Account Health Triage
Account Health Rating shifts faster than ever under the 2025 expanded visibility.
Sellers see numbers move green to yellow to red and do not know which violation
matters most. This skill reads the snapshot and tells you what to fix first.
When to use this
- AHR dropped and the seller does not know why.
- Multiple violations are open and the seller does not know which to address first.
- The account is yellow or red and a suspension is plausible.
- Routine quarterly review.
The framework. The Enforcement Ladder
Not all violations weigh equally. They sit on rungs by how fast they can take the
account down, from the top rung that can suspend you on a single complaint down to
cosmetic flags. Climb the ladder from the top. fix the highest occupied rung first,
then work down.
- Rung 1. Suspension-trigger violations. IP complaints, authenticity, restricted,
safety. These can suspend a listing or account on a single bad complaint. Fix
immediately, regardless of any point value.
- Rung 2. Order-defect rate (ODR). The headline performance metric. negative
feedback, A-to-z claims, chargebacks. A single bad week pushes ODR over the
threshold.
- Rung 3. Late shipment rate and valid tracking rate. Fulfillment reliability.
Fast to move with operational changes.
- Rung 4. Cancellation rate. Seller-caused cancellations from stockouts or
pricing errors.
- Rung 5. Policy violations with point values. Each carries weight in the AHR
score. address by point cost.
- Rung 6. Listing quality issues. Lowest urgency, but cumulative weight.
Within each rung, sort by recency. fresh violations weigh more than aging ones.
A note on point values. Amazon's Account Health Rating runs on a points system, but
Amazon does not publish a fixed per-violation point table, and the displayed AHR
score behaves more like a band (green / yellow / red) than a precise tally. Any point
numbers in this skill are illustrative estimates for sequencing the work, not
official figures. use them to decide what to fix first, never as a promised score
change.
Step by step
-
Collect the snapshot. Current AHR score, list of open violations, current
performance metrics, recent listing flags.
-
Classify each item onto one of the six rungs above.
-
Identify the binding rung. Where the score is dropping from. Often one rung
accounts for most of the loss.
-
Build the remediation order. Rung 1 first, then by point cost within each
rung. Each action has its expected directional impact on the score.
-
Tie to existing skills. Suspension-trigger items route to
. Listing-suppression items route to the suppression
recovery skill. Compliance items route to
.
-
Set the review cadence. Weekly check during recovery, monthly when stable.
-
Run the quality check, then deliver.
Output format
## Account Health Triage. [account]
Current AHR: [score] Status: [green/yellow/red]
### Triage list (in order to address)
1. [rung] . [violation] . [point cost or impact] . [route to]
2. ...
### Binding rung
[which rung accounts for most of the drop]
### This week
[top 3 actions, ordered]
### Review cadence
[weekly / monthly]
Worked example
Account AHR at 180, yellow. Open: two IP complaints, an ODR at 1.2 percent, a late
shipment rate at 6 percent, three listing-quality flags.
Triage: IP complaints first (rung 1, suspension trigger), route both to suspension
appeal workflow with evidence. ODR next (rung 2), respond to the negative feedback
and dispute the open A-to-z. Late shipment rate (rung 3), shift to FBA or fix the
handling-time process. Listing-quality flags last. Any point figures here are
illustrative estimates for sequencing, not Amazon-published values. as a rough
illustration, clearing the IP complaints might recover on the order of 50 to 60
points and fixing ODR another 20 to 30, but Amazon does not publish a fixed per-
violation point table and the real weights move. use them only to order the work,
not as a promised score. The triage shows the path back to green without scattershot
action.
Quality check
- Every violation is classified onto one of the six rungs.
- Suspension-trigger items (rung 1) are first, regardless of point value.
- The binding rung is named.
- Each action has an expected directional score impact, with any point figures
flagged as illustrative estimates, not Amazon-published values.
- Items requiring other skills are routed to those skills, not duplicated.
Common mistakes
- Treating all violations equally. A minor listing-quality flag is nowhere near
as dangerous as an IP complaint that can suspend the account outright.
- Reacting to the score, not the violations. Trying to "raise the number"
without identifying which specific items are pulling it.
- Ignoring fresh violations. Recent violations weigh more than old ones in the
formula.
- No review cadence. AHR drifts. a weekly check during recovery catches new
issues before they compound.
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