Amazon Listing Optimizer / Audit — Score, Keyword Gaps & Rewrites 🛒

Paste an ASIN and get back a decision, not a data dump: a 0–100 audit score with a title / bullets / keywords / images breakdown, the high-volume keywords your listing is missing (with real monthly search volume), a rewritten title and bullet points inside Amazon's character limits, and the top 3 things to fix first. One dataset row per ASIN, up to 50 ASINs per run. Works with n8n, Make, Zapier and AI agents (Claude, ChatGPT) through the Apify API and MCP.

Most Amazon tools hand you keyword lists or raw scrapes and leave the thinking to you. This one answers the question a seller actually has: how good is this listing, what is it missing, and what exact copy should replace it?

---

What you get per ASIN

One dataset row per ASIN. Here is a real one, trimmed for readability:

{
  "asin": "B09B8V1LZ3",
  "marketplace": "amazon.com",
  "mode": "audit",
  "status": "ok",
  "source": "primary",

  "audit_score": 72,
  "audit_breakdown": { "title": 78, "bullets": 65, "keywords": 60, "images": 85 },

  "keyword_gaps": [
    { "keyword": "hot pot electric", "volume": 31000 },
    { "keyword": "phone case magsafe", "volume": 93000 },
    { "keyword": "google home compatible", "volume": 105000 }
  ],

  "current_title": "Electric Hot Pot, 1.5L Portable Non-Stick Cooker",
  "rewritten_title": "Electric Hot Pot 1.5L - Portable Non-Stick Cooker for Ramen, Soup & Pasta, Dual Power, Rapid Boil, Dorm & Travel",

  "current_bullets": [
    "1.5L capacity",
    "Non-stick coating"
  ],
  "rewritten_bullets": [
    "1.5L CAPACITY, COOKS FOR TWO - enough for ramen, soup, oatmeal or pasta without a stove; ideal for a dorm, office or RV.",
    "NON-STICK & EASY TO CLEAN - PFOA-free coating wipes clean in seconds; no scrubbing, no stuck-on residue."
  ],

  "top_actions": [
    "Add 'electric' and capacity to the title - 31k/mo of demand your title never names.",
    "Rewrite bullets as benefit-led lines with keywords; current bullets are 2-4 words each.",
    "Target 'google home compatible' (105k/mo) in bullets or backend if the product qualifies."
  ],

  "cached": false,
  "fetched_at": "2026-07-21T12:00:00Z"
}

Field by field:

Field	What it means
audit_score	Overall listing quality, 0–100. A single number to sort a catalogue by
audit_breakdown	The score split into title, bullets, keywords, images — so you know where the points are being lost
keyword_gaps[]	High-demand keywords your listing does not use, each with real monthly Amazon search volume (not a guess). Sorted by volume
current_title / current_bullets	What is live on the listing right now
rewritten_title	A drop-in replacement, always ≤ 200 characters (Amazon's title limit)
rewritten_bullets	Benefit-led bullet points, each ≤ 250 characters (Amazon's bullet limit)
top_actions	Exactly 3 ranked fixes — the highest-leverage changes first
source	primary or fallback — which data path answered
status	ok, or not_found when the ASIN has no live listing (still a delivered answer)
cached	true when served from the 24-hour cache at no per-row charge

Every field is present on every row. A not_found row carries the same keys with empty values, so your downstream parser never has to special-case it.

---

Two modes — full audit and lite

You choose how much to pay per ASIN with the mode field.

	audit (default)	audit_lite
Keyword gaps	From our Amazon keyword dataset, with real search volume	From your own keywords list or Amazon autocomplete (volume: 0)
Best for	Finding demand you did not know existed	Scoring & rewriting against keywords you already have
Price (FREE tier)	$0.40 / ASIN	$0.08 / ASIN
Everything else	Score, breakdown, rewrites, top-3 actions — identical	Score, breakdown, rewrites, top-3 actions — identical

Use audit when you want the tool to discover the high-volume keywords a listing is missing. Use audit_lite — about 5× cheaper — when you already have your keyword set and just want the score, the rewrites and the priority list. In lite mode the volume field is 0, because neither your list nor autocomplete carries volume; everything else is the same audit.

