08:32
Amazon PPC
Successful Amazon PPC requires continuous, data-driven optimization rather than a 'set and forget' approach: sellers should expect targeting to shift dramatically after launch, must 'buy the data' by spending real ad dollars before judging a keyword, and should adapt tactics for Q4 tent-pole events, rising CPCs, and new data/AI tools.
PPC targeting is dynamic, not static — expect keyword performance and relevance to look 'wildly different' a week or a month after launch.
New sellers with no reviews or reputation often can't convert efficiently even on an 'obviously relevant' exact-match keyword; product-name matching alone doesn't guarantee results.
Advertising is described as a 'constant battle upwards' requiring ongoing optimization, not letting a campaign 'run its course.'
For tent-pole events (Black Friday, Cyber Monday), agencies decide per client whether they're actively running a deal; if not, ads are turned off entirely on those days to protect spend, since CPCs spike and non-deal listings convert poorly.
For seasonal/gift products that do participate, gift-related keywords get notably more expensive in Q4; Brand Analytics top search terms are used to prioritize the most reasonable keywords rather than 'boiling the ocean.'
Two paths to determine keyword relevance: (1) run auto/broad/phrase campaigns on core keywords and mine the search term report, or (2) do upfront keyword research with tools like Data Dive, Jungle Scout, or Helium 10.
The 'buy the data' rule: don't try to anticipate a keyword's performance — spend a set dollar amount on it, then decide to keep or kill based on actual sales results, because both Amazon's relevance algorithm and human behavior are unpredictable.
Historically ('old school'), the strategy was to start broad, let Amazon's algorithm feel it out, then segment down; newer tools (Product Opportunity Explorer, Brand Analytics, Data Dive, Helium 10, Jungle Scout) now let less experienced advertisers pick better starting targets.
Two structural shifts versus the past: CPCs are dramatically higher (down from roughly two-cent clicks historically), raising the cost of testing, while data availability has increased — making precise, intentional targeting more important as costs rise.
Recent high-value tactics cited by panelists: AMC audience targeting and B2B advertising (for products with business-buyer appeal) both reported as delivering strong recent results.
For AI-assisted market research, one panelist avoids feeding the AI social/Reddit/TikTok content and instead directs it to pull directly from competitor websites to map the biggest and newest players in a market.
AI tools reportedly cannot crawl Amazon's own site because Amazon blocks it, and Amazon is suing Perplexity — a constraint expected to persist, pushing AI research toward competitor/DTC sites rather than Amazon data.
Charm.io is cited as a tool for spotting newer market entrants, and Charm account data can be connected directly into Perplexity for combined market analysis.
Auto/Broad/Phrase Campaign Testing — A keyword-discovery method where sellers run auto, broad, and phrase match campaigns on core keywords and use the resulting search term report to surface additional relevant keywords. Apply: Launch all three match types on seed keywords, monitor the search term report regularly, and promote high-performing search terms into more targeted campaigns.
"Buy the Data" Keyword Decision Rule — Mina's rule, credited to 'Ritu,' that keyword relevance can't be predicted, so a set dollar amount must be spent on a keyword before deciding whether to keep or cut it. Apply: Set a spend threshold tied to product price, then only judge or kill a keyword once that spend has produced a sale-or-no-sale data point.
Tent-pole Event On/Off Ad Strategy — A Q4-specific tactic of deciding per client whether they're actively participating in deal events like Black Friday/Cyber Monday, since CPCs spike during those windows. Apply: If a client isn't running a deal for a tent-pole event, turn ads off entirely on those specific days to protect spend; if they are participating, advertise aggressively and budget for elevated CPCs.
Gift-Keyword / Brand Analytics Prioritization — For seasonal/gift products, gift-related keywords become notably more expensive in Q4, so top search terms from Amazon's Brand Analytics are used to gauge search volume and pick the most reasonable keywords rather than targeting all of them. Apply: Pull top search terms from Brand Analytics for the category before Q4, estimate search volume, and selectively bid on the most reasonable gift keywords instead of covering every term.
Old-School Broad-to-Segmented Strategy — Elizabeth's description of the traditional approach from when CPCs were low: start in broad match, let Amazon's algorithm 'feel it out,' then segment down into more specific targeting. Apply: Begin new targeting in broad match to gather algorithmic signal cheaply, then split winning terms into narrower match types once data accumulates.
