March 11, 2017

Librarians’ knowledge and attitudes about print-on-demand: an informal study

Print-on-demand printing, or POD, is a technology that allows publishers and individuals to have books printed one-at-a-time, instead of doing a whole print run of hundreds or thousands of copies at a time. The per-unit cost is higher, but there can be a savings in not having to deal with warehousing a large inventory or dealing with remainders. POD has ushered in a new era of self-publishing, so many people associate it with self-published books exclusively. It has also allowed small companies to make money by printing a large catalog of out-of-copyright classics from Project Gutenberg. I noticed recently that some catalogers on the AUTOCAT list used the term POD interchangeably with this kind of publishing, which I knew to be a misunderstanding of how the technology is currently used in the publishing industry. I think it is important for librarians to understand how POD is being used, for reasons that I will get into, so I decided to do a study to test librarians’ knowledge of POD. I designed a survey to find out what librarians think about POD, how knowledge that a book is POD would affect their treatment of books that are printed this way, and how they believe they can tell if a book is POD when they encounter it. I ran the survey and have some results that I will share here. It was an informal survey, and I am not using the kind of statistical techniques that you would find in an academic journal article, but I think the implications of the study turned out to be very clear nevertheless. In this report, I will say a bit about my methods and share some numbers.

First, some facts about POD in today’s publishing industry:
– Traditional publishers are now using POD publishing to maintain a large backlist, to economically publish academic titles that are expected to have low sales figures, and to deal with sudden spikes in demand.
– Inks, paper, and binding have improved to the point where POD paperbacks are superior in quality to paperbacks printed traditionally a few decades ago, and are not easily discernable from traditionally-printed books by their physical attributes.
– Since traditional publishers are using POD, POD books are often professionally designed and edited. The printing method doesn’t necessarily imply anything about the content of the book.
– POD books are not necessarily hard to catalog, if they come from a traditional publisher. Catalog records are readily available, often from the Library of Congress’ CIP program. Although the CIP program is not intended for POD books, some publishers ignore this, and many academic publishers will do an initial print run and switch to POD when demand has subsided, without calling it a new edition.
– Ordering systems such as GOBI, Title Source 360, and Oasis often do not show any indication that a book is POD. To be sure of this, I tested three titles I know to be POD, one from the University of Georgia Press and two from Library Juice Press. In all three ordering systems, the University of Georgia Press title was not identified as POD. In the case of the two Library Juice Press titles, one title was indicated as POD in one of the three ordering systems but not the other two, and one title was missing from one of the ordering systems and not identified as POD in the other two. (The ISBNs for these titles are 978-0-8203-3891-0, 978-1-63400-021-5, and 978-1-936117-01-7.)

I had a conversation with the director of the University of Georgia Press, Lisa Bayer, about their use of POD publishing. Lisa had the following to say about POD:

“Most if not all university presses are using POD for their monographs and trade paperbacks. It’s very widespread and allows us to manage our inventory investment. We use the POD programs at Lightning Source (Ingram), Baker & Taylor, and Amazon… We certainly don’t do our illustrated art books or jacketed hardcovers in POD. It’s best for paperbacks and printed case library editions. The quality has improved tremendously in the past few years. I’m not aware of vendors designating POD editions; one of the benefits of putting a title into the system is that it is always ‘available.'”

As I mentioned, I noticed that in discussions on the AUTOCAT list some librarians did not seem to be aware of these facts, and as you will see from the numbers, my survey bore out this impression. Some details on the survey:

I asked five questions; two were demographic and three were designed to gauge librarians knowledge and opinions about POD, as well as their ideas about how to identify POD titles. The demographic questions were to find out what kind of a library the respondent was from and what kind of position they held in their libraries. The other questions asked for some words they associated with POD, how knowledge that a book is POD would affect their treatment of it in their jobs, and how they know a book is POD if they think they know that it is. The questions that asked for words that they associate with POD and how POD status would affect their treatment of a book revealed value judgments, while the question about how they know if a book is POD revealed whether they were knowledgable about POD or clearly held misconceptions.

Since this was an informal study that did not pretend to have the rigor of scholarly research, I went ahead and coded these responses myself, without worrying about such issues as intercoder reliability and the like. A future study on this could use more rigorous methods.

