Episode 21: Thad McIlroy on Metadata Quality, the Backlist, and more

In this episode I am joined by Thad McIlroy, an electronic publishing analyst and author based in San Francisco. Thad writes in-depth articles on publishing topics for his own website, The Future of Publishing, as well as for Publishers Weekly. We started off the conversation talking about the need to understand metadata from a strategic perspective, not just from a clerical or practical perspective, even as far up the food chain as the publisher or CEO.

With that strategic understanding in place, the next step for any publishing company is to experiment with your metadata and find what works for your books. Every company is different, and every list is different, so it is important to put in the time and effort to see what will be most effective for you.

Thad also provided his thoughts on the value of addressing metadata on backlist titles, choosing BISAC subject codes, creating keywords, and the differences between discoverability, findability, and conversion.

I highly recommend you check out Thad’s website, The Future of Publishing, and read his article, “We Need to Talk About the Backlist”.

Transcript

Joshua Tallent 

So this week on the BookSmarts Podcast, I’m excited to have Thad McElroy. Thad is an electronic publishing analyst and an author based in San Francisco and Vancouver. He has a site called The Future of Publishing that provides in-depth coverage of the book publishing industry. If you haven’t seen any of Thad’s articles, you need to go take a look and read what he writes. He is a partner in Publishing Technology Partners, and an adjunct professor in the Masters of Publishing program at Pace University in New York. Hey, Thad, thanks for joining me.

Thad McIlroy 

Oh, thank you, Joshua, love to get together and talk metadata with you.

Joshua Tallent 

Oh, yeah, this is gonna be fun. I, I love talking about metadata. And you’re one of those people that, you know, we should be able to talk about metadata forever, although we’re gonna, we’re gonna limit it, we’re gonna keep it shorter. But yeah, it’s the kind of thing that I think we need to talk about more. So we were chatting before we started recording about how metadata is handled by publishers and what we think some of the big issues are. So what do you think one of, just picking an issue, that you think publishers need to be thinking about when they think about metadata?

Thad McIlroy 

Let’s go to fundamentals. And, you know, where, what’s sort of the state of metadata today? So if you go into an average publisher, and you say “metadata”, well, when I started in this, you know, the conversation ended right there. It doesn’t end there anymore, you know, you get a degree of familiarity, you get a degree of comfort, but what you don’t get is a comprehensive understanding. They have, in my view, a very superficial view of metadata, because it’s technical, and they’re not technical people.

Joshua Tallent 

Yeah. So it’s kind of underappreciated in many ways.

Thad McIlroy 

I think so.

Joshua Tallent 

So obviously, large publishers get that, right? A lot of large publishers have very large metadata teams, they have people who are dedicated to dealing with metadata issues and thinking about how the data goes out. I think you’re 100%, right, though, that especially midsize and smaller publishers tend to kind of run aground on the issue. And think that it’s a bigger problem—like something beyond my scope, or something beyond my ability. And that kind of puts a blockade in the mind to keep them from wanting to pursue it too deeply, because they’re just worried about how deep you have to get.

Thad McIlroy 

Yeah. One of the things I’m thinking as you’re describing that, it’s it’s, you know, is metadata clerical or is it strategic, and there’s, you know, I think probably is the hole they fall into, they think they can hire a clerk to do it, and just fill in those little boxes. And having filled in those little boxes, they sit back and feel satisfied. And my, you know, major push to all of them is you have to think of this as strategic, you have to realize that there is an ROI. And I try and point out to them that metadata is not merely quantitative, it’s qualitative. You know, you can say to someone, make sure you’ve got a description of the book, well, of course, they’re going to have a description of the book. But what does that description say. And when I look at the average publisher’s description, it’s very passive, because they grew up creating flaps, you know, creating the hardcover flap, or the back cover. And those tended to be, you know, written in a passive form. And the description that’s going to sell a book has to be very active, it has to convert.

Joshua Tallent 

Yeah, and