Third-Party Keyword Research Tools (Helium 10, Jungle Scout, Data Dive) — External software platforms used to research keywords and estimate search volume/opportunity before launching campaigns. Apply: Use one of these tools to build an initial keyword list and estimate volume before testing, to pick better starting targets with less guesswork.
Amazon Product Opportunity Explorer — An Amazon-native data tool cited by Elizabeth as giving sellers more visibility into market opportunity data than was previously available. Apply: Use it alongside Brand Analytics to identify promising niches/keywords before committing PPC spend.
AMC Audience Targeting — Targeting ads based on Amazon Marketing Cloud (AMC) audience segments, named by a panelist as the most valuable recent change to their campaign strategy. Apply: Build and target AMC-defined audiences within campaigns to improve on recent results, per the panel's experience.
B2B Advertising — Running ads specifically targeting business buyers on Amazon, reported by a panelist as performing quite well for eligible products. Apply: If a product has business/bulk-buyer appeal, add B2B targeting to the campaign mix alongside standard consumer targeting.
AI-Assisted Competitor Market Research (via Perplexity) — Noah's workflow of using an AI tool (Perplexity) for market research, deliberately excluding Reddit/social/TikTok sources in favor of direct competitor website data to map the biggest and newest players in a market. Apply: Prompt the AI to pull directly from competitor sites rather than social threads, and use the output to scope market leaders and emerging competitors rather than granular consumer sentiment.
Charm.io + Perplexity Data Integration — Charm.io is a market-intelligence tool for spotting newer players; its account data can be connected directly into Perplexity to combine sources for category/market analysis. Apply: Link a Charm.io account into Perplexity so competitor/market data flows directly into AI-assisted research sessions.
Exact-Match Top-of-Search Trap — A common beginner mistake where new sellers assume their product's obvious keyword (e.g., 'Mentos mints' for a mints product) will convert well via exact match, top-of-search placement, ignoring that they lack reviews/reputation. Apply: Avoid over-committing budget to exact-match, top-of-search bids on 'obvious' keywords when launching with no reviews; expect low initial conversion and budget accordingly.
The 'buy the data' philosophy reframes keyword testing as a paid research expense you're expected to incur, not a targeting mistake to avoid — you spend to learn rather than guess correctly upfront.
The tent-pole strategy goes beyond reducing budget: agencies fully pause ad spend for non-participating clients during major sale windows, treating 'no deal = no chance of converting efficiently' as a hard rule rather than a soft optimization.
Rising CPCs are framed as a reason data and tools matter more, not less — precision targeting becomes cost mitigation as testing gets more expensive, not just a nice-to-have best practice.
The AI market-research workflow deliberately filters out user-generated/social content in favor of direct competitor-site data, implying social sources are treated as noisier or less reliable for this use case.
Amazon's active blocking of AI crawlers plus its litigation against Perplexity is presented as a structural constraint reshaping how sellers can use AI for research — pushing it toward competitor and DTC intelligence rather than direct Amazon marketplace data.
«have the expectation going in that whatever it is that you end up targeting any keywords that you have anything like that will be wildly different a week or a month from when you first started it.»
— 00:17
«I have Mentos mints and therefore I'm going against the keyword Mentos mints and so I should do amazing, right? No, that's not really how advertising works at the end of the day»
— 00:36
«You have to actually work towards advertising. It's not a set and forget. It's a constant battle upwards.»
— 01:28
«but if they aren't, we actually turn off ads for especially those days when we know that if they have no deal, they have no chance of converting efficiently and so we just like protect their ad spend and not not even run ads during that time.»
— 02:43
«You don't want to boil the ocean and try to go after every every one of them. Be be strategic is what I would say.»
— 03:44
«I learned this from Ritu is I wait until I spend a certain amount of dollars on a keyword, and then I make a decision. And I have to buy that data.»
— 04:22
«There's zero chance you're going to anticipate because not only is it, you know, Amazon's relevance, da da da, it's also human behavior, uh, which none of us can also predict.»
— 04:40
«I mean you used to be able to get two cent clicks. I would kill for two cent clicks. It's not going to happen.»
— 06:19
«all AI can't actually crawl Amazon sites cuz Amazon doesn't want to, right? And now Amazon is also suing Perplexity, so I don't think that's going to change anytime soon.»
— 07:55
Reception
No comments are available to gauge audience reception.
The video is a multi-speaker practitioner panel trading concrete Amazon PPC tactics rather than a single-author lecture, blending evergreen principles like data-driven keyword testing with time-bound specifics on current CPC levels, Q4 tent-pole strategy, and AI-assisted market-research tools.
08:32