---
title: "How to Research a KDP Niche on Amazon"
description: "A durable KDP niche research order: seed demand, category depth, BSR spread, price band, formats, and gaps—without fake profitability scores."
source: "https://smartkdp.com/blog/how-to-research-a-kdp-niche"
date: "2026-08-09"
---

# How to Research a KDP Niche on Amazon

“Profitable niche” lists go stale the week they publish. What does not go stale is the **evidence pack** you need before you spend weeks on a manuscript, cover, and keywords: is there real demand, how hard is the shelf, what do winners charge, which formats are missing, and can you truthfully own a category path?

This guide is a **repeatable research order** for Amazon KDP niche decisions. You will gather public storefront signals in a fixed sequence, write them down once, and only then decide go / niche-down / walk away. For unit ranges from **overall** Best Sellers Rank, use the free [Amazon BSR Sales Calculator](https://smartkdp.com/tools/amazon-bsr-calculator?via=guide:how-to-research-a-kdp-niche). For full browse paths you may later validate in KDP, use the free [KDP Category Finder](https://smartkdp.com/tools/kdp-category-finder?via=guide:how-to-research-a-kdp-niche). For faster SERP scanning while you browse Amazon, see the [Seller Insights browser extension](https://smartkdp.com/extension) (beta).

> **Not affiliated with Amazon.** Amazon, Kindle, KDP, and related marks are trademarks of Amazon.com, Inc. or its affiliates. SmartKDP is not affiliated with, endorsed by, or sponsored by Amazon. Rank and category behavior below follow public Amazon/KDP help pages. **Amazon does not publish an official rank-to-sales table.** Treat any unit numbers as model estimates—including ours.

> **No magic thresholds.** This post will not invent “BSR under X = gold niche” or “fewer than N results = easy money.” Those rules are marketing copy, not KDP policy.

## The durable core: evidence first, score second

A niche decision is a **portfolio of observables**, not a single keyword tool score.

```
1. Seed        → what readers type / what shelf they browse
2. Depth       → category path + related paths (honest fit only)
3. Demand      → overall BSR spread of titles that actually match
4. Price band  → list prices + formats (ebook / paper / hard)
5. Quality bar → covers, reviews shape, differentiation room
6. Gaps        → format, angle, audience, packaging holes
7. Decision    → go / niche-down / kill — with reasons written down
```

If you reverse that order—pick a “hot niche,” then force categories to fit—you build a book for a badge, not a reader. KDP’s category and metadata rules expect **accuracy to content**, not shelf gaming ([Categories](https://kdp.amazon.com/help/topic/G200652170); [Metadata Guidelines](https://kdp.amazon.com/help/topic/G201097560)).

![Seven-step KDP niche research order from seed to written go/no-go decision](https://smartkdp.com/blog/how-to-research-a-kdp-niche/research-order.webp)

## Step 1 — Seed: name the reader problem, not the product type

Start with a **reader intent** you can state in one sentence:

- “Adults who want short daily crossword breaks at work”
- “Parents buying a first maze book for ages 6–8”
- “New managers who need a short plain-language 1:1 guide”

Then collect **seeds** in three free places:

| Seed source                                | What you capture                   | What it is not                   |
| ------------------------------------------ | ---------------------------------- | -------------------------------- |
| Amazon autocomplete (Books / Kindle Store) | Phrase shapes people type          | Not a volume number              |
| Related searches / “customers also bought” | Adjacent demand clusters           | Not proof you will rank          |
| Best Seller / New Releases category trees  | Shelf language Amazon already uses | Not permission to mis-categorize |

Autocomplete is a **demand signal**, not a search-volume API. For the mechanism and honesty limits, see [How Amazon Search Suggestions Work](https://smartkdp.com/blog/how-amazon-search-suggestions-work) and the free [Amazon Keyword Suggestion Expander](https://smartkdp.com/tools/amazon-keyword-suggestion-expander?via=guide:how-to-research-a-kdp-niche).

Write 10–20 seed phrases. Cluster them. Kill seeds that only describe a **format** (“paperback journal”) with no topic or audience.

## Step 2 — Depth: map honest category paths

For each surviving cluster, open Amazon’s category tree (or search, open a true competitor, and read **Product details → Best Sellers Rank** category lines). Capture the **full path**, not just the leaf name.

Use the [KDP Category Finder](https://smartkdp.com/tools/kdp-category-finder?via=guide:how-to-research-a-kdp-niche) to:

1. Search by a topic word.
2. Read parent → child chains.
3. Copy path strings you can re-find in KDP’s picker later.

**Rule that bites:** every path must still describe **your** future book when a stranger lands from that shelf. “Easy bestseller” micro-shelves that do not match the manuscript are a compliance risk, not a growth plan. You get a small number of category slots at title setup—treat them as scarce, accurate placement, not lottery tickets. Detail: [How to Choose KDP Categories](https://smartkdp.com/blog/how-to-choose-kdp-categories).

Note **marketplace**: US, UK, DE, and others do not share one identical tree. Research the store you will price and promote first.

