75:37
Amazon PPC
Crescent Kao argues that Amazon PPC only becomes profitable when sellers replace guesswork with a structured system: rigorous multi-tool keyword research segmented into separate lists, campaigns siloed by keyword list and match type for clean data, and a disciplined, metrics-ordered (impressions to CTR to conversion rate to ACoS) optimization process rather than reflexive bid-tweaking.
Doing PPC correctly can make a listing appear twice in search (sponsored + organic), doubling visibility, clicks, and sales.
Most beginners do keyword research and PPC backwards: dump every keyword from a tool into one campaign and hope Amazon figures it out.
Reverse ASIN lookup via Cerebro reverse-engineers what's already working for top competitors instead of guessing at keywords from scratch.
Cleaned keywords should be segmented into four parallel lists (Master, Main, Related, Misspelled) so each can be targeted and tested independently.
ChatGPT can be prompted to generate high-intent Amazon-search typos for a misspelled-keyword list.
Amazon ranks organic listings by sales volume, not search volume, so high-conversion 'buyer intent' keywords beat high-search-volume keywords for both listing indexing and PPC economics.
Market Availability (percent of niche sales outside the top 3 sellers) and CPR (units needed over 8 days to reach page 1) are used to judge how winnable a keyword's ranking race is.
Keyword Hunter Pro (Sellerise) surfaces keywords by actual historical sales generated, filtering out keywords that never converted.
A full keyword list is built by combining seed-keyword research with keywords 'stolen' from the top 3 related competitor listings via a Keyword Re-ranker tool, then deduplicated and manually verified against real Amazon search results.
Amazon PPC works like an auction: even a brand-new seller can outrank established best-sellers in ad placement through bidding, which matters most during the 'honeymoon period' rank boost right after launch.
Campaigns should be built as four separate automatic campaigns (Close Match, Loose Match, Substitutes, Complements) rather than one mixed automatic campaign, to keep performance data clean.
Of the three keyword match types (Exact, Broad, Phrase), the video deliberately excludes Phrase match from its campaign structure because anything it would catch is already covered by Broad match.
The video's full manual structure totals roughly nine separate campaigns (four automatic plus main/related exact, master/misspelled/suggested/sales-volume broad).
Negative keywords are added in two phases: common-sense pre-filtering before launch, then ongoing removal of non-converting search terms found in PPC reports.
Setting a campaign's daily budget artificially high increases how often Amazon shows the ad throughout the day without necessarily increasing spend, since payment only happens on click, not impression.
Campaigns should run 7-10 days before optimizing, to avoid reacting to daily sales noise rather than real trends.
Most PPC failures trace back to two overlooked root causes: not diagnosing the funnel in order (impressions/CTR/conversion rate before ACoS), and poor campaign structure that mixes keyword strategies together.
ACoS should be judged relative to actual profit margin (not against a universal low number) and analyzed last, since it's the result of everything upstream in the funnel, not the root problem.
Downloaded PPC reports get triaged by color: red keywords (ACoS far over margin, or high no-sale spend) get turned off, yellow/borderline keywords get bids lowered gradually, and green/profitable keywords get moved into their own more aggressively-funded campaign.
PPC alone isn't sufficient for a launch — converting traffic requires a parallel review-generation strategy since shoppers gravitate to listings with social proof.
Reverse ASIN Lookup (Cerebro) — A Helium 10 Cerebro method that feeds top-ranking competitor ASINs back into the tool to reveal which keywords those listings actually rank and sell on, rather than guessing keywords from scratch. Apply: Search your main product keyword on Amazon, pull the ASINs of the top organic (non-sponsored) listings in rank order, paste them into Cerebro, and click Get Keywords.
Three-Filter Cerebro Narrowing — A sequential filter set — Match Type = Organic, Competitor Rank <=60, Number of Competitors >=2 — that shrinks a raw reverse-ASIN pull of thousands of keywords down to a manageable ~200-300. Apply: Apply the three filters in order right after a reverse ASIN pull, before doing any manual clean-up of the keyword list.
Four-List Keyword Taxonomy (Master/Main/Related/Misspelled) — A segmentation scheme that splits cleaned keywords into four non-overlapping lists so each can be targeted and tested independently in separate PPC campaigns. Apply: Pass through the cleaned master list multiple times, tagging highly-relevant 'main' terms, broader 'related' terms, and natural misspellings into separate saved Helium 10 folders.
