How watch dealers and appraisers pull comps and price trends by reference
Pricing a watch to trade or to appraise comes down to two questions: what have close comparables actually fetched, and which way is the reference moving? Both answers live in realized sales, and both turn on the reference number, not the model name. The connector lets you ask for either in plain language and get back sold results with the house, the date and a link on each.
Always include the reference number. “Rolex Daytona 116500LN” returns a tight, high-confidence range; a bare “Rolex Daytona” returns a wide one flagged low. Chain a price check, a model index and a trend to get level, position and direction. Separate realized from asking prices by status, never by whether a price is present.
Realized watch sales, asked for in plain language
The connector adds auction price data to your AI assistant. For watches the useful tools are price_check for a single reference, market_index for a model-level index, price_trend for direction over time and market_stats for a segment or a house. All read-only, all normalized to USD, each result linking to the sale.
From a reference to a defensible number
Lead with the reference. Price-check it, read the model index, then show the trend. The three together give you the level, where this example sits within it, and the direction of travel.
Price-check the exact reference
A fair-value range from comparable realized sales for that reference, with the comparables listed, each linking to its sale, and a confidence label. A specific reference with a tight band of recent sales reads high. A result might come back as a range in the low tens of thousands with dozens of supporting sales, rather than a single point.
Read the model-level index
A precomputed index for the model: low, 25th percentile, median, 75th percentile and high, with the sample size and a trend read. Watches are one of the covered categories, so this gives you a clean model-level band without assembling it by hand.
Show the direction of travel
Realized prices bucketed by quarter, year or month, with the count and median per period. This is the evidence for “this reference has firmed” or “this has come off its peak” in a valuation or a buying note.
Check a segment or a specific house
A realized-price summary for that maker and house: count, median, quartiles, average, range, the date span and top vendors by volume. Useful for setting a reserve or reading where a particular saleroom prices a maker.
Confidence, caveats and why the number is defensible
Rolex Daytona 116500LN, matches a tight pool and reads high. A model-only query, such as a bare Rolex Submariner, matches thousands of disparate examples, is flagged loose_key and reads low. A reference whose realized prices are genuinely spread wide is flagged wide_quality_range and capped down, which is the honest read. The practical rule: if a result reads low, add the reference or the configuration and ask again. One more caution specific to marketplace listings: realized prices and asking prices are separated by the lot status, never by whether a price is present, so filter on status when you build an index or a comp set rather than assuming a priced row is a sale.What the data does and does not cover
Keep these in view when you price from the connector.
- It reports realized auction results normalized to USD. Some marketplace sources carry asking prices for live listings; those are separated by status, so filter on status for a realized-only comp set.
- Live single-item comparables and the model index cover watches, handbags and cars. Watches are fully covered here.
- Confidence depends on how specific the request is. A model name without a reference reads low on purpose.
- Condition, dial and provenance drive watch prices and vary across lots, so read the returned detail before treating two results as identical.
- It is market evidence for pricing, not a formal appraisal or personalized investment advice.
Connect it once, then ask in plain language
Access requires a paid ALT/FNDATA API key. Get one in the app, add the connector to your assistant with the setup page for your tool, and start with a reference number.
Questions
Frequently asked questions
Why does a reference number matter so much?
What is the difference between price_check and market_index?
Does it include asking prices or only realized sales?
Can I cite these in an appraisal?
Is there a free tier?
Put a sourced number on it today.
Connect the ALT/FNDATA connector to the AI assistant you already use, and pull realized auction comparables, market statistics and price trends in plain language, each with a link to the sale.
The ALT/FNDATA connector is read-only and returns realized auction results across 20 categories from 650+ auction houses, normalized to USD, each with the auction house, sale date and a source link. It reports what items sold for at auction, not retail or private-treaty prices. Methodology · All datasets · All guides