SKILL.md
Deal Finder
Amazon deals are a paid visibility event, not a discount. You pay a deal fee to be placed on the Deals page during a window of concentrated traffic. A deal pays off when the visibility lift outlasts the deal itself. This skill decides which deal, when, and whether the math works.
When to use this
- A seller is considering a Lightning Deal or Best Deal and is not sure it pays.
- Planning the deal calendar around Prime Day, Prime Big Deal Days, or Q4.
- A product needs a rank push and the seller is weighing deal versus coupon.
- A past deal felt expensive and the seller wants a go or no-go rule.
The framework. The Deal Worth-It Test
A deal is worth running when three things are true. miss one and skip the deal.
- The product is deal-ready. Solid rating (4.0-plus), enough reviews to convert
a cold visitor, and stock to survive a spike. A deal that floods a weak listing with traffic just buys cheap proof that it does not convert.
- The math clears. Deal price, minus referral fee, minus FBA fee, minus unit
cost, minus the deal fee spread across expected units, is not deeply negative. A small loss can be acceptable as a rank investment. a large one is not.
- The timing concentrates traffic. A deal during a major event reaches far more
shoppers per dollar of deal fee than a deal on a random Tuesday.
The deal types
| Type | Best for | Notes |
|---|---|---|
| Lightning Deal | A short, sharp spike and rank push | A few hours, a fixed fee, limited slots |
| Best Deal | Sustained visibility | Runs days to weeks, higher fee, broader reach |
| Prime-exclusive discount | Event windows, Prime members | Strong during Prime Day and Big Deal Days |
| Coupon | Always-on visibility badge | Lowest cost, see amz-coupon-strategy |
| Outlet | Clearing aged or overstocked inventory | Built for stock you need gone |
Step by step
- Collect inputs. Price, unit cost, fees, current rating and review count, stock
on hand, the goal (rank push, event sales, clearance), and any target event date.
- Run the deal-ready check. Rating, reviews, and stock. If the listing is not
deal-ready, recommend fixing that first. a deal does not fix a weak listing.
- Pick the deal type from the table, matched to the goal and timing.
- Run the math. Per-unit contribution at the deal price, with the deal fee spread
across a realistic unit estimate. State the number and whether a loss, if any, is an acceptable rank investment.
- Time it. Anchor the deal to the highest-traffic window available. an event
deal is worth far more than the same deal off-peak.
- Plan the after. A deal spikes rank. Define how to hold the gain: ads ready to
defend the new rank, stock to stay in stock, no price jump right after.
- Run the quality check, then deliver.
Output format
## Deal Plan. [product]
Goal: [goal]
Deal-ready: [yes / no, with what to fix first]
### Recommended deal
Type: [deal type] Timing: [event or date]
Deal price: [$]
### Math
Per-unit contribution at deal price (deal fee included): [$]
Verdict: [profitable / acceptable rank investment / do not run]
### Hold the gain
[post-deal plan to keep the rank]Worked example
Product at 50 USD, 4.3 stars, 600 reviews, healthy stock. Goal: rank push before Q4.
Deal-ready: yes. Recommended: a Lightning Deal timed to Prime Big Deal Days, deal price 40 USD. Math: per-unit contribution roughly 4 USD after fees and the deal fee spread, a thin but positive number. The deal is run as a rank investment. After the deal, Sponsored Products bids are raised for two weeks to defend the new rank and the price returns to 50 without a gap that would stall momentum.
Quality check
- The deal-ready check ran first. rating, reviews, stock.
- A listing that is not deal-ready is told to fix that before any deal.
- The math spreads the deal fee across a realistic unit estimate.
- Any loss is explicitly labeled an acceptable rank investment or rejected.
- The deal is timed to the highest-traffic window available.
- There is a written plan to hold the rank gain after the deal ends.
Common mistakes
- Dealing a weak listing. Pouring event traffic into a 3.8-star listing buys
proof that it does not convert and wastes the fee.
- Ignoring the deal fee. The fee can turn a "profitable" deal price into a loss.
- Off-peak deals. The same fee buys a fraction of the reach off-event.
- No after-plan. The deal spikes rank, then rank slides back because nothing was
in place to defend it.
- Stocking out mid-deal. A deal that sells out early wastes the visibility window.
Built by Jay GPT Pro
Part of Amazon Pro Skills. Production-grade skills for serious Amazon sellers. Free and open. Built by Jay Margaliot.
I share a new AI play for Amazon sellers every week, free, in my WhatsApp group. Join here: https://chat.whatsapp.com/ILX65p1yWcaIG3c9WGHpTY