1) DealTruth score
"86/100"
Weighted score from discount, rarity, quality, trust, and urgency.
What you get on every deal card
Every pick includes evidence, not just a red "-50%" badge.
2) Real discount
"28% below median"
Compared to observed history baseline, not list price.
3) Rarity
"Lowest in 180 days"
Shows whether this is unusually good or ordinary.
4) Confidence
"High"
Based on history depth, offer stability, and seller quality.
5) Buy/Wait signal
"~12% drop odds"
Optional short-window prediction for timing decisions.
Why this converts better than generic deal lists
Real savings filter
We normalize effective price and compare to observed baselines, so inflated anchors get filtered out.
Buyer safety layer
Prime/FBA, seller trust, and scam penalties stop low-quality listings from topping rankings.
Honest provisional labels
When 90-day history is still building, we label the score as short-window or provisional instead of overclaiming certainty.
Decision-ready outputs
Users get clear reasons to buy now or wait, improving trust, return visits, and shareability.
How to use DealTruth in 30 seconds
- 1. Start with score + confidence, not promo percent.
- 2. Validate "Real discount" against baseline mode (90-day, short-window, provisional).
- 3. Use rarity + Buy/Wait to time checkout.
- 4. Skip low-trust listings even if headline discount looks huge.
Score guide
- 80+: exceptional
- 65-79: strong buy
- 50-64: worth tracking
- Below 50: usually pass
Start browsing
Jump directly to pages where this model is applied.
Last updated: 2026-03-02