Seed-Expansion Loop — Re-running the top-volume keyword from the Main list as a direct keyword search in Cerebro (instead of reverse-ASIN) to surface additional terms the ASIN-based pull missed. Apply: After building the Main list, search its top keyword directly in Cerebro, apply the Organic filter, and merge new results into the existing lists (Helium 10 auto-dedupes).
ChatGPT Misspelling-Harvest Technique — A method of exporting the master keyword list as CSV and feeding it to ChatGPT with a specific prompt to generate common high-intent Amazon-search typos. Apply: Prompt ChatGPT with 'Give me a downloadable CSV file containing all the common Amazon search only high-intent typos for the following keyword phrases,' paste the keyword list, then strip the original-keyword column before reimporting as the Misspelled Keywords list.
Buyer-Intent / Sales-Driven Keyword Strategy — Prioritizing keywords with high conversion rate over high search volume in listing copy and PPC targeting, on the claim that Amazon ranks by sales volume, not query volume. Apply: Compare candidate keyword sets' conversion rates and choose the higher-converting (even if lower-volume) set for both listing indexing and ad targeting.
Market Availability metric — The percentage of a niche's total sales not captured by the current top 3 sellers for a given keyword, used to gauge competitive saturation. Apply: Favor keywords with high market availability (e.g., 70% or more) since a new listing won't launch in the top 3 anyway; treat low availability (e.g., 10%) as a red flag.
CPR (units-to-rank) metric — An estimate of how many units must sell over an 8-day window to reach page 1 of search for a given keyword. Apply: Compare CPR across candidate keywords/strategies to pick the cheaper, faster path to first-page organic ranking.
Keyword Hunter Pro (Sellerise) — A keyword research tool that reports actual historical sales generated per keyword (via a 'semantic core' metric), filtering to only keywords that produced organic sales. Apply: Enter a seed keyword to pull a sales-validated keyword list, then filter it by conversion rate and market availability before cleanup.
Two-Metric Cleanup Filter (conversion rate + market availability) — A screening pass that removes keywords with conversion rate below ~20% and keeps only keywords where sellers outside the top 3 capture at least ~40% of sales. Apply: Apply both filters to a raw Keyword Hunter export to shrink hundreds of keywords down to a workable, higher-quality list.
Relevance-Sort Cleanup — Sorting a keyword list by a 'relevance' score to quickly spot and remove keywords unrelated to the exact product variant being sold. Apply: Treat relevance scores of roughly 10 or below as likely-unrelated, and manually strip competitor brand names and wrong-variant terms.
Competitor Keyword Stealing (Top Related Products + Keyword Re-ranker) — Identifying the top 3 related/competitor listings via a 'Top Related Products' tab, then running each through a separate 'Keyword Re-ranker' tool to pull every keyword that listing ranks for. Apply: Merge the resulting competitor keyword spreadsheets, deduplicate (e.g., via Numbers' 'is distinct' filter), and re-apply the sales/relevance filters before combining with your own list.
Manual Amazon-Search Verification & Highlighting — A QA pass where every candidate keyword is manually searched on Amazon (All Departments), and keywords whose results don't match the product are highlighted rather than deleted, reserved for listing-indexing use only. Apply: Run this check as the final step before finalizing keyword lists, separating unhighlighted (PPC-ready) keywords from highlighted (indexing-only) keywords.
Four-Step PPC Framework — The video's overall build sequence: (1) automatic campaigns, (2) keyword research, (3) manual keyword campaigns, (4) negative keywords. Apply: Follow this order when building a PPC account from scratch, treating manual campaigns as the source of long-term profitability and negative keywords as the step most sellers skip.
Automatic Targeting Sub-Groups (Close/Loose Match, Substitutes, Complements) — Amazon's four automatic-targeting match groups, each with distinct bid behavior and buyer-intent level. Apply: Isolate each group into its own separate campaign (starting with Close Match, disabling the other three), rather than bidding all four together, to keep performance data clean.
Keyword Match Types (Exact/Broad/Phrase) — Three ways Amazon can match a targeted keyword to a shopper's search — Exact (word-for-word), Broad (related/loosely connected), and Phrase (same order, extra words allowed) — ranging from most controlled to most permissive. Apply: Use only Exact and Broad in campaign structure, skipping Phrase since anything it would catch is already covered by Broad match.
