The Amazon PPC Keyword Research Guide

Why Most Keyword Research Fails

Most Amazon sellers do keyword research once — at launch. They pull a list from a third-party tool, stuff the best ones into their campaigns, and move on. Months later they wonder why their ACOS is climbing and their organic rank isn't moving.

The problem isn't the keywords. The problem is the process. Keyword research isn't a one-time event. It's an ongoing system that informs your PPC structure, your listing copy, your launch strategy, and your ranking campaigns. Get it right and it compounds. Get it wrong and you spend months chasing the wrong traffic.

This guide covers the complete methodology: how to think about keywords strategically, which tools to use (and how to use them), how to validate demand with Amazon's own data, how to organize keywords into a workable architecture, and how to keep the system improving over time.

Root Keyword Strategy

Start with Root Keywords, Not Long-Tail

The most common mistake in Amazon keyword research: starting with long-tail keywords and building upward. The right approach is the opposite — start with root keywords (the core one- or two-word phrases that define what you sell) and expand outward.

Every niche has a handful of root keywords that matter. For a silicone baking mat, those might be "baking mat," "silicone mat," and "oven liner." For a protein powder, it's "protein powder," "whey protein," and "protein supplement." These root keywords define the branches — every phrase match, every long-tail variation, every related search is downstream of them.

How to Identify Your Root Keywords

Start here:

1. Name the product category exactly: What is the shortest, most accurate phrase that describes what you sell? Don't overthink it. Start there.
2. Ask the customer question: If a customer needed your product and had never heard of your brand, what would they type into Amazon's search bar? Not what you want them to type — what they actually would.
3. Check your competitors' top search terms: Run a reverse ASIN lookup on your two or three closest competitors. The highest-volume keywords showing up for all of them are likely root keywords for your category.
4. Look at Amazon's autocomplete: Type your product name into the Amazon search bar and note what autocompletes. These are real searches being made — Amazon shows the most popular completions.

Root Keyword Grouping

Once you have your root keywords, group them into themes. A theme is a set of root keywords that share the same core intent. For a blender brand, you might have: protein shakes theme (protein blender, smoothie maker, shake blender), kitchen appliance theme (countertop blender, kitchen blender, home blender), and travel/portable theme (personal blender, mini blender, travel blender).

Each theme will eventually become its own campaign cluster — the same keywords, the same lifestyle imagery, the same headline. This is how keyword research connects directly to campaign structure and creative strategy.

✦ ADLABS FEATURE

SQP Data That Shows Which Root Keywords Actually Drive Your Sales

AdLabs surfaces your Search Query Performance data at the keyword level — impression share, click share, purchase share — so you can see which root keywords your customers are actually using to find (and buy) your products. No more guessing which themes matter most.

Try Free →

The Keyword Research Tool Stack

No single tool tells the complete story. Effective Amazon keyword research requires layering multiple sources — Amazon's own first-party data, reverse ASIN tools, and dedicated keyword analytics platforms. Here's how we think about the stack.

3rd Party Tools: The Recommended Research Hub

Using additional software is an easy choice for getting started with serious keyword research. There are lots of tools out there that offer excellent research tools for Amazon sellers.

These tools generally combine reverse ASIN lookup, keyword database search, competitor analysis, and relevancy scoring systems into one platform built specifically for Amazon.

Amazon First-Party Data (SQP & POE)

Why Amazon's Own Data Wins

Third-party tools estimate search volume based on various signals and models. Amazon's Search Query Performance report shows you actual search counts for your category. When there's a conflict between what a tool says and what SQP shows, trust SQP. It's the most accurate source of search volume data available to Amazon sellers.

Search Query Performance (SQP)

SQP is available in Seller Central under Brands → Brand Analytics → Search Query Performance. It shows you:

• Total search volume for each query in your category
• Your brand's impression share, click share, and purchase share on each term
• The full ICAP funnel: Impressions → Clicks → Add to Cart → Purchases

For keyword research, this is invaluable. You can see exactly which search terms your target customers are using — and whether they're buying from you, from competitors, or from nobody (low purchase share across the board = opportunity).

How to Use SQP for Keyword Research

1. Export 12 months of data (use the Seller Central Utilities Chrome extension to bulk-export multiple months at once)
2. Filter for terms with meaningful volume — ignore queries with fewer than 100 searches/month
3. Sort by purchase volume, not just impressions — high impressions with low purchases means people are searching but not buying (often a conversion rate issue, not a targeting one)
4. Look for high-volume terms where your purchase share is low — these are the gaps. You're visible but not winning. These keywords deserve aggressive campaign investment.
5. Classify terms as brand, non-brand, or competitor — the non-brand terms with high purchase shares across the category but low shares for your brand are your biggest opportunities

Product Opportunity Explorer (POE)

POE (Seller Central → Growth → Product Opportunity Explorer) shows you category-level demand data: niche search volume, seasonality trends, average selling price, number of products, and click concentration.

