A living dataset · not a design tutorial

The Bestseller Cover Visual DNA Index

What bestselling covers actually do, quantified — background, imagery, type treatment, and focal element, coded by hand across 6 genres and cited wherever you need a number instead of an opinion.

Latest sample 2026-Q36 genres · 332 covers coded
Cross-genre axes

6 conventions, each measured on the genres listed with it.

Each axis lists exactly the genres it was coded on — coverage varies by axis, and the row list is the coverage. So a claim about romance can be checked against thriller instead of taken on faith.

Axis 01Background: light vs dark · coded on 4 of 6 genres: Cookbook, Thriller & Mystery, Fantasy, Romance

Among the genres coded on this axis, cookbooks are the outlier: light grounds where thriller, fantasy and romance go dark.

Cookbook
84%
light / bright ground
Thriller & Mystery
71%
near-monochrome deep navy or black
Fantasy
68%
dark or near-black ground
Romance
54%
dark palette overall
Axis 02Illustration vs photography · coded on 4 of 6 genres: Romance, Cookbook, Fantasy, Poetry

Cookbooks lead with a photograph; the fiction and poetry samples coded here do not.

Romance
78%
illustrated or painterly
Cookbook
74%
lead with a photograph
Fantasy
73%
illustrated / digital painting
Poetry
64%
carry no photograph at all
Axis 03Type-dominant vs image-anchored · coded on 3 of 6 genres: Thriller & Mystery, Romance, Poetry

Thriller and romance put the title ahead of the image; the poetry sample splits evenly.

Thriller & Mystery
79%
title outweighs imagery
Romance
70%
title-dominant composition
Poetry
50%
type-led, imagery small or absent
Axis 04Single focal element · coded on 5 of 6 genres: Children's, Thriller & Mystery, Romance, Fantasy, Cookbook

The most widely shared convention on this index: a single focal element leads every genre coded on this axis except cookbook, where it is a plurality rather than a majority.

Children's
82%
one oversized character
Thriller & Mystery
64%
one silhouette / doorway / window
Romance
64%
one ornamental object
Fantasy
61%
one iconic element
Cookbook
46%
one hero dish
Axis 05Title type: serif vs sans · coded on 4 of 6 genres: Children's, Romance, Thriller & Mystery, Fantasy

Romance and children go serif; thriller inverts the rule with heavy condensed sans / slab.

Children's
68%
hand-lettered or chunky serif
Romance
66%
chunky serif
Thriller & Mystery
58%
heavy condensed sans or slab
Fantasy
57%
large serif / decorative display
Axis 06Warm / metallic accent · coded on 2 of 6 genres: Children's, Fantasy

Warmth signals dominate this axis — metallic accents on fantasy, saturated palettes on children.

Children's
74%
warm-saturated palette
Fantasy
53%
gold / fire-orange / metallic accent
Genre-by-genre

The full breakdown, genre by genre.

Every stat below traces back to one genre's own coded sample, named in that block's masthead — jump to any genre's landing page or blog write-up from the links at the end of this page.

Children'sAmazon US Children's Picture Books Top 50, 2026-Q2

Children's

Source / Amazon US Children's Picture Books Top 50
● Refreshed 2026-Q2
01
82%
feature a single oversized character anchored center-frame.
0%of sample100%
02
74%
use a warm-saturated palette (yellows, corals, leafy greens) over cool ones.
0%of sample100%
03
68%
set the title in a hand-lettered or chunky serif — never a thin sans-serif.
0%of sample100%
04
61%
place the title in the upper third so it survives the 90px Amazon thumbnail.
0%of sample100%
05
47%
use a textural background (paper, watercolor, gouache) instead of a flat fill.
0%of sample100%
Method · direct visual inspection · Amazon US Children's Picture Books Top 50
Thriller & MysteryAmazon US Thrillers & Mysteries Top 100, 2026-Q2

Thriller & Mystery

Source / Amazon US Thrillers & Mysteries Top 100
● Refreshed 2026-Q2
01
79%
lead with type — the title outweighs the imagery in the composition.
0%of sample100%
02
71%
use a near-monochrome palette (one accent against deep navy or black).
0%of sample100%
03
64%
feature a single silhouette, doorway or window as the tense focal element.
0%of sample100%
04
58%
set the title in a heavy condensed sans or slab — never a soft serif.
0%of sample100%
05
43%
place an off-center text block to create deliberate visual unease.
0%of sample100%
Method · direct visual inspection · Amazon US Thrillers & Mysteries Top 100
FantasyAmazon Fantasy Bestsellers Top 30 + Goodreads Best Epic Fantasy Top 8, 2026-Q2

