidly Trending Items

Items experiencing sudden viral demand or price spikes (e.g., a celebrity endorsement, TikTok trend) may not yet be reflected in sold data. Our eBay data updates frequently but can lag 24-48 hours behind market shifts.

Bundled & Lot Items

When scanning a photo of multiple items together, individual valuations are estimates. Spread & Scan works best with items spaced apart on a contrasting background.

Authentication-Dependent Items

Luxury goods, designer items, and high-value collectibles where authenticity dramatically affects price. Our AI can identify brands but cannot authenticate items — always verify authenticity independently for items over $200.

When our AI is less certain, you'll see a "low confidence" or "medium confidence" badge on your scan results, along with a warning banner. Use these as signals to do additional research before buying.

Ready to See It in Action?

Grab something off your shelf and scan it. The first 25 scans are free — no credit card, no commitment.