Aggressive Overbid Technique — Deliberately overbidding Amazon's suggested bid (e.g., by $1, or with odd-penny amounts) on exact-match, high-intent keywords to near-guarantee top ad placement, since charges only occur per click. Apply: Use during product launch on your best-converting keyword lists to accelerate data collection, choosing odd penny amounts to beat sellers who default to round bid increments.
Inflated Daily Budget Pacing — Setting a campaign's daily budget artificially high (e.g., $100) relative to intended spend so Amazon's pacing algorithm shows the ad more often throughout the day, since payment only occurs on click, not impression. Apply: Apply on exact-match campaigns, manually checking and pausing the campaign once actual spend hits your real target, resuming the next day.
Dynamic Bidding Strategy Choice (Up/Down vs Down Only vs Fixed) — Amazon's three bidding-strategy options controlling whether the platform can raise or only lower your bid in real time based on conversion likelihood. Apply: Use 'up and down' during launch for aggressiveness, 'down only' for a tighter budget or non-launch periods; the video states it never uses fixed bids.
Placement Bid Adjustments (Top of Search / Rest of Search / Product Pages) — Separate bid multipliers controlling where within search results or product pages an ad appears, since Top of Search reportedly converts best. Apply: Raise Top of Search by roughly 20-30% (or test in 10% increments), leaving Rest of Search and Product Pages at 0% until data justifies a change.
Placement-Adjustment Down-Shift Hack — A workaround for the fact that Amazon's placement bid adjustments can only raise bids by a percentage: raising a strong placement's adjustment while simultaneously lowering every keyword's base bid by the same percentage keeps the strong placement's effective bid unchanged while cheapening weaker placements. Apply: Apply after identifying one high-ACoS placement (e.g., Rest of Search) alongside one strong one (Top of Search), making this global change before touching individual keyword bids.
Negative Keyword Technique (two-phase) — A two-stage process for excluding non-buying searches: common-sense pre-filtering of obviously irrelevant/similar-sounding terms before launch, then ongoing removal of non-converting keywords found in PPC reports. Apply: Set negative keywords to exact or phrase match (not broad), refining the list continuously as report data accumulates to cut wasted ad spend.
7-10 Day Data-Collection Rule — A cooldown period of at least 7 to 10 days before making optimization changes to a new or adjusted campaign, to avoid reacting to daily sales noise. Apply: Wait out this window between each round of bid, keyword, or negative-keyword changes rather than checking and adjusting campaigns daily.
Four Foundational PPC Diagnostic Metrics (Impressions to CTR to Conversion Rate to ACoS) — A funnel-ordered audit sequence for diagnosing why a campaign underperforms, treating each metric as revealing problems at a different stage. Apply: Check impressions first (targeting/budget issues), then CTR (listing/targeting/reviews), then conversion rate (listing quality/pricing/traffic relevance), and only then evaluate ACoS.
ACoS (Advertising Cost of Sale) — Ad spend expressed as a percentage of the sales revenue those ads generated; framed as the funnel's outcome metric rather than a root cause. Apply: Compare ACoS against your actual profit margin (not a universal low number) to judge whether a keyword or campaign is truly profitable, evaluating it last in any diagnostic pass.
ROAS (Return on Ad Spend) — An alternative profitability metric to ACoS that some sellers prefer, per the video, functioning as an inverse measure of ad efficiency. Apply: Compare ROAS to your break-even ROAS threshold as an alternative lens to the ACoS-vs-margin comparison.
Color-Coded Keyword Triage (Red/Yellow/Green) — A PPC-report review system: red = ACoS far over margin or high spend with zero sales (turn inactive), yellow = borderline performer (lower bid gradually), green = profitable (migrate to a dedicated, more aggressively-budgeted campaign). Apply: Apply during each 7-10 day optimization review after sorting a downloaded PPC report by ACoS and by spend.
3x-Profit-Margin Spend Threshold — A rule for flagging a zero-sale keyword as a likely loser once its cumulative spend exceeds roughly three times the per-sale profit margin in dollars. Apply: Use when sorting a PPC report by spend to decide which no-sale keywords to cut versus leave running for more data.