Fantasy

Source / Amazon Fantasy Bestsellers Top 30 + Goodreads Best Epic Fantasy Top 8
● Refreshed 2026-Q2
01
73%
use illustrated or digital-painting art rather than photography or photo-compositing.
0%of sample100%
02
68%
set imagery against a dark or near-black background, using contrast to make the focal element pop.
0%of sample100%
03
66%
carry explicit series name and book-number labelling in a secondary type band, visible even at thumbnail scale.
0%of sample100%
04
61%
anchor the composition around a single iconic element — a wing, sword, snake, or silhouette — rather than a narrative scene.
0%of sample100%
05
57%
set the title in a large serif or decorative display face; thin geometric sans-serifs are essentially absent.
0%of sample100%
06
53%
introduce at least one gold, fire-orange, or metallic warm accent against the dark ground.
0%of sample100%
Method · direct visual inspection · Amazon Fantasy Bestsellers Top 30 + Goodreads Best Epic Fantasy Top 8
RomanceAmazon US Best Sellers in Romance Top 50, 2026-Q2

Romance

Source / Amazon US Best Sellers in Romance Top 50
● Refreshed 2026-Q2
01
78%
are illustrated or painterly — photographic couple covers now cluster only in the sports-romance niche.
0%of sample100%
02
70%
are title-dominant — oversized typography is the hero of the composition, not the figure.
0%of sample100%
03
66%
set the title in a chunky serif; geometric sans-serif titles are essentially absent (4%).
0%of sample100%
04
64%
center on a single ornamental object or symbol — a crown, dagger, rose, dragon — rather than a couple or scene.
0%of sample100%
05
56%
use a solid flat color block as the background, with dark-solid being the single most common treatment (34% of all).
0%of sample100%
06
54%
sit in a dark palette overall — warm and cool accents punch through a dark base, not a soft pastel one.
0%of sample100%
Method · direct visual inspection · Amazon US Best Sellers in Romance Top 50
CookbookAmazon Best Sellers — Cookbooks, Food & Wine Top 50, 2026-Q2

Cookbook

Source / Amazon Best Sellers — Cookbooks, Food & Wine Top 50
● Refreshed 2026-Q2
01
84%
set the cover on a light or bright background — the inverse of the dark grounds that dominate fiction covers.
0%of sample100%
02
74%
lead with a photograph — a finished dish, ingredients, or the author/chef — rather than illustration or pure typography.
0%of sample100%
03
46%
anchor on a single hero subject — usually one plated dish shot close or overhead — instead of a multi-dish grid.
0%of sample100%
04
28%
use a multi-dish grid or collage, a pattern concentrated in beginner, diet, and appliance (Instant Pot / griddle) cookbooks.
0%of sample100%
05
12%
feature the author's or chef's face on the cover — a celebrity-/author-branding convention almost absent from fiction.
0%of sample100%
Method · direct visual inspection · Amazon Best Sellers — Cookbooks, Food & Wine Top 50
PoetryAmazon Best Sellers — Poetry (n=44), 2026-Q3

Poetry

Source / Amazon Best Sellers — Poetry (n=44)
● Refreshed 2026-Q3
01
89%
wear a muted, monochrome, or otherwise restrained palette rather than bright saturated color.
0%of sample100%
02
66%
build the cover around open negative space — one small motif or the title alone in an empty field — instead of a full-bleed image.
0%of sample100%
03
64%
carry no photograph at all, reaching for illustration, hand-lettering, or pure type as the cover image instead.
0%of sample100%
04
50%
are type-led — the title or the poet’s name is the dominant element, with imagery kept small or dropped entirely.
0%of sample100%
05
36%
place a human figure or portrait anywhere on the cover — so roughly two-thirds show no person at all.
0%of sample100%
Method · direct visual inspection · Amazon Best Sellers — Poetry (n=44)
Methodology

How this dataset is built.

Direct visual inspection of current category bestseller lists, coded by hand against a fixed set of axes. Observational, not causal — these are patterns on the shelf, not evidence that any pattern causes sales.

332
Covers coded
324 Amazon · 8 Goodreads
Sources
6
Genres
2026-Q3
Latest sample

Evan Kane is the founder of MakeMyBookCover and runs the cover-archetype research behind the product — coding category bestseller lists by hand for every genre in the Bestseller Cover Visual DNA Index, and re-coding them as the lists move, to keep the genre-signal models honest.

How to use this

Cite it, or go deeper on one genre.

Every number on this page is reference-first — link to it, quote it, or check it against your own shelf. For the full write-up and worked examples per genre:

Cite this page

The citation, and the licence.

Kane, E. (2026-Q3). The Bestseller Cover Visual DNA Index. MakeMyBookCover. https://makemybookcover.com/book-cover-design-data

These figures are published under a Creative Commons Attribution 4.0 licence. Reuse them anywhere, including commercially, with attribution to MakeMyBookCover and a link back to this page.