How DevDealRadar Tracks Price History and Identifies Deals

DevDealRadar uses recorded prices to put a current offer in context. The purpose is to distinguish a supported price movement from a promotional label. The resulting score is a signal for investigation, not a prediction of service quality or a reason to buy a product you do not need.

This guide describes the existing implementation. It does not claim every merchant is monitored, every promotion is captured or any product has been tested first-hand.

Where the catalog data comes from

Configured collectors import available merchant or feed data into the catalog. Collection depends on which sources are configured and reachable. Prices, billing periods and specifications are checked against the supplied data; that is not an independent benchmark or a checkout guarantee.

The scheduler runs collection when configured to do so. Successful access to a source does not guarantee that its information is complete or current. An expired offer, a price change between runs or a feed error can make a catalog entry differ from checkout. Confirm the final terms with the merchant.

Demonstration products use synthetic observations and are clearly labelled DEMO OFFER. Their destinations are examples. Those entries do not establish evidence about a real merchant and are excluded from product search indexing.

What a history row means

History begins when a product is first tracked. The application does not reconstruct prices from before that moment. It records an initial price and subsequently records a new observation when the price changes. An unchanged price does not create another row, so the most recent stored row is not necessarily the last time a collector ran.

The chart presents available history within a 90-day window. A newly tracked product may have much less history. If an older observation exists, the last known price is carried into the start of that window. The endpoint is extended to the current catalog price for display. Neither extension is a new merchant observation. A flat segment represents the last known price carried forward, not continuous checkout verification.

Our product pages distinguish the latest recorded change from a fresh price check. Newly tracked products may have very little history; lack of a recorded reduction should not be interpreted as proof that no earlier discount existed.

How a recorded discount is calculated

The recorded percentage drop compares the current catalog price with the previous stored price. An increase produces no positive discount. An advertised original price is shown separately and cannot by itself establish a recorded saving.

The lowest price is calculated from available observations. A supported historical-low flag needs more than one observation and a current price at or below the observed low. “All-time” means all observations in this database, not all prices ever offered by the merchant.

The 30-day average is time-weighted. A price that remained in effect longer contributes more than a short-lived price. When available, a prior observation is carried into the window. Missing history is not filled with an invented market price.

The five scoring inputs

The existing formula assigns up to 100 points. Specifications help you compare plans but do not directly add points to this pricing score:

• Recorded price drop: up to 40 points, using the percentage drop multiplied by 0.8.
• Supported historical low: 20 points when the historical-low condition is met.
• Price value: up to 20 points, using half the percentage reduction below the 30-day time-weighted average.
• Merchant quality: a configured input clamped between 0 and 10 points. This is not a customer-review rating or a benchmark.
• Freshness: up to 10 points, decreasing with days since the latest recorded price observation.

The formula rounds the combined score and caps it at 100. An advertised discount above 10% that lacks support from the available history triggers the suspicious-discount condition and caps the score at 39. That flag indicates limited supporting evidence, not a finding of merchant misconduct.

What the labels do and do not promise

Scores of 90 or higher are Exceptional, 75–89 Excellent, 60–74 Good, 40–59 Fair and below 40 Normal. The default qualifying-deal threshold is 60 and can be configured. These labels summarize the pricing formula; they are not review ratings.

Affiliate commissions do not enter the formula. A merchant cannot buy a higher ranking through an affiliate relationship. Because freshness uses the last stored observation, a stable price can age even if a collector has checked it again. This is a limitation of the current signal, not proof that the source is unavailable.

Use the catalog, specifications and side-by-side comparison together. Read our methodology for the concise reference, and contact us if an entry appears wrong. Include the product URL, disputed field and a supporting merchant source so the discrepancy can be investigated.