## Step 3 — Demand: overall BSR spread, not one hero rank

Open 8–15 **true comps**—same problem, similar audience, same product type—not every book that shares one keyword.

For each, record:

- **Overall store BSR** (Books or Kindle Store as shown—do not mix them blindly)
- **Category ranks** (up to three bestseller lists may appear; KDP’s [Sales Ranking](https://kdp.amazon.com/help/topic/G201648140) help notes ranks are based on customer activity relative to other books)
- Title age signals if visible (publication date, edition)
- Whether the rank looks **stable** across a few checks or is a promo spike

Amazon states Best Sellers and Category Ranks reflect **relative customer activity**. It does **not** hand you copies/day. When you need a unit **range** from overall rank, plug that rank into the [Amazon BSR Sales Calculator](https://smartkdp.com/tools/amazon-bsr-calculator?via=guide:how-to-research-a-kdp-niche) and keep the low–mid–high band in your notes—not a single fantasy number.

What you are looking for in the **spread**:

| Pattern                                             | Reading                                                             |
| --------------------------------------------------- | ------------------------------------------------------------------- |
| Several matching titles with stronger overall ranks | Demand exists; competition is real                                  |
| One superstar + a cliff of dead ranks               | Demand may be winner-take-most; hard to be #2 without a sharp angle |
| All matching titles weak overall                    | Thin demand **or** wrong shelf—verify you sampled true comps        |
| Ranks jump wildly day to day                        | Promo/ad noise; recheck on a quiet day                              |

Do not stop at “#1 in a tiny subcategory.” Subcategory badges without overall velocity often mean a quiet shelf, not a business.

![Evidence pack diagram: demand, depth, price band, formats, quality bar, gaps → decision](https://smartkdp.com/blog/how-to-research-a-kdp-niche/evidence-pack.webp)

## Step 4 — Price band and format spread

For the same comps, table:

- List price (ebook vs paperback vs hardcover)
- Page count / trim if print
- KU / non-KU where visible for ebooks
- Series vs standalone

You are answering: **what buyers already pay** for this problem, and whether the shelf is ebook-led, print-led, or mixed. That feeds royalty math later ([KDP Royalty Calculator](https://smartkdp.com/tools/kdp-royalty-calculator?via=guide:how-to-research-a-kdp-niche))—not a guess that “everyone prices $2.99.”

**Format gaps** are research gold when they are real:

- Strong ebook comps, almost no quality paperback
- Puzzle books only in one trim readers complain about in reviews
- Workbooks without downloadable extras (if you can deliver extras **inside policy**)

Gaps that are “nobody made a clone of the #1 cover in blue” are not gaps.

## Step 5 — Quality bar and differentiation

Skim covers and Look Inside (where available) for the **bar you must clear**, not for theft:

- Cover promise vs title promise alignment
- Description themes (benefit clusters—not phrases to paste)
- Review themes: what buyers praise and punish
- Production tells for low-content/puzzle: margin, contrast, page economy

For a field-by-field listing read, use [How to Analyze a Competitor’s Book Listing](https://smartkdp.com/blog/how-to-analyze-competitor-book-listings). Optional: paste **your** draft description into the free [Word Cloud Generator](https://smartkdp.com/tools/word-cloud-generator?via=guide:how-to-research-a-kdp-niche) later to see theme density—never as a substitute for reading.

Write one sentence: **“I win if I am the book that \_\_\_ for people who \_\_\_.”** If you cannot finish it without naming a competitor brand, you do not have a position yet.

## Step 6 — Decision rules (write them down)

Kill or niche-down **before** cover spend when:

1. You cannot name three **honest** category paths.
2. True comps show no meaningful overall activity (and you confirmed they are true comps).
3. The only “gap” is lower price on an identical promise.
4. You would have to mislead metadata or categories to look like a fit.
5. You cannot clear the quality bar with the time and skills you actually have.

Go—or pilot with a minimum viable title—when:

1. Demand shows up in overall rank **spread**, not one lucky leaf badge.
2. Price band leaves royalty room after print cost (run the numbers).
3. You have a differentiation sentence a stranger understands.
4. Category paths stay accurate after the manuscript outline is real.

Re-run the pack after outline lock. Page count and trim change spine, cost, and sometimes the competitive set for print.

## Worked example (illustrative)

You are considering **“large-print word search for seniors — animals theme.”**

1. **Seeds:** Amazon autocomplete around large-print word search; related “for seniors,” “animals,” “easy.”
2. **Paths:** Books → … → activity / puzzle leaves that actually fit large-print word search—not a random craft leaf. Copy full paths via [KDP Category Finder](https://smartkdp.com/tools/kdp-category-finder?via=guide:how-to-research-a-kdp-niche).
3. **Comps:** 12 paperbacks that are large-print word search (not mixed sudoku bundles). Note overall Books BSR for each; convert a few mid-pack ranks through the [BSR Sales Calculator](https://smartkdp.com/tools/amazon-bsr-calculator?via=guide:how-to-research-a-kdp-niche) for a velocity **range**.
4. **Price/format:** Most comps paperback-only in a tight price band; few hardcovers → optional later, not required for entry.
5. **Quality:** Review themes mention print too light / puzzles too hard. Differentiation: higher contrast, clearer grid sizing, honest “easy” labeling.
6. **Decision:** Go **if** outline locks grid size + trim + page count you can produce cleanly; kill **if** every animal large-print book is already a saturated clone war and you have no production edge.

