Price is not a property of a product. It is a property of a request: the region it arrives from, the currency it is paid in, the device it is viewed on, and the promotions the marketplace decides to attach that morning. Retailers call it localized pricing. Buyers who notice it call it arbitrage.
Where the spread comes from
Three mechanisms do most of the work. First, marketplaces localize by region to match what local competitors charge. Second, A/B tests change a price for a slice of traffic and revert it later. Third, automated repricing rules move a listing against a competitor’s catalog and overshoot.
None of these are errors in the sense of a bug. They are deliberate, temporary differences that exist because a global storefront is being served to local buyers. A twelve-week look at the network logs shows the same pattern again and again: the gap is real, it is reproducible, and it almost never lasts a full day.
How long a gap stays open
Duration decides whether a find is worth anything. A gap that closes in four minutes only matters to someone watching at that minute. A gap that survives an hour is already public. The useful window sits between those two, and it is measured in minutes, not days.
This is why polling frequency matters more than scraping depth. A node that reads one page every few minutes will miss most glitches. A node that reads the same page on a tight schedule catches the moment a repricing rule fires.
A price gap is not a discount. It is a difference that exists until somebody confirms it, and most of them are gone before confirmation arrives.
What a node actually reports
A report is not a screenshot and it is not a promise to buy. It is a small, structured observation: which product, which platform, what the original price was, what the discounted price is, and when it was seen. The server then compares it against history and against every report already stored.
- The identifier must be stable, so the same product is not counted twice under two spellings.
- The original price anchors the claim; without it, a low number proves nothing.
- The timestamp matters, because a find that arrives after the seller fixes the price is only history.
Where the arbitrage ends
Shipping cost, import duty, stock depth, and account restrictions each cut into the spread. A gap that looks like 40 percent on the listing can be 10 percent after delivery. Some marketplaces refuse to ship outside the region that produced the price. Others hold stock in a warehouse that serves only local orders.
The honest version of this market is narrower than the screenshots suggest. The work a node does is not finding money on the floor. It is measuring, quickly and repeatedly, how far apart two prices are before one of them moves.
Why the network needs many devices
One phone can watch a handful of listings. A network watches the same listing from many regions at once and learns whether the gap is regional, device-specific, or simply a stale cache. That comparison is the product. The find is the unit of payment, but the coverage is the reason the find exists.
Verified finds pay a tiered DASH reward. Duplicates pay nothing. Both rules exist for the same reason: the network is buying an observation it has not seen yet, not a copy of one it has.