Why Amazon sales move: the full diagnostic playbook

flowchart TD
  A(["📉📈  Sales moved sharply — down or up"]) --> K1["CHECK 1 — BSR relative check
your BSR %Δ vs last / compare date
same %Δ calc for each tracked competitor's BSR"]
  K1 --> K2{"Your BSR %Δ ≈
competitors' %Δ?
(moved together)"}
  K2 -->|"Yes — whole niche moved"| M(["🌊 MARKET MOVE — not the listing
category demand shift / seasonality / external event.
Down: hold strategy, don't panic-change price or listing.
Up: it's a tide, not your genius — check stock cover to ride it"])
  K2 -->|"No — your BSR diverged drastically
from competitors (either direction)"| B["CHECK 2 — discovery-driven or conversion-driven?
Pull Business Report → Detail Page Sales & Traffic by ASIN
two equal windows: current vs prior (e.g. 30d vs previous 30d)"]
  B --> C{"Compare
Sessions - Total vs Unit Session %"}

  C -->|"Sessions moved ≥25% vs norm
CVR flat (±10%)"| D["🔍 DISCOVERY-DRIVEN
visibility changed — listing itself unchanged.
Down: find what broke. Up: find what's working & feed it"]
  C -->|"Sessions moved <25% (flat)
CVR moved >10% relative"| E["🛒 CONVERSION-DRIVEN
same traffic, different buy rate.
Down: something on the page broke. Up: recent change worked — lock it in"]
  C -->|"Both moved
(sessions ≥25% AND CVR >10%)"| F{"Day-by-day:
which moved first?"}
  C -->|"Neither moved
(sessions <25% AND CVR ±10%)"| N["💲 PRICE / MIX-DRIVEN — funnel didn't move
same traffic, same buy rate, different $ per order. Check:
avg selling price change (sales $ vs units diverging) ·
units-per-order / bundle mix · coupon or deal margin ·
Buy Box share shifting sales to another seller ·
if none of these: window too short — re-run on a longer compare"]
  F -->|"Sessions led"| D
  F -->|"CVR led"| E

