U.S. commercial printing establishments from 2010 to 2022 (the most recent year for which we have data). This data series, organized by NAICS (North American Industrial Classification System), is from County Business Patterns, an annual Census Bureau publication that also includes number of employees, first quarter payroll, and annual payroll.

We include the following commercial printing NAICS:

• 323 (Printing and Related Support Activities)
• 32311 (Printing)
• 323111 (Commercial Printing, except Screen and Books)
• 323113 (Commercial Screen Printing)
• 323117 (Books Printing)
• 32312 (Support Activities for Printing—aka prepress and postpress services)

Five- and six-digit NAICS are subsets of the larger thee-digit NAICS.

Establishment counts are also broken down by establishment size:

• 1–4 employees
• 5–9 employees
• 10–19 employees.
• 20–49 employees
• 50–99 employees
• 100–249 employees
• 250–499 employees
• 500–999 employees
• 1000 employees

As an example of how to read this data series, in 2010 there were 14,533 establishments having 1 to 4 employees in NAICS 323, which had declined to 12,851 by 2016.

These data are provided to assist WhatTheyThink members in their understanding, forecasting, and planning related to trends in the North American commercial printing markets.

About the Data

Why are these data important? Demographics is an aspect of most all business decisions, especially for B2B sales and marketing. The most common question in sales is “how many prospects and customers are there?” which is essentially a question about demographics—that is, the number of establishments in a particular business category.

Beyond quantifying potential prospects, the number of business establishments is an important measure of economic activity. Once a year, the Census Bureau publishes establishment data. A basic question we should answer at the outset is, What's an “establishment,” and how does it differ from a “firm”? It’s not technically accurate, but think of an establishment as a location. A single firm can own many establishments. For example, a company like Wal-Mart would have multiple establishments, although it’s only one firm. There are all kinds of technical exceptions, but that’s the general gist of it.

A second question is, Where do the data come from? There are two reports issued by the government that provide the demographics of businesses and organizations. This data series comes from County Business Patterns, which is an annual report based on the filings of Social Security and Medicare taxes by businesses with employees. (Businesspeople will know this as form 941.) Most large businesses file and submit payments with every payroll, but smaller businesses may file quarterly. The County Business Patterns report is based on the first form 941 of the year, which includes the count of employees as of the week of March 12.

There is a second report called Nonemployer Statistics, which counts sole practitioners, such as accountants or lawyers—or, in graphic communications, freelance designers, photographers, and even self or independent publishers. What these businesses have common is that they do not have employees per sebut pay their Social Security and Medicare taxes with their annual tax returns. These include proprietors, partnerships, schedule S corporations, LLCs, and other businesses without employees, as well as freelancers and “moonlighters” who have regular jobs but also have outside income from extracurricular activities. They therefore do not use form 941, which is what County Business Patternsis based on. Ergo, Nonemployer Statistics. The overlap is very small in the grand scheme of things. In the case of NAICS 323—general commercial printing—Nonemployer Statisticsis not applicable, except perhaps for “hobbyist” printers who, say, run a letterpress shop in the garage to produce limited-run wedding invitations and the like. In general, though, when looking at printing establishments, we don’t worry too much about these businesses, except when we start talking about the creative markets (graphic design, ad agencies, PR, and publishers).  

These reports have detailed breakdowns of data by primary business classification. Who does these classifications? Typically, the business or the business’s tax accountant. For most of these businesses, they self-classify when they file for employer identification numbers (EINs) or file their tax returns. And then every five years, businesses participate in the Economic Census and provide detailed information about their businesses that adds further detail to the data. In the case of the printing industry, businesses report in a general category, and then report their sales in specific categories, and are classified by the Census on that basis. It must be noted that this is not a perfect process, and that inaccurate business classifications can go on for years before correction. For example, in the 1970s, many letterpress printers had shifted to offset, but their data were being collected in the old category. More recently, “commercial printers” may offer a full range of marketing services and, whether or not they consider themselves “marketing services providers,” may or may not classify themselves in a manufacturing NAICS (323) or a service NAICS (541). Businesses that come into the printing industry from legacy NAICS (photolabs now offering wide-format commercial printing, for example) may or may not be counted in a commercial printing NAICS or their original NAICS. These conundra are why industry analysts don’t sleep very much.

Statisticians do the best they/we can to resolve these issues. At the same time, beware of data that imply an undeserved level of precision. Greater detail introduces a wider range of error, so be wary of overly granular data.

Still, County Business Patternsis the best available data with a collection methodology that includes possible imprisonment for lack of compliance, so that’s a motivating factor that many of us who conduct surveys wish we had at our disposal. One downside of this series is that the data are at minimum two years out of date. (The good news is that this is a significant improvement over what it used to be.) That may sound like these data are not especially helpful, but that’s not the case at all. Trends often take place over a substantial span of time and while it’s true that technology changes very rapidly, the fact remains that demographic trends and changes take some time to manifest themselves. Looking at industry changes from 2010 to 2019 is still a useful barometer by which to gauge what is happening vis-à-vis establishments however nice that might be. And, as we always say in our forecast reports, it is only by looking at the past that we can predict the future. Even the best forecasting software can only ever “predict the past”—that is, use historical data to extrapolate the future.

Background on the NAICS system can be found on the Census website, and users who wish to look beyond NAICS 323 can search by specific business classifications. Bear in mind, that over time WhatTheyThink will be expanding the range of business classifications for which we provide data.