For keyword research, POE is most useful for understanding category-level intent and finding adjacent niches you might be missing. If your primary niche has high competition and you're losing on price, POE can surface related niches where search volume is significant but competition is lower.

Validating Keywords

You've identified candidate keywords from multiple sources. Now you need to validate them — confirm they have real search volume, that the competitive landscape is winnable, and that they're actually relevant to your product. Third party tools allow for this validation step.

Reverse ASIN Analysis

Start by running reverse ASIN lookups on your top 3–5 competitors. Enter their ASINs into a 3rd party tool and see every keyword they rank for organically and paid, along with estimated volume.

Focus on keywords that appear for multiple competitors with high organic/paid rank and relevance — these are the core terms in your category. If your top 5 competitors all rank for the same 20 keywords, those are your must-have targets.

Filtering for Actionable Keywords

A raw keyword list from any third party tool might have hundreds or thousands of terms. Before you act on it, filter it down using these criteria:

• Remove irrelevant terms: Filter out keywords that describe different materials, sizes, use cases, or product types that don't apply to you. This is your first n-gram pass — look for word patterns that are categorically irrelevant.
• Bucket by intent tier: High-volume root keywords (broad targeting, research campaigns), mid-volume phrase variations (phrase match), proven long-tail converters (exact match).
• Check conversion viability: A keyword with massive search volume but poor conversion economics (high competition, low margins at winning CPC) may not be worth targeting at your current scale.
• Cross-reference with SQP: Any keyword you plan to invest heavily in should have confirmation from SQP that actual purchases happen on that term in your category.

✦ ADLABS FEATURE

Automated Keyword Harvesting — Move Validated Keywords to Exact Match Automatically

Once you've identified your keyword targets, AdLabs automates the harvesting workflow — surfacing converting search terms from your campaigns and moving them to exact match with criteria that actually make sense. No more manual bulk sheet work for every new keyword.

Try Free →

Competitive Keyword Research

Your Competitors Are Your Keyword Map

The fastest path to a comprehensive keyword list is your competitors' organic rankings. Every keyword a top competitor ranks for organically is a keyword that Amazon's algorithm has associated with their product — and that association was built through PPC traffic, conversions, and click history on that term.

Your goal: find the overlap between "keywords my competitors rank for" and "keywords that are relevant to my product." That overlap is your core keyword target list.

The Competitor Keyword Research Process

1. Identify 5–10 true competitors: Products that a shopper searching your main keyword would seriously consider buying instead of yours. Not aspirational comparisons — actual alternatives.
2. Run all of them through a 3rd party research tool: Collect their organic keyword rankings, estimated volumes, and relevancy scores.
3. Find keyword clusters: Group keywords that appear across multiple competitors. Terms in 4 or 5 competitor ASINs are the heart of your category.
4. Look for gaps: Keywords that appear in competitor listings but not in yours (in your actual product listing copy or backend keywords) may be relevancy gaps limiting your organic indexing.
5. Identify emerging keywords: New search terms appearing in competitor data that weren't there six months ago signal shifting customer language — get ahead of them before competition intensifies.

Competitor Negative Keywords: The Offensive Reverse

Competitive keyword research also informs your negative keyword strategy. When you see the search terms a top competitor has negated (visible in their search term report history, or inferable from their bid patterns), you're seeing terms that don't convert for products like yours. That saves you from expensive testing.

Building Your Keyword Architecture

Keywords don't live in isolation — they need to be organized into a structure that maps directly to your campaign architecture. The goal is a system where every keyword has a home, every campaign has a purpose, and you can trace performance back to individual keywords with precision.

The Three-Tier Keyword Structure

Tier 1: Research Tier

Root keywords and broad variations go into auto campaigns and broad/phrase research campaigns. The goal here is data collection and keyword expansion. These campaigns tell you which search terms customers actually use, which ones convert, and which ones waste money. Your search term reports from these campaigns are the primary source of keywords for Tier 3 These campaigns help occupy one-off searches and long tail keywords also.

Tier 2: Phrase Match Tier

Validated phrase variations — terms you've confirmed have real volume and relevancy but haven't fully proven at the exact search term level — go into phrase match campaigns. These bridge the gap between broad research and surgical exact match targeting.