---

Pricing — flat per ASIN, not per keyword or per field

Event	Price (FREE tier)	When it fires
Actor start	$0.00, once per run	Run starts; covers validation and the cache check
Audit (full)	$0.40 per ASIN	Once per delivered audit row
Audit (lite)	$0.08 per ASIN	Once per delivered audit_lite row

Tiered by your Apify Store tier:

	FREE	BRONZE	SILVER	GOLD+
Audit (full)	$0.40	$0.38	$0.35	$0.32
Audit (lite)	$0.08	$0.075	$0.07	$0.06

The score, the full breakdown, every keyword gap, both rewrites and the top-3 actions are all included in that one price. There is no per-keyword fee and no per-field charge.

You are never charged for: ASINs that fail upstream · rows served from the 24-hour cache · runs that stop at your plan's charge cap before doing any paid work. Those cost the start fee only (which is $0.00).

You are charged for an ASIN that returns not_found, because "this listing does not exist / could not be read" is a delivered answer.

What it actually costs

Who	Setup	Runs/month	Monthly total
Single seller	20 ASINs audited monthly	1	$8.00
Seller, re-audit after edits	20 ASINs, twice	2	$16.00
Agency, lite scan	50 ASINs, audit_lite, weekly	4	$16.00
Catalogue sweep	50 ASINs, full audit	1	$20.00

There is no seat and no subscription. A one-off catalogue audit of 50 ASINs is a $20 line item, not a $99/month commitment.

How it compares

	This Actor	Helium 10	CopyMonkey
Price	from $0.06 / ASIN	$99/mo+ (Platinum)	$49–99/mo
Commitment	none — pay per ASIN	monthly seat	monthly seat
Audit score + breakdown	included	Listing Analyzer	✗
Keyword gaps with volume	included	Cerebro (separate)	limited
Title + bullet rewrites	included	Frankenstein/Scribbles	✓
Output	JSON / CSV / API, schedulable	dashboard-first	dashboard-first
Seats	none — API keyed to your account	per seat	per seat

If you live in a suite all day, a subscription is a reasonable purchase. If you want a listing audit in your own pipeline — a sheet, a database, an LLM report, a Slack message — paying $99/month for a UI you open twice is the expensive option.

---

How the audit is built (honest scope)

Every audit is grounded in the live listing fetched at run time and real keyword demand — nothing is invented.

• The listing is read fresh. current_title and current_bullets come from the live Amazon product page at run time. fetched_at records when.
• Keyword volumes are real, not modelled. In audit mode every keyword_gaps entry carries an actual monthly Amazon search volume (measured examples from live runs: an electric hot pot keyword at 31,000/mo, a phone-case keyword at 93,000/mo, a smart- home keyword at 105,000/mo). In audit_lite the volume is 0 by design — see above.
• The score and rewrites come from a language model reading the listing and the keyword gaps. The rewrites are validated against Amazon's limits on every row: title ≤ 200 characters, each bullet ≤ 250 characters, and always exactly 3 top_actions.
• A not_found is an answer. If an ASIN has no readable live listing, the row says status: "not_found" rather than failing silently. It is delivered and billed like any other answer.
• source: fallback. If the primary data path fails during a run, a secondary one takes over so you still get a row. The row records which path answered via source.

Scope limits in this version

• Up to 50 ASINs per run.
• audit keyword gaps carry real volume; audit_lite gaps are volume: 0.
• The score is a relative quality signal, not a ranking guarantee — use it to prioritise which listings to fix, then verify the rewrites before publishing.
• Rewrites are suggestions. Review them before you paste them into Seller Central — you know your brand voice and compliance constraints; the model does not.