  D --> Q1{"1. Listing
suppressed?"}
  Q1 -->|"Yes"| R1(["🚫 Suppression is the cause. Fix the flagged
attribute / policy issue, republish, confirm searchable"])
  Q1 -->|"No"| QG{"2. Product type / item type /
GL code / category (browse node)
changed?"}
  QG -->|"Yes"| RG(["🗂️ Reclassification broke your indexing. A changed product-type,
item-type keyword, GL code, or browse node re-maps which queries you
index for AND which category's BSR you compete in — a rank 'collapse'
may really be a category shift. Diff listing attributes vs the stored
snapshot; correct the classification or open a case to re-map it"])
  QG -->|"No"| QT{"3. Title or key content
changed recently?"}
  QT -->|"Yes"| RT(["✏️ Content change reset indexing. A title edit re-indexes the
ASIN — keywords you ranked for can drop or deindex entirely, and it
looks like a mystery rank collapse. Compare stored title vs today;
if it changed near the sales break: revert or re-optimize, then verify
indexing keyword-by-keyword (SQP impressions > 0)"])
  QT -->|"No"| Q2{"4. Hijacker
on the listing?"}
  Q2 -->|"Yes"| R2(["🏴 Hijacker splitting the Buy Box.
Test buy → IP complaint / Brand Registry takedown;
reprice meanwhile to hold the box"])
  Q2 -->|"No"| Q3{"5. Buy Box
share > 90%?"}
  Q3 -->|"No"| R3(["📦 Buy Box leak. Check price vs the competing offer,
account health, fulfillment channel — reprice to reclaim"])
  Q3 -->|"Yes"| Q4{"6. Coupon
still active?"}
  Q4 -->|"No"| R4(["🏷️ Coupon expired or budget exhausted. The green
badge lifts tile CTR — its silent disappearance reads as a
discovery drop. Relaunch or accept the new baseline"])
  Q4 -->|"Yes"| Q5{"7. Two-day / Prime
delivery promise showing?"}
  Q5 -->|"No"| R5(["🚚 Delivery promise degraded. The shown delivery date
sits on the search tile — a slower promise cuts CTR & CVR.
Cause is almost always inventory placement → run check 9"])
  R5 -.-> Q7
  Q5 -->|"Yes"| Q6{"8. Total inventory ≥ ~30 days
at current daily sales?"}
  Q6 -->|"No"| R6(["📉 Low stock. Amazon dampens traffic to offers about
to strand buyers; reorder now + slow velocity deliberately
(trim PPC, lift price) rather than stock out"])
  Q6 -->|"Yes"| Q7{"9. Per-warehouse: any top-moving
region's stock at / near 0?"}
  Q7 -->|"Yes"| R7(["🗺️ Regional stockout — the hidden killer. Delivery promise is
computed per customer from the nearest FC's stock. Zero units in a
high-velocity region = no two-day promise there → badge gone for those
shoppers → CTR/CVR fall regionally → regional rank decays. Total inventory
looks fine while sales bleed. Rebalance / inbound to that region's FCs"])
  Q7 -->|"No"| QA{"10. Own tile changed:
price · discount % · rating?"}
  QA -->|"Yes"| RA(["💲⭐ Your tile got weaker. Price up, strikethrough / discount %
shrunk, or rating display ticked down (4.5 → 4.4 stars) — each cuts
CTR at the SAME rank, which reads as a discovery drop. Quantify:
which attribute changed, on what date, and vs which competitor price"])
  QA -->|"No"| QB{"11. SERP top-10 changed: new listing
priced BELOW yours above you, or a
new player took the top position?"}
  QB -->|"Yes"| RB(["🥊 Someone re-shaped the shelf. A low-price entrant or a competitor
jumping to top position absorbs clicks even when YOUR rank held.
Compare the top-10 lineup vs the compare date: who's new, their price
vs yours, their deal/coupon badges, their review count. Respond on
price/coupon — or outbid them on top-of-search placement"])
  QB -->|"No"| Q8{"12. PPC spend reduced
vs compare window?"}
  Q8 -->|"Yes"| R8(["💸 Name the exact keywords whose spend fell —
budget cap hit, bids outcompeted, campaign paused,
or dayparting gap. Restore spend on those terms first"])
  Q8 -->|"No"| Q9{"13. Organic rank: any money keyword
fallen from top 1–25 to 25+ ?
BA keyword count vs yesterday?"}
  Q9 -->|"Yes"| R9(["🔻 Organic slip — point to the exact terms: list each keyword
that left the top 25 with old → new position, and the net Brand
Analytics keywords lost vs yesterday. Rank-recovery PPC on
precisely those terms while rank is still recoverable"])
  Q9 -->|"No"| QC{"14. External-traffic signature?
Sessions ↓ · PPC spend flat (<±10%) ·
no listing change · any BRANDED keyword
(or other) dropped out of Brand Analytics?"}
  QC -->|"Yes"| RC(["🌐 External traffic dried up. A branded keyword vanishing from BA
means people stopped searching your brand — the demand source that
fed those searches stopped: influencer / TikTok / social post faded,
Google ads or email campaign paused, affiliate link died. Amazon looks
unchanged because the leak is OUTSIDE Amazon. Check external campaign
status & Amazon Attribution tags; restart the source or replace it"])
  QC -->|"No"| R10(["🧐 All fourteen gates pass — widen the compare window
and run the SQP cross-check below"])
  R10 --> G["Cross-check Search Query Performance per keyword:
impressions collapsed → rank / indexing loss (true discovery)
impressions OK but clicks ↓ → tile lost the click: main image, price, rating
clicks OK but purchases ↓ → it's actually detail-page conversion"]
  R9 -.-> G
  R8 -.-> G
  G --> BS["📝 Known blind spots — log for the record (no data source today):
· a different variation (child ASIN) showing for the keyword instead of yours
· product displayed in an upper / wrong category (browse-node change)
· top-of-search ad share & SERP layout shifts (extra sponsored slots, video / editorial widgets pushing organic down)
· badge changes: Amazon's Choice / Best Seller / Deal badge gained or lost
· strikethrough (reference) price removed by Amazon — kills the perceived discount"]
  G --> H(["✅ Down: restart / raise PPC on historical terms,
rank-recovery campaign, resolve suppression, reprice for Buy Box.
✅ Up: feed the winning lever — raise budget on the scaling
keywords, protect rank, verify stock covers the new velocity"])