Tier 3: Performance Tier (Single Keyword Campaigns)

Your highest-priority, proven-converting exact match keywords get their own single keyword campaigns (SKCs). One keyword, one ASIN, one campaign. Maximum bidding control, clean performance data, dedicated budget. These are your ranking keywords — the terms you're actively trying to dominate organically.

How Keywords Move Between Tiers

The architecture is dynamic. Keywords get promoted when they show consistent relevance, and move across tiers when they've proven they convert. Terms not driving performance can be evaluated for negation or harvested to control. The system is constantly refining.

Prioritization Framework

With hundreds of potential keywords identified, you can't target everything at once — and you shouldn't. Smart prioritization means spending your limited budget and attention on the keywords most likely to drive ranking, revenue, and long-term organic growth.

The Keyword Priority Score

We evaluate keywords across four dimensions:

The Quick Prioritization Filter

When you need to move quickly, apply this filter:

1. Must-have (launch immediately): High volume, high relevance, you currently rank poorly or aren't ranking at all. These are revenue-positive and ranking gaps simultaneously.
2. Should-have (launch within 30 days): Medium volume, high relevance. Good profitability at target ACOS but lower immediate ranking impact.
3. Test and watch: High volume, medium relevance. Could work but needs data. Budget for testing, not heavy investment.
4. Skip for now: High competition, low purchase share across all competitors, low relevancy score. Not the right battle at your current scale.

Seasonal Keyword Timing

Some keywords are stable year-round. Others spike predictably or (unpredictably). 3rd party tracking tools and POE both show search trend data over time — use this to plan your keyword expansion calendar. If you sell outdoor products, you want to be established on summer keywords before April, not after. If you sell gifts, you want Q4 terms in your exact match campaigns by October.

✦ ADLABS FEATURE

Placement-Level Data That Tells You Which Keywords Are Worth Ranking For

AdLabs shows you CVR and RPC by keyword and placement — so you can identify which of your target keywords actually convert at top of search vs. which ones just generate impressions. Prioritize the keywords worth investing in for ranking, not just traffic.

Try Free →

Ongoing Keyword Harvesting

Keyword research isn't a launch task — it's a continuous process. Your campaigns are constantly collecting data on which search terms customers actually use, which ones convert, and which ones are wasting money. Harvesting that data and acting on it is what separates accounts that improve over time from accounts that plateau.

The Harvesting Workflow

Every two to four weeks, run this process:

1. Download your search term report for the past 30–60 days across all campaigns.
2. Aggregate search term performance — pivot table all terms, sum spend, clicks, orders, and sales. Don't look at individual rows; look at the aggregated performance per unique search term.
3. Identify harvestable terms: Search terms with 2+ orders (or same-SKU orders), relevant to the advertised product, that aren't already in an exact match campaign.
4. Identify negative candidates: Search terms that are irrelevant to your product (wrong material, wrong use case, wrong demographic). Look for patterns using n-gram analysis — one irrelevant word negated as a phrase match can block hundreds of irrelevant searches at once.
5. Act on both lists: Add harvested terms to exact match campaigns. Add negative phrases to source campaigns.

Harvesting Criteria That Actually Work

The n-Gram Shortcut for Negation

When reviewing search terms for negation, don't negate individual terms one by one. Aggregate your search term data and look at individual word components — unigrams and bigrams. A single word that's irrelevant to your product (like "steel" for a silicone seller) might appear in 50 different search terms. One negative phrase match blocks all of them at once, instead of playing whack-a-mole with individual terms.

For a complete breakdown of this technique, see our N-Gram Analysis Guide.

Keywords Are a Living System

The biggest mindset shift in Amazon keyword research: stop thinking about it as research you do before you launch, and start thinking about it as a system you build and refine over time. The keywords you launch with aren't the keywords you'll be running six months from now — and that's exactly how it should be.

The Complete System at a Glance

Key Principles That Don't Change

1. Start with root keywords, expand outward. Long-tail follows from root keyword authority, not the other way around.
2. Validate with Amazon's data. Third-party volume estimates are approximations. SQP purchase share is real.
3. 3rd party tools for discovery, SQP for validation. They answer different questions — use both.
4. Keywords need a home. Every keyword you target should have a specific campaign type and match type assigned to it. No keyword orphans.
5. Negate themes, not individual terms. One phrase match negative is worth 50 exact match negatives when you find an irrelevant word pattern.
6. Harvest what converts, cut what doesn't. Your search term reports tell you the truth about customer language. Trust the data.
7. Keyword research is never finished. The system improves with every optimization cycle. Set the cadence and stick to it.

This guide is based on years of managing keyword strategy across hundreds of Amazon advertising accounts, combined with research from That Amazon Ads Podcast and That Amazon Ads Masterclass.