← Back to Marketing Analytics Consultants

Case study · Market and audience research

The audience finding that changed a publisher’s mind

A seasonal illustrated series was pitched to a national publisher. The category was assumed to be a niche — pleasant, small, not worth a multi-volume commitment. The pitch was going nowhere on enthusiasm alone. So the question became a research question: how big is this actually, and who is already paying attention?

1. The category is not small

Halloween is measured every year by the National Retail Federation, which surveys around eight thousand consumers. The numbers are public, and nobody in publishing appears to look at them.

$13.1BUS Halloween spending, 2025 — a record
+12.9%on the year before
73%of consumers take part
$114.45spend per person, also a record

Chart 1

Total US Halloween spending, 2022–2025

Total US spending$10.6B2022$12.2B2023$11.6B2024$13.1B2025 2025 is a record, and up 12.9% on the year before.

The 2024 dip is real and worth noting rather than hiding — spending fell from the 2023 record before recovering sharply. A category that can drop a year and then set a record is behaving like a healthy consumer market, not a fad.

Source: National Retail Federation annual survey, conducted by Prosper Insights & Analytics.

Chart 2

Where the money goes

2025 spending by categoryCostumes$4.3B71% buyDecorations$4.2B78% buyCandy$3.9B66% buyGreeting cards$0.7B38% buy More people buy decorations than costumes — and spend more on them than on candy.

Decorations are the surprise. More people buy them than buy costumes, and they spend more on them than on candy — $4.2 billion, which is $1.6 billion more than in 2019. Decoration is a display behavior, and display behavior is driven by looking at pictures of other people’s displays.

2. What people actually do

Spending tells you the size. Activity tells you whether there is a subject worth writing about.

Chart 3

Participation by activity

Share of consumers taking part, 2025Hand out candy66%Dress in costume51%Decorate51%Carve a pumpkin46%Throw or attend a party32%Visit a haunted house24%Dress up a pet23%

Two thirds hand out candy, half dress up, half decorate, and nearly half carve a pumpkin. A quarter visit a haunted house. These are not passive purchases — they are practices, with history behind every one of them. That is what a book about the subject is for.

Chart 4

When the season really starts

When Halloween shopping begins 49% 51% September or earlier October Half the season is over before October begins. For a book, that moves the publication window forward by a full quarter — an August delivery for an August release the following year, which is exactly how the series is scheduled.

Half of all Halloween shopping happens before October. For a publisher this is a scheduling finding as much as a marketing one.

3. Where demand actually begins

This is the finding that mattered most, and it is one line in the NRF survey that nobody quotes.

Chart 5

Leading source of Halloween inspiration

Leading source of Halloween inspiration37%Online search27%Retail stores21%Friends and family Search beats the store. That is the whole argument for a book about this subject.

Online search is the single largest source of Halloween inspiration — ahead of walking into a store, and well ahead of friends and family. People go looking. They type questions. They want to know where this came from and why we do it.

What this means for a book

A category where the primary demand behavior is searching for information is a category that wants reference material. The audience is not waiting to be sold a book — it is already asking the questions the book answers.

4. The audience nobody had counted

The market data establishes that the category is large. It does not establish that anyone is paying attention to the specific subject. So the second half of the research was a census: find every social property dedicated to Halloween history, film and culture, and size it.

Chart 6

The assumed audience against the measured one

What the category was assumed to reach “a niche readership” What counting the platforms actually found 247,000,000 combined following across 484 platforms Bars are to scale. This was a census of what already existed, not a projection.

Four hundred and eighty-four platforms, counted and sized individually. Not an estimate and not a projection — a count of what already existed.

484platforms identified and sized
247Mcombined following
73%of the country already takes part
37%start with a search

Put the two halves together and the argument writes itself. A $13.1 billion category. Nearly three quarters of the country participating. The largest single source of inspiration is online search. And 247 million people already following accounts devoted to the subject.

None of that is a niche.

5. What happened

The conversation with the editor changed the moment the figures appeared. It stopped being a discussion about whether people like the subject and became a discussion about a measurable audience the category had simply never counted.

Before the researchAfter
“A pleasant niche subject”A $13.1 billion category with 73% participation
Audience assumed, never measured247 million followers across 484 counted platforms
Demand assumed to be impulse at retail37% of demand begins with a search
One speculative volumeA six-volume commission
Outcome

A six-volume contract with a national publisher, scheduled from 2027 through 2032. The research did not make the books better. It made the case that the books should exist.

What this kind of work is for

Every figure on this page came from published survey data and a platform-by-platform count. No proprietary access, no privileged information, nothing that any publisher could not have looked up.

They did not look it up. That is usually the finding.

Start here

What would this look like for your category?

Send a web address and we will send back three specific findings, in writing. No meeting, no pitch, nothing attached.

Prefer email? info@marketinganalyticsconsultants.com