---

Input

Field	Required	Default	Notes
asins	✅	—	Up to 50. 10 characters, e.g. B08N5WRWNW. Invalid ones are reported and skipped before anything is billed
mode		audit	audit (full, keyword gaps with volume) or audit_lite (cheaper, your keywords / autocomplete)
keywords		—	Up to 100. Used only in audit_lite; ignored in audit
amazonDomain		amazon.com	Marketplace the listing is scored against

Minimal input:

{
  "asins": ["B08N5WRWNW"]
}

Lite mode with your own keywords:

{
  "asins": ["B08N5WRWNW", "B09B8V1LZ3"],
  "mode": "audit_lite",
  "keywords": ["electric hot pot", "portable cooker", "ramen cooker"]
}

---

Quick start

1. Gather your ASINs

An ASIN is the 10-character identifier in any Amazon product URL:

https://www.amazon.com/dp/B08N5WRWNW
                          ^^^^^^^^^^

Add up to 50 under ASINs. Invalid entries (a full URL, an SKU, a typo) are reported and skipped — the run does not spend anything on them.

2. Pick a mode

• audit — the tool finds the high-volume keywords your listing is missing. Start here if you want discovery.
• audit_lite — about 5× cheaper. Supply your own keywords; the tool scores and rewrites against them. Start here for a bulk quality sweep of a catalogue.

3. Run, read the top 3, act

Each row's top_actions lists the three highest-leverage fixes in order. Do those first, re-run to confirm the score moved, and move to the next listing.

4. Schedule re-audits (optional)

Add an Apify Schedule — a monthly cadence matches how often listings meaningfully change. Re-audit after you edit a listing to confirm the score improved.

---

Use with AI agents & LLM workflows

The output is shaped for software, not just for reading.

n8n / Make / Zapier

1. Trigger → run this Actor with a list of your ASINs
2. Read the dataset
3. Branch on audit_score < 60 → open a task, or append top_actions to a Google Sheet

AI agents (Claude, ChatGPT) via Apify MCP

Connect the Apify MCP server and ask directly:

"Audit these 10 ASINs, then give me only the ones scoring under 60 and their top 3 fixes, sorted by how much search volume they're leaving on the table."

Because each row carries audit_score, audit_breakdown and keyword_gaps with volumes, an agent can prioritise your catalogue without you writing that logic.

LLM reporting

Feed the dataset to a model and ask for a per-listing action plan. top_actions and rewritten_* are already written for exactly that — you can lift them straight into a client deck.

---

Recipes

Run it from the API

curl -X POST "https://api.apify.com/v2/acts/doesaiknow~amazon-listing-optimizer/runs?token=$APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{ "asins": ["B08N5WRWNW", "B09B8V1LZ3"], "mode": "audit" }'

Fetch the results:

$curl "https://api.apify.com/v2/acts/doesaiknow~amazon-listing-optimizer/runs/last/dataset/items?token=$APIFY_TOKEN&format=json"

Add &format=csv for a spreadsheet, or &clean=true to drop empty fields.

Flag the listings worth fixing first

const items = await fetch(datasetUrl).then(r => r.json());

const worst = items
  .filter(r => r.status === "ok" && r.audit_score < 60)
  .sort((a, b) => a.audit_score - b.audit_score)
  .map(r => `${r.asin} scored ${r.audit_score}: ${r.top_actions[0]}`);

if (worst.length) await notifySlack(worst.join("\n"));

Total the demand you are missing

import requests

items = requests.get(DATASET_URL).json()

for row in items:
    if row["status"] != "ok":
        continue
    missed = sum(g["volume"] for g in row["keyword_gaps"])
    print(f'{row["asin"]}  score {row["audit_score"]:>3}  '
          f'missing {missed:,}/mo across {len(row["keyword_gaps"])} keywords')

---

Troubleshooting

"Skipping N invalid ASIN(s)" An ASIN is exactly 10 characters, letters and digits, found after /dp/ in a product URL. Pasted a full URL or an SKU? It is skipped before anything is spent.

