Garbage in, garbage out

In the software industry, the expression garbage in, garbage out is well known. It means that no matter how sophisticated a software system you build, its success depends on the data it holds. If you feed lousy input data to it, the output will be lousy as well.

Think of a brand new, top-notch espresso coffee maker: in spite of its high pressure, temperature control and built-in adjustable coffee bean grinder, you’ll still get a very poor cup of coffee, if you use inferior quality coffee beans.

The coffee bean image fits the relation that many publishers have with their book product metadata. Despite the investment in new title management systems and systems for metadata handling and distribution, the metadata creation activity itself is often seen merely as tiresome recording of administrative facts about the books, and hence not as the valuable marketing effort that it is.

The activity is typically left to assistants who fill-in a minimum necessary set of metadata to let the book go through production and distribution. This must be due to a general lack of knowledge about the power and growing importance of metadata. And that is why you will want to communicate to your organization what good metadata actually can accomplish for book marketing and sales downstream in the value-chain.

(You can read about the use and benefits of book product metadata in various reports, e.g. from BISG or BIC)

If you want readers to be able to find your books  when they look for them online, your metadata need to deliver clear identification of products and authors, and they need to have a search engine optimized textual description as well as detailed and precise subject classification.

If you want to let readers select their format of preference, or if you want your books to be recommended by algorithms, you’ll need your metadata to communicate all possible information about a book’s relations:

  • in which other formats (EPUB, audio, paperback, …) is the book available?
  • – do any earlier, commented, abbreviated or translated editions exist?
  • which are the other books in the same series, and what is their sequence? In particular, readers would want to know exactly which book is next in the same series.
  • – which are the other books by the same author(s)?

Furnishing this information to retailers, will let them guide readers to exactly those of your books, that they will be most likely to buy. And who wouldn’t want that? One could argue, that a publishing house owes that to the authors whose interests it attends to.

But if the metadata you deliver do not have the necessary quality, how can they provide the discoverability you wish for your books? How can retailers use the metadata to maximize sales?

So you’ll need the relevant people in your organization to assume responsibility for exactly those parts of the metadata set that they own, be they from editorial, production, marketing or sales. Metadata are best and most effectively created as close to their source as possible, and therefore the responsibility must be distributed across functions.

To make this happen, somebody has to tell the involved people about the actual downstream use of the metadata and how their efforts can mean a great difference for the short and long term sales of a book. Most of us need to understand the goals of our activities and to feel that what we are doing is meaningful.

It might help to see your metadata as if they were babies: They need a safe home and they need caring adults to be responsible for them. All the care you give them while they are growing up will definitely make them stronger and more confident and reliable. And that will all help them to succeed and achieve their goals later in life.

That should be well worth the effort.