  E --> E0{"CHECK 3 — did clicks move too?
PPC clicks + SQP clicks
vs compare window"}
  E0 -->|"Yes — clicks moved with sales"| E5(["Not pure conversion — the tile gained/lost the click:
main image, price shown, rating, badge.
Treat as discovery/CTR → run the SQP cross-check"])
  E5 -.-> G
  E0 -->|"No — clicks held, purchases moved
= confirmed conversion issue"| V1{"C1. Price & offer changed?
price moved · coupon gone ·
'lowest price in 30 days' basis
changed or strikethrough disappeared"}
  V1 -->|"Yes"| W1(["💲 The offer got worse on the page. Price hike, lost coupon
badge, or Amazon dropped the 30-day-lowest reference price —
the strikethrough vanishes and the perceived deal dies at identical
traffic. Compare the full offer stack vs the compare date; restore
the coupon or hold price steady 30 days to re-earn the reference"])
  V1 -->|"No"| V2{"C2. Delivery & Buy Box intact?
two-day / fast shipping showing?
Buy Box present on the page?"}
  V2 -->|"No — either missing"| W2(["🚚 Detail-page offer degraded. A slow delivery date beside the
buy button cuts CVR hard; NO Buy Box at all ('See All Buying Options')
collapses it. Causes: regional/total stockout, price outside guardrails,
account health. Fix placement or eligibility → also run discovery check 9"])
  W2 -.-> Q7
  V2 -->|"Yes"| V3{"C3. Social proof shifted?
rating down · top-10 reviews reordered ·
a negative review now on top ·
review videos changed ·
'What customers say' AI summary changed"}
  V3 -->|"Yes"| W3(["⭐ Social proof is the killer. Shoppers read the top 3 reviews —
a fresh negative rising to the top, a reshuffled top-10, a removed
review video, or a harsher 'What customers say' AI summary tanks CVR
with zero listing change on your side. Respond to the negative, drive
fresh positives, address the complaint in bullets & images"])
  V3 -->|"No"| V4{"C4. Page content changed?
main image · bullet points · description / A+ ·
item details · user guide / docs"}
  V4 -->|"Yes"| W4(["🖼️ Content regression. Diff the stored listing snapshot vs today —
image swap, edited bullets, changed A+ / description / item details. Revert whichever
change coincides with the CVR break date, then re-test"])
  V4 -->|"No"| V5{"C5. Competitor running a deal
or coupon beside you?"}
  V5 -->|"Yes"| W5(["🔥 Competitor deal siphoning buyers. Their Lightning / Best Deal
or coupon makes your unchanged page look expensive. Check their
price-drop dates (Keepa). Usually self-resolves — decide: match with a
coupon or wait it out; don't permanently reprice against a 7-day deal"])
  V5 -->|"No"| E2{"Page Views ÷ Sessions
ratio rising?"}
  E2 -->|"Yes"| E3(["🤔 Shoppers returning & comparing
→ hesitation — price issue most likely"])
  E2 -->|"No"| E4(["📄 Straight conversion loss
→ content, images, reviews, offer"])