In total, there were 408 responses. 27 respondents stated that they did not have knowledge of POD and couldn’t answer the questions. 37 gave responses that were too difficult to understand to use in the study. 32 respondents understood the notion of POD in a different way, interpreting it as a service offered by the library. While not incorrect, that is a different understanding of the term POD than I was intending to study, so I disregarded those responses. So, of the 408 responses, 312 were from people who had clear ideas about POD (whether or not they gave opinions about it). Of those 312 people, 271 showed in their answers that they held mistaken ideas about POD and how it is used in the industry, while 40 understood POD more or less correctly. That means that 87% of librarians surveyed have incorrect ideas about POD. That is a strong result, and somewhat disturbing, as it has implications, in some cases, for how librarians do their jobs.

I also coded responses to see whether respondents viewed POD positively, negatively, or neutrally. Of the 312 otherwise usable responses, for 40 of them I couldn’t really tell how they felt about POD, leaving 272 usable responses. Of 272 usable responses to these questions, 198 librarians viewed POD negatively (73%), 12 viewed it positively (4%), and 62 viewed it neutrally (23%). I counted responses as neutral if they indicated that a book’s status as POD would not affect their treatment of it in the context of their work.

In terms of comparing librarians in different settings, the demographic questions didn’t reveal much. The data on the work role of the respondents was not usable, because such a high number of librarians worked in more than one role, and the combination of roles didn’t follow a strong pattern. The data on type of library showed that the majority of respondents were in academic libraries. A comparison across library types showed no real difference as far as their understanding. Of the 408 responses, 234 were from academic libraries (170 usable); 103 were from public libraries (86 usable) and 64 (55 usable) were from a mix of other types of libraries and institutions (school, corporate, government, non-profit, consortia, vendors, and LIS programs). This is good for the study, from my point of view, because it is primarily in academic publishing where traditional publishers are using POD.

So the data from this survey show that librarians mostly have an inaccurate understanding of POD, as well as viewing it negatively in a way that affects their treatment of POD books in their job roles. Let’s look a little more closely at the responses themselves to see what we can learn, since the questions were open-ended. Bearing out my initial impression from discussions on the AUTOCAT list, most librarians associate POD books with self-publishing and reprints of out-of-copyright works, and they also think that the signs indicating that a book is self-published or a reprint are what tell them that a book is POD. Common reasons given for identifying a book as POD were ugly covers, poor layout, poor editing, no ISBN, no catalog records available, known self-publishing imprints like CreateSpace, Authorhouse or iUniverse, and a lack of publishing dates. Many librarians who gave this type of response said that they could “always tell” when a book was POD. So that raises the question, aren’t they effectively just biased against self-published books and reprints? If they don’t recognize a traditionally-published book as POD when it crosses their desk, what difference does it make if they have a negative view of POD? I think the answer is that it is not impossible to know that a book is POD even if it is of high quality and easy to catalog. As several respondents pointed out, most POD books have printing on the back page with some arcane data:

Some catalogers and others look for that tell-tale information and may draw conclusions from it about the book as a whole. Also, although ordering systems may not usually give an indication that an academic title is POD when it is, sometimes they do give that indication. Some librarians in an acquisitions role, as they stated in their responses, will avoid purchasing a book on that basis. This means that librarians’ decisions are sometimes affected by a bias against POD that is not borne out by the facts about POD.

For those interested in the raw data from my survey, you can look at it here

I’d like to make a final note of full disclosure. I am personally interested in this issue because I am a former librarian who owns a book publishing company that uses print-on-demand printing. We primarily publish library science titles, so I view our customers as my colleagues as well. It bothers me that many librarians would have a bias against our titles if they knew about how we print them. We use POD because in our field sales volumes are typically low. I decided to go with POD when I formed my company after extensive conversations with a head of another publishing company that publishes a lot of LIS titles. In those conversations I gleaned important and useful information about the publishing business, and I learned that most publishers in the LIS field are using POD, for the same reason I decided to – being in a small, niche market. Because of the stigma attached to POD, however, publishers typically don’t reveal this fact. I hope that with this report I can do a little bit to de-stigmatize this common method of printing books.


  1. Our problem is with publishers who print out Wikpedia articles and sell them or just print out inferior titles.

    I don’t see a problem with POD if it for reputable books that might not be available otherwise.

    Thanks for doing your study.

    Comment by Scott Cohen — March 12, 2017 @ 1:59 pm

  2. When I first started cataloging PODs long when they were very new in the publishing trade, I was very impressed by the fact that authors could get their books published economically and reach out to their readers.
    Very informative survey finding.

    Comment by Prema Desai — March 14, 2017 @ 2:38 pm

RSS feed for comments on this post. | TrackBack URI
You can also bookmark this on or check the cosmos

Leave a comment

XHTML ( You can use these tags): <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong> .