It’s a vexing question. I like to explain ONIX for Books by separating out the standard (what you track as metadata) from the technical side (XML, which can be thankfully ignored in this post). ONIX as a standard is a way for two companies to exchange information about books without ambiguity by using published definitions. Using the standard involves those companies agreeing between them on what level of detail they need from the ONIX data fields to support their systems and sales. That simple case is the start, but almost all companies supply or receive ONIX files from a number of sources. Once a community of data suppliers is willing to support certain details, the community of data aggregators will begin to accept it. And once enough data is available, the aggregators are motivated to invest in programming to take a new data element into their systems because it should increase their sales or their market knowledge. Once their systems are able to take in a new data element, they can take that information from any of their data suppliers. Better data from some creates use and a need for better data from all.
On a macro level, better metadata—or what we’ve been calling “enhanced” metadata (including reviews, prizes, descriptions and more)—can help publishers differentiate their products and allow consumers to make a better buying decision. On a more micro level, granular metadata—parsing out the information into smaller component parts—means greater accuracy and flexibility in use and display for the end user. It’s as simple as supporting a clear subtitle in addition to the main title and then supporting indexing by isolating the title prefix. Or splitting contributors out as people and corporate entities, breaking down a person’s name into component parts, and providing an ISNI identification number. Such granularity done consistently across a supply chain can support all sorts of good things: Indexes work better, consumers can find authors, and deeper data linkages are made between systems.
There’s no clear line between “granular” and “macro” data. Yesterday’s granular data becomes today’s macro data by use. No one (who has implemented it, at least) would call the Related Products section of ONIX “granular” data today. But branding is a new data point that shows how ONIX develops. Better support for branding has been added to ONIX 3.0 with limited support in ONIX 2.1. If you’re selling a math learning aid and have the rights to put a trademarked character on it, you’ll want parents to put those two things together. Finding a branded product matched to the right content can increase sales. At this level, a “brand” is closest to a granular addition to a product title, but it allows retailers to group and present all products associated with a third party’s identification.
Related Products has been fully developed in ONIX for years. It was identified early and is used heavily by most publishers and all retailers. Branding has been argued about for years, but only recently has a good way to trade information about it been available. I’m not aware of any use of it yet, but I know it’s needed because SalesData tells me publishers and retailers sell branded products.
Metadata is not static, but changes steadily to try to answer the identified needs of the supply chain—it grows because of need and our ability to better define the business requirements of publishing and sales.
What’s the difference between 2.1 and 3.0?
A publisher need to describe their products fully—and that advice is not specific only to ONIX—because thorough product description are needed for selling the product. If you imagine an idealized publisher with a database that fully describes everything about their products, then the difference between ONIX 2.1 and 3.0 is accuracy and completeness. Both standard versions track author, title and supply, allow for marketing information and subjects. What ONIX 3.0 allows for, largely, is improved accuracy and granularity of data. In other words, the data sender can add more clarity and definition to their records. OK, ONIX 3.0 changes are more complex than that. There are new ways to present series and sets, much greater attention to accuracy in rights and supply, an explicit need to include ROW statements, and of course ebook metadata is handled more completely with bells and whistles from rental through enhancement. But, mostly, it’s the same data with more potential and accuracy built in.
I like to present ONIX as a shopping list of values you might consider supporting. Every publisher and each of their books are unique. Not every book record needs to match the same shopping list, but if you want to understand how to place your book in the market there is no better place to look than the ONIX for Books standard. Supply chains from around the world have contributed to it. Looking to increase sales of your ebooks? Have you considered support for accessibility? That might be worth as much as a 10% increase in sales in the long term. And the fact that it’s a supported value in ONIX means you should think about it because others have already acted on this.
All products have contracts, restrictions, prices, suppliers and markets. ONIX 2.1 handles these adequately for local markets and a bit ambiguously internationally—particularly ambiguously given the way many publishers currently present their data for local markets. ONIX 3.0 assumes an international market and supports it fully. It assumes you have access to full product information and makes it harder to provide partial information. If you are only supporting a local market, in ONIX 3.0 you’ll need to say that clearly and in a way that your book record can be distinguished from one for another market.
Implementing ONIX 3.0 means that your company understands how it wants to sell its products and that it fully understands each product in an international context. That’s true even if you only sell in Canada—the new twist is that you really need to say it in the record. That’s not a big change. Having that understanding you should review the ONIX 3.0 standard to see what you might have missed or not considered and to make sure you understand how to communicate what you need to say in it.
If you need help interpreting the ONIX 3.0 standard, or if you’ve identified a value you need to communicate to help you sell products that doesn’t appear to be supported, contact email@example.com. Help is available.