Honeymoon Period — Amazon's claimed temporary organic-rank boost given to new listings, said to start the moment inventory goes live and in-stock. Apply: Finish all listing content (photos, copy) before launch so PPC — and the ranking boost — can start the instant the listing goes live, rather than delaying.
Setting a high daily budget on an exact-match campaign is framed as a pure exposure lever, not a spend lever, since Amazon paces impressions to the budget it thinks it needs but only charges per click.
The video recommends overbidding using odd, non-round penny amounts specifically because most competing sellers bid in round multiples of five, giving a cheap way to win ties.
A high Market Availability percentage is presented as a green flag for a keyword, not a red flag — it means the top 3 incumbents are NOT monopolizing the niche, which favors a new listing.
Since Amazon's placement bid adjustments can only raise bids by a percentage, the video's workaround for effectively lowering a weak placement's bid is to raise the strong placement's adjustment and simultaneously lower every keyword's base bid by that same percentage.
Keywords with zero sales but low spend are deliberately left running rather than cut, on the logic that there's insufficient data yet and the next click could convert; the cutoff for action is spend exceeding roughly 3x the per-sale profit margin.
The claimed conversion-rate gap between a high-volume keyword (6%) and a low-volume 'buyer intent' keyword (52%) for the same product is used to argue that chasing search volume is a systematically worse strategy than chasing conversion rate.
Reviews are framed as inseparable from PPC success: driving traffic (PPC) and converting it (reviews/social proof) are presented as two halves of the same launch problem, not sequential concerns.
«There's five key steps to setting up Amazon PPC campaigns correctly, and if you miss even one of them, you end up wasting money.»
— 00:01
«Most beginners are doing keyword research and PPC completely backwards.»
— 02:43
«What I see over and over again is people obsessing over finding the best keyword tool, then dumping every keyword it spits out into one campaign, and hoping Amazon figures it out for them. That's not a strategy. That's guessing.»
— 02:54
«Free tools just give you ideas, whereas Cerebro gives you validated data.»
— 05:20
«You want to focus on using high conversion keywords in your product listings, not high search volume keywords, because you want the Amazon search algorithm to index your listing for high conversion keywords, since it's sales that determine your listing rank, not search volume.»
— 16:55
«We don't care about keywords that didn't generate any sales because they're not going to help us rank.»
— 19:52
«PPC is hands-down the most frustrating part of selling on Amazon. It's always the fastest way to drain your wallet if you don't know exactly what you're doing.»
— 29:31
«Do you see the word sponsored? Those listings didn't earn their spot organically. They paid to be there.»
— 30:44
«Amazon PPC works like an auction.»
— 31:07
«There's a honeymoon period where Amazon helps boost your product rank whenever you launch a new product.»
— 31:58
«Amazon is a search engine.»
— 39:32
«Because once you know that, everything changes. You're no longer guessing, you're no longer wasting clicks. You're telling Amazon exactly where you want your ads to show up.»
— 41:13
«You don't pay when your ad is shown, you only pay when someone clicks. So, higher budgets doesn't automatically mean higher spend. It means more exposure.»
— 50:07
«What a negative keyword is is it tells Amazon, if someone searches for this, do not show my ad.»
— 55:01
«Most Amazon sellers don't struggle with PPC because they don't know how to adjust bids or pick keywords. They struggle because they're optimizing the wrong things.»
— 57:06
«ACOS should always be analyzed last, not first. It's not the root problem. It's the result of everything happening earlier in the funnel.»
— 62:04
«There's no shortcut here. This part requires testing.»
— 73:39
«A keyword that's profitable today isn't guaranteed to stay that way forever.»
— 74:18
«Shoppers naturally gravitate towards listings with the most social proof. It's human behavior.»
— 75:10
Reception
Viewers are largely appreciative and engaged, praising the creator's clear, high-quality PPC guidance and asking detailed follow-up questions, though some express frustration with their own Amazon ad performance and click fraud rather than with the video itself.
A procedurally dense, tool-centric tutorial that lays out Crescent Kao's specific keyword-research and campaign-structuring system end to end, built around paid third-party software (Helium 10 Cerebro, Sellerise Keyword Hunter/Re-ranker) the presenter has affiliate ties to; its value is in the exhaustive, replicable steps and named metrics rather than in explaining Amazon's underlying ad-ranking mechanics beyond what's needed to justify the workflow.
75:37