Nearly 100% of Publishers Have Seen E-Booksellers Get Their Metadata Wrong

Print Friendly

Woman reading a book in a bookstore where metadata is hardly ever wrong.

By Jeremy Greenfield, Editorial Director, Digital Book World, @JDGsaid

Imagine if when a book hit Barnes & Noble store shelves it had a different cover than when the proof left the publisher.

That’s what nearly all publishers are experiencing with their e-books, but in a digital way.

According to an upcoming study from the Book Industry Study Group set to come out in a month, 95% of publishers have had the experience of creating their e-books with one set of metadata and seeing an altered set of metadata at the point of sale, online booksellers like Amazon, Barnes & Noble and Apple*.

Initial findings from the study, presented at the BISG Making Information Pay 2012 conference in New York today, suggest that many publishers have little control over their e-book metadata and little knowledge of when and where it gets processed and changed.

Metadata is a buzzword in digital publishing today. Publishers know they need to do it right, but there seems to be a poor industry-wide understanding of what exactly it is.

Think of it as the digital version of everything you would find on a physical book’s book jacket: author; title; ISBN number; blurbs from luminaries praising the book; the book cover; a summary of the content and description of the book; an author’s biography; and so on.

Why it’s so important for e-books is that when a consumer lands on a book’s page at an online book retailer, the metadata is what tells them what the book is about, who it’s by and generally indicates whether they should buy it – just like a book jacket.

(To be sure, there is some metadata that is not completely analogous to what is on a book jacket or on the inside pages of a book, but that stuff is less important by and large than spelling the author’s name right.)

Many publishers, relatively new to e-books, aren’t keeping tabs on such information. Almost half of publishers use an automated system to check that metadata is appearing correctly on online booksellers’ sites; a little over a third check manually and about a fifth don’t check at all, according to the study.

For those publishers that are aware that their metadata has been altered, about half don’t know where in the e-book supply chain the data was altered – at the publisher, the e-book converter, the distributor or the e-bookseller. That’s a big problem for publishers.

“When something is wrong at retail, it can be hard to fix because it isn’t clear where the change took place,” said Brian O’Leary of digital publishing consultancy Magellan Media, who is authoring the study and made the presentation today. “If you don’t fix it at the source, it keeps coming back.”

As part of the upcoming study, BISG and O’Leary will make recommendations for how the industry should treat metadata. One clear recommendation: “There has to be a better discussion of how metadata is edited; it can’t be a black box,” said O’Leary.

The BISG study was conducted by Magellan Media between February and April among a cross-section of just over 100 publishers, wholesalers, service-providers, retailers and manufacturers. About of a quarter of the responses were gathered by phone and the rest online.

Write to Jeremy Greenfield

Woman in bookstore photo via Shutterstock

* Editor’s note: This sentence has been edited slightly to reflect a more accurate portrayal of the story per Brian O’Leary’s comment below.

Related Posts:

13 thoughts on “Nearly 100% of Publishers Have Seen E-Booksellers Get Their Metadata Wrong

  1. We now have 72 titles available at Cool Gus Publishing and we’re constantly having to go to various outlets and check not only the metadata, but also customer reviews as people will note if there are formatting problems or other issues. It’s a full time job redoing every title every few months not just with new metadara for SEO, but also links to new titles.

  2. The question in the survey asked publishers if their metadata had been “altered”, not specifically if it was “wrong”. Almost every publisher reported having seen their metadata altered. When I answered you, I should have been clearer about this. My fault for not taking an extra couple of minutes to re-read my response.

    One thing that I think I was clear on: publishers don’t necessarily know where their metadata is altered. In that light, “e-booksellers” are not necessarily responsible for altered metadata. Changes can be made upstream from a retailer and simply presented as “what we were given”.

    Survey questions were asked about both physical and digital books. The headline here might be interpreted as e-books alone, but responses to both the interviews and the survey indicate that metadata is changed for both.

    • Thanks for the comment, Brian.

      Having re-read the piece, I think everything is accurate except the “altered” versus “wrong” thing. I will be going in and making a small change.

      Also, if I could do it all over again, I would perhaps change the headline from “E-Booksellers” to “Online Booksellers”

      Thanks so much for your great presentation and I look forward to the full report when it comes out.

  3. Anyone who works in the ebook retailer/vendor world will be screaming at this article. Not necessarily because it is wrong but that it ignores the major problem with poorly represented bibliographic data in the marketplace; that the original metadata is incorrect or inconsistent with the agreed industry standards. Or that many publishers pay far less care to how they represent an individual book through its metadata than many schoolchildren spend on representing this years lemonade sale.
    For the article to just say that retailers are the problem in altering metadata ignores the main problem in the industry, that publishers are consistently bad at creating and preparing the key data associated with their one and only product. It is at worst ignorant of the key role of metadata in securing the best sales and potentially derelict in their duty of representing and protecting their authors interests appropriately.

    • Just to be clear, there is no blame being assigned here. All the article is saying is that publishers have observed that the metadata gets altered at some point in the process and many of them don’t know where.

  4. Part of the problem is that each ebook retailer has a different idea of what some of the basic metadata pieces are. Take a simple one: author. Every retailer requires that the author field be filled in. Many REQUIRE the author name be split into FIRSTNAME and LASTNAME. But what do you do when the author is The Society for the Study of American Culture to name one example? Or when the author is Cher? If the author’s pseudonym is Captain Kangaroo, should Captain be the first name, and Kangaroo be the last name, even though universally people will look to alphabetize that under C and not K? Does Kermit the Frog go under Kermit or Frog?

    Now add the fact that almost every trade book published has both a title and subtitle, but pretty much none of the major ebook retailers makes any allowance for subtitle as a separate data category. Your only choice is to treat it as part of the title, leading to some unwieldy titles out there.

    The lack of metadata standardization across ebook retailers has meant our metadata database has five separate fields for author name, configured all different ways, and many other duplications to try to satisfy all the quirks. The alteration of our metadata into something wrong often has to happen on our end before we can even upload, in order to fit what weirdnesses a given retailer might require.

  5. We’ve seen data changed months after upload. One specific example is with Amazon and they change our publisher name from our name to someone else. I believe, this happens in part because we published a backlist title that while the previous publisher has taken off its for sale, it hasn’t removed the data from whatever distributor they are using to push the information to Amazon. What is bothersome to me as a small publisher is the idea that we have a completely different ISBN than the previous publisher, obviously, but that that the vendor, in this case Amazon, changes the publisher name. Amazon has been good about fixing this (its happened only with one author but on all her books), but its one of those things that needs to be address.

    I agree with the lack of standardization in ebooks. We upload direct to the 4 major platforms and each one is done very differently with different information required.

  6. It is true that matter of the e-books changing from their actual matter. This only because of the reader and then commenter. Mata Data of course changed again it is because of the reader’s thinking then the thinking become the part of that e-book. Just take it like writer wrote the e-book and then publisher nothing to do with it as he is only a publisher.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>