A row came back not_found The ASIN has no readable live listing at that marketplace — discontinued, region-locked, or mistyped. It is a delivered answer and is billed as one.

Keyword gaps show volume: 0 You ran audit_lite. That mode uses your own keywords or autocomplete, neither of which carries search volume. Switch to audit for gaps with real monthly volume.

Results came back cached and I wanted fresh ones Identical runs within 24 hours are served from cache at no per-row charge. Wait out the 24-hour window or change the input to force a fresh audit.

The run stopped early with a charge-limit message Your Apify plan's per-run charge cap was reached. The Actor stops before doing any paid work it could not bill, so you are never charged for uncounted rows. Upgrade the plan or reduce the number of ASINs, then re-run.

---

FAQ

How is this different from Helium 10 or CopyMonkey? Price model and delivery. They sell a monthly seat and a dashboard; this sells per ASIN audited, with output landing in your systems via API, JSON or CSV. A one-off audit of 50 ASINs is about $20 here versus a $49–99/month subscription there. If you want a full suite and a UI, buy the suite — this is for people who want the audit in a pipeline.

Is the search volume Amazon's own data? It is a monthly search-volume estimate for Amazon, not a figure Amazon publishes. Use it to prioritise which keyword gaps to close, not as a precise traffic forecast.

Can I trust the rewrites enough to paste them straight in? Treat them as a strong first draft. They are always inside Amazon's character limits and built from your live listing plus real demand, but you should review brand voice and any category compliance rules before publishing. That final check is yours.

What is the difference between audit and audit_lite? audit finds the high-volume keywords your listing is missing (real volumes) and costs about 5× more. audit_lite scores and rewrites against keywords you already supply (or autocomplete), with volume: 0. The score, breakdown, rewrites and top-3 actions are identical between the two.

Do I get charged if an ASIN is invalid or not found? Invalid ASINs are skipped before any spend. A not_found ASIN is billed, because a definitive "no live listing" is a delivered answer. Upstream failures and cache hits are never billed.

How many ASINs can I audit at once? Up to 50 per run. There is no limit on how many runs you make, so split larger catalogues across several runs or scheduled tasks.

Can I export to CSV or Google Sheets? Yes. Every Apify dataset exports to CSV, JSON, Excel and more, through the UI or the API, and integrates with n8n, Make and Zapier.

Does it change my listing for me? No. It reads the public listing and returns suggested copy and actions. Applying them in Seller Central is a deliberate step you take, so nothing is edited on your account.

---

Legal & compliance

This Actor reads publicly available Amazon product pages — the same pages any visitor sees without logging in.

• No login, no account access, no gated content. Nothing behind authentication is touched. It does not connect to Seller Central and does not change your listings.
• No personal data. Output contains product listing content — titles, bullet points, keywords and a generated audit. No customer names, reviewer identities or contact details are collected.
• Public data only. Product pages and search-volume estimates are derived from public web content.
• Suggestions, not actions. Rewrites and recommendations are drafts for you to review; the Actor never publishes anything to Amazon.

You are responsible for using the output in line with Amazon's terms and any laws that apply to you, and for reviewing suggested copy before you publish it.

---

Related actors

• Amazon Keyword Research — search volume and reverse-ASIN. Find the keywords worth targeting, then audit the listings that should rank for them.
• Amazon Rank Tracker — once you have fixed a listing, track where its ASIN actually ranks for the keywords that matter.
• Amazon Bestseller Radar — watch what is winning in a category and reverse-engineer why.

---

Support & feedback

Found a bug, need another marketplace, or want a field that is not here? Open an issue on the Actor's page. Bug reports get fixed; feature requests shape the roadmap — a competitor- comparison mode and non-.com volume coverage are the two most requested so far.

If the Actor is useful, a review or bookmark on the Store page helps other sellers find it.