Numbers in your real sheet will differ; the **order** does not.

## How to use the free tools in this workflow

### Amazon BSR Sales Calculator

1. Open [Amazon BSR Sales Calculator](https://smartkdp.com/tools/amazon-bsr-calculator?via=guide:how-to-research-a-kdp-niche).
2. Enter a competitor’s **overall** store rank (and marketplace/format context the tool asks for).
3. Record the **range**, not only the midpoint.
4. Repeat for several comps so you see a band for the niche, not one title’s lucky hour.

### KDP Category Finder

1. Open [KDP Category Finder](https://smartkdp.com/tools/kdp-category-finder?via=guide:how-to-research-a-kdp-niche).
2. Search topic language from your seed list.
3. Save full paths that still fit after you outline the book.
4. Prefer paths you can defend under metadata accuracy rules.

### Seller Insights extension

The [Seller Insights extension](https://smartkdp.com/extension) is a browser-side helper for Amazon research sessions (beta). Use it to speed scanning of public listing signals while you work the checklist above—not as a substitute for reading comps or for inventing backend keyword claims. Backend keywords are **not** public; do not pretend a niche tool revealed them.

## FAQ

### What is KDP niche research?

It is the process of gathering **public demand, competition, price, format, and category-fit evidence** before you commit to writing and publishing a book for a specific reader problem. It is not a single “niche score” from a paid dashboard.

### How do I find a low-competition KDP niche?

Look for a **real reader problem** where true comps show some demand, the quality bar is clear, and you have a differentiation you can execute—then niche **down** on audience, angle, or format until the competitive set is specific. “Low competition” with **no** overall velocity is usually a dead shelf.

### What BSR means a niche is good?

There is no official good-BSR cutoff. Judge **overall rank spread** among true comps against your goal (validation, payback, etc.). Use ranged estimates from the [BSR Sales Calculator](https://smartkdp.com/tools/amazon-bsr-calculator?via=guide:how-to-research-a-kdp-niche). See also [What Counts as a Good BSR on Amazon?](https://smartkdp.com/blog/what-is-a-good-bsr-on-amazon).

### Should I use Amazon’s Best Sellers list for niche research?

Yes as a **seed and density** source—read the list, open detail pages, record overall ranks and paths. No as a “copy the #1 title” brief. Bestseller lists are relative activity snapshots that move.

### How many competitors should I research?

Enough true comps to see a **pattern** (often on the order of a dozen for a focused niche), not one hero and not 200 random keyword matches. Quality of match beats quantity of rows.

### Can I trust niche finder tools that show “monthly sales”?

Only if you understand the model. Amazon does not publish an official BSR→sales table. Prefer tools that show **ranges and methodology**. Discard anything that claims exact daily sales or secret backend keywords for competitors.

### Do categories or keywords matter more for niche research?

Both, for different jobs. **Categories** are shelf placement with accuracy constraints. **Keywords** (storefront language + your seven backend boxes later) describe search phrasing. Research both; do not stuff category names into fields that forbid that pattern. See [KDP 7 backend keywords](https://smartkdp.com/blog/kdp-7-backend-keywords-explained).

### When should I stop researching and write?

When your evidence pack answers demand, depth, price, formats, quality bar, and a one-sentence position—and a re-check a day later does not reverse the go decision. Endless research is often fear of shipping, not missing data.

## Related free tools

- [Amazon BSR Sales Calculator](https://smartkdp.com/tools/amazon-bsr-calculator?via=guide:how-to-research-a-kdp-niche) — ranged unit estimates from overall BSR
- [KDP Category Finder](https://smartkdp.com/tools/kdp-category-finder?via=guide:how-to-research-a-kdp-niche) — full browse paths for honest placement
- [Amazon Keyword Suggestion Expander](https://smartkdp.com/tools/amazon-keyword-suggestion-expander?via=guide:how-to-research-a-kdp-niche) — autocomplete expansion without fake volumes
- [Word Cloud Generator](https://smartkdp.com/tools/word-cloud-generator?via=guide:how-to-research-a-kdp-niche) — theme density on text you already own
- [Seller Insights extension](https://smartkdp.com/extension) — browser helper for Amazon research sessions (beta)
- [All free tools](https://smartkdp.com/tools)

## Disclaimer

Amazon, Kindle, and KDP are trademarks of Amazon.com, Inc. or its affiliates. SmartKDP is an independent site and is not affiliated with, endorsed by, or sponsored by Amazon. Category trees, ranking behavior, and storefront UI change without notice. Confirm paths and product details against the live Amazon store and your KDP account before you publish. Estimates from any BSR tool—including ours—are models, not Amazon-reported sales.