Amazon KDP Keyword Research: Step-by-Step Guide

You type in a seed phrase, a generic tool gives you 40 keywords alongside a neat “monthly search volume” number, and you pick the 7 with the biggest numbers. That workflow has a fundamental flaw: Amazon does not publish search volume data for books, meaning those figures are calculated guesses or outright fabrications. Here is the authentic, evidence-based process: transforming a seed phrase into 7 high-performing KDP backend keywords using live Amazon autocomplete, intent classification, and real-world demand signals.

Step 1: Start with a Seed Phrase, Not a Closed Keyword List

Start with a concise phrase describing your book’s core premise — not a preconceived list of keywords you assume you want to rank for. The seed phrase is the starting point of discovery: beginning with fixed assumptions is the fastest way to repeat the same saturated keywords instead of finding untapped search queries.

Step 2: Extract Candidates from Live Amazon Autocomplete

From your seed phrase, Markfolio queries the real-time autocomplete suggestions that Amazon returns to shoppers — the exact dropdown readers see when typing into the Amazon search bar. This is not a static database of generic book keywords: every suggestion reflects active shopper behavior in your target marketplace.

Step 3: Classify Each Candidate by Search Intent

Not all keywords serve the same purpose. Markfolio categorizes suggestions by intent type — topic, problem solved, promised outcome, buyer persona, reader skill level, methodology, format, or in fiction: genre, subgenre, trope, character archetype, setting, tone, and central conflict. Two keywords of the same type compete for the same slot in your 7 backend keywords; distinct types open completely different search pathways.

Step 4: Score Candidates Using Real Demand Signals, Not Fake Volume

Instead of relying on fabricated search volume numbers, Markfolio calculates a relative Demand Index from 0 to 100 based on verifiable signals:

Signal	What It Measures	Weight in Demand Index
Autocomplete Depth	How many downstream variations Amazon suggests for that phrase	40%
Total Search Results	The overall catalog competition returned for that phrase	20%
Exact Match Saturation	Percentage of top-ranking books with the exact phrase in their title	25%
Ranking Stability	Whether the top-ranking titles remain steady or churn frequently	15%

If data for a signal is unavailable, the weight dynamically redistributes across available signals rather than inserting a placeholder average. Each score also includes an explicit confidence rating — low, medium, or high — based on observation history.

Step 5: Build a Balanced 7-Keyword Portfolio

Selecting your final 7 backend keywords is not just sorting by score and picking the top 7. Choosing near-identical phrases (“easy vegan recipes” and “vegan recipes easy”) wastes precious slots. Markfolio penalizes keyword redundancy and rewards semantic diversity across intent types. The result is a balanced portfolio of 7 keywords that capture maximum reader discovery angles.

Amazon explicitly prohibits terms like “free”, “bestseller”, or “number one” in backend slots — high-risk terms are automatically filtered out before scoring.

Step 6: Copy Directly into Your Amazon KDP Dashboard

Once selected, your 7 backend keywords are formatted and ready to paste directly into your KDP book details panel.

Frequently Asked Questions

Why don’t you show exact monthly search volume figures? Because Amazon does not disclose search volume data for books. A relative 0–100 demand index built on verified signals is the only transparent, dependable alternative.

How many keyword candidates should I review? Evaluating at least 30–50 candidates gives the algorithm enough diversity to find 7 keywords spanning distinct intent categories.

What does “low confidence” mean? It indicates limited historical observation data for that specific query — not that the keyword is ineffective.

Can I use the same keyword research across books in a series? Each book has unique themes and character arcs, but shared subgenre and trope candidates can certainly be reused across series titles.

To learn how to pick the best Amazon browse categories alongside your keywords, read our guide on finding profitable Amazon KDP categories. For the full strategic overview, explore our complete KDP keyword research guide.