Executive Summary
Every deals product faces the same trust problem on day one: how do you know a discount is real?
A "50% off" badge is a claim, not a fact. It may be 50% off a price nobody paid, off an inflated reference, or off a genuine street price — and only one of those is a deal worth alerting a user about. A deals feed that forwards every discount badge uncritically becomes a firehose of fake markdowns that erodes user trust fast. The value of clearance and deals data is not in finding discounts; discounts are everywhere. It is in distinguishing a genuine price drop from a manufactured one — and that requires a price baseline the badge itself never provides.
This report covers clearance price data scraping built for real deal detection: what a deals feed must capture, why the baseline is everything, and what the data looks like.
This report is published by Product Data Scrape. Sample figures are illustrative of structure, not a live census.
Why "Detecting a Deal" Is Harder Than It Sounds
The advertised discount is a claim. "Was 2,000, now 1,000" is a statement by the seller. Whether 2,000 was ever the real selling price is a separate, unverified question. A deals feed that trusts the "was" price inherits every inflated reference in the market.
A real deal needs a real baseline. The only way to know a drop is genuine is to compare the current price against what the product actually sold for recently — a trailing street price you captured yourself. Without that history, "50% off" is unverifiable.
Timing matters. A deal is time-sensitive. A feed that surfaces a price drop hours late surfaces an expired deal. Detection has to be fast.
Not every low price is a deal. A permanently cheap product is not "on sale." A deals feed has to distinguish a genuine markdown from a normally low price — again, a baseline question.
The Traps
Trap one: trusting the advertised "was" price
The single biggest error. The seller's reference price is unverified and often inflated. A credible deals feed computes the discount against its own captured baseline, not the badge.
Trap two: no price history
Without a trailing price series per product, genuine drops cannot be separated from fake ones or from normally low prices. The baseline is the product; the current price is only half of it.
Trap three: latency
A deal surfaced after it ends is worse than useless — it sends users to an expired offer. Detection frequency has to match how fast clearance prices move.
Trap four: no drop magnitude or depth context
A 2% dip and a genuine 45% clearance are different events. The feed has to quantify the drop against the baseline and flag depth, not just report a lower number.
What a Usable Deals Feed Captures
| Field group |
Fields |
| Identity |
product_id, store, brand, product_url |
| Current |
current_price, advertised_was_price, advertised_discount_pct |
| Baseline |
trailing_median_price, lowest_ever, price_30d_ago |
| Verified drop |
genuine_discount_pct, drop_vs_baseline, is_genuine_deal |
| Context |
stock_signal, deal_type, detected_at |
| Capture |
captured_at |
The Baseline and Verified drop groups are the entire value: they turn an unverified badge into a computed, trustworthy discount.
Sample Data: Real Deal vs Fake Deal
An illustrative comparison of two products both advertising a discount.
| Product |
Advertised |
"Was" price |
Current |
Our trailing median |
Genuine drop |
Verdict |
| Product A |
50% off |
2,000 |
1,000 |
1,050 |
−5% |
Fake — was never 2,000 |
| Product B |
30% off |
1,500 |
1,050 |
1,480 |
−29% |
Genuine clearance |
Illustrative figures.
The two products advertise similar discounts. Product A claims 50% off but its trailing median was 1,050 — the "2,000 was" price is fiction, and the real drop is 5%. Product B claims 30% off, and its trailing median (1,480) confirms the current 1,050 is a genuine 29% clearance. A deals feed that trusted badges would rank Product A higher; a feed with a baseline correctly flags A as fake and B as the real deal. That distinction is the product.
The structured record for Product B:
{
"product_id": "DEAL-ELEC-4471",
"store": "retailer_a",
"product_url": "https://.../p/4471",
"captured_at": "2026-07-15T13:00:00+05:30",
"detected_at": "2026-07-15T13:00:00+05:30",
"current": {
"current_price": 1050,
"advertised_was_price": 1500,
"advertised_discount_pct": 30
},
"baseline": {
"trailing_median_price": 1480,
"lowest_ever": 1050,
"price_30d_ago": 1499
},
"verified_drop": {
"genuine_discount_pct": 29,
"drop_vs_baseline": -430,
"is_genuine_deal": true,
"is_lowest_ever": true
},
"context": {"stock_signal": "low", "deal_type": "clearance"}
}
The is_lowest_ever flag and the baseline-verified genuine_discount_pct are exactly the fields a deals app needs to alert a user with confidence — "genuinely the lowest price we've seen" rather than "the seller says 30% off."
What the Feed Enables
Trustworthy deal alerts. Users are alerted to genuine drops, verified against a baseline, not to fake badges.
"Lowest ever" features. A real price history powers the price-tracker feature users trust most.
Ranked deals by genuine depth. Deals are sorted by real discount against baseline, not advertised percentage.
Fast, time-sensitive alerting. Drops are detected and surfaced while the deal is still live.
Who Uses Clearance and Deals Data
Deals and bargain-hunter apps surface genuine, baseline-verified discounts — the need behind "a deals/top-deals channel with real-time and past pricing."
Price-tracker products power "lowest ever" and price-drop alerts from real history.
Resellers and arbitrage tools find genuinely underpriced clearance stock quickly.
Cashback and coupon platforms validate that advertised discounts are real before promoting them.
Limitations
Genuine-deal detection depends on a trailing baseline, which requires prior capture — a brand-new product has no baseline yet. Advertised "was" prices are captured as claims, not truths. Latency is bounded by detection frequency. We capture publicly available pricing; sample figures illustrate structure rather than audited statistics.
About the Data
This report was produced using clearance price data scraping methods from Product Data Scrape. We build deals feeds across marketplaces and retailers that compute genuine discounts against a captured trailing baseline — current and advertised price, trailing median, lowest-ever, verified drop magnitude, genuine-deal flags, and deal type — detected fast enough to surface live deals.
Delivered as JSON, CSV, via API, or a webhook stream, with the baseline history that makes discounts verifiable.
Building a deals or price-tracker product? Product Data Scrape will stand up a sample deals feed on your categories, with baseline-verified discounts, so your users see real drops instead of fake badges.
Product Data Scrape — turning marketplace complexity into decision-ready data.
Singapore
EMAIL
PHONE
.webp)