  classDef default fill:#121722,stroke:#1c2333,color:#e8ecf4,font-size:13px
  classDef decision fill:#241d0e,stroke:#8a6d1f,color:#ffd97a
  classDef discovery fill:#0e1a2e,stroke:#3b6db0,color:#9ec5ff
  classDef conversion fill:#0f231a,stroke:#2e8f63,color:#7fe0b2
  classDef fix fill:#0f231a,stroke:#34d399,color:#a7f3d0
  classDef start fill:#2a1216,stroke:#b04a4a,color:#ffb3b3
  class C,F,E2,K2,E0,Q1,QG,QT,Q2,Q3,Q4,Q5,Q6,Q7,QA,QB,Q8,Q9,QC,V1,V2,V3,V4,V5 decision
  class D,G,E5,R10 discovery
  classDef culprit fill:#2a1216,stroke:#b04a4a,color:#ffc9c9
  class R1,RG,RT,R2,R3,R4,R5,R6,R7,RA,RB,R8,R9,RC,W1,W2,W3,W4,W5 culprit
  classDef note fill:#0b0e14,stroke:#3a4356,stroke-dasharray:5 4,color:#7e8aa3
  class BS note
  class E conversion
  class H,E3,E4 fix
  class A start
  class K1 default
  classDef market fill:#1a1330,stroke:#7c5cbf,color:#cdb8ff
  class M market
  classDef price fill:#231a0e,stroke:#b0813b,color:#ffd9a0
  class N price

flowchart LR
  A(["📉📈 Sales moved sharply"]) --> K1["BSR relative check
your BSR %Δ vs compare date ·
same calc per competitor"]
  K1 --> K2{"moved
together?"}
  K2 -->|"Yes"| M(["🌊 MARKET MOVE — not the listing
Down: hold, don't panic-reprice
Up: it's a tide — check stock cover"])
  K2 -->|"No — you diverged"| NEXT(["→ Phase 2
Funnel Split"])
  classDef default fill:#121722,stroke:#1c2333,color:#e8ecf4,font-size:13px
  classDef decision fill:#241d0e,stroke:#8a6d1f,color:#ffd97a
  classDef market fill:#1a1330,stroke:#7c5cbf,color:#cdb8ff
  classDef start fill:#2a1216,stroke:#b04a4a,color:#ffb3b3
  classDef go fill:#0e1a2e,stroke:#3b6db0,color:#9ec5ff
  class K2 decision
  class M market
  class A start
  class NEXT go
    

flowchart TD
  C{"Compare Sessions - Total
vs Unit Session %"}
  C -->|"Sessions ≥25%Δ · CVR flat"| D(["🔍 DISCOVERY
→ Phase 3 gate chain"])
  C -->|"Sessions flat · CVR >10%Δ"| E{"Clicks test: did PPC + SQP
clicks move with sales?"}
  C -->|"Both moved"| F{"day-by-day:
which moved first?"}
  C -->|"Neither moved"| N(["💲 PRICE / MIX
avg price · units-per-order ·
coupon margin · Buy Box share leak ·
else: widen the window"])
  F -->|"sessions led"| D
  F -->|"CVR led"| E
  E -->|"clicks moved too"| D2(["🖼️ Tile lost/gained the click
image · price · rating → treat as
DISCOVERY, run SQP cross-check"])
  E -->|"clicks held"| V(["🛒 CONVERSION confirmed
→ Phase 4 gate chain"])
  classDef default fill:#121722,stroke:#1c2333,color:#e8ecf4,font-size:13px
  classDef decision fill:#241d0e,stroke:#8a6d1f,color:#ffd97a
  classDef disc fill:#0e1a2e,stroke:#3b6db0,color:#9ec5ff
  classDef conv fill:#0f231a,stroke:#2e8f63,color:#7fe0b2
  classDef price fill:#231a0e,stroke:#b0813b,color:#ffd9a0
  class C,E,F decision
  class D,D2 disc
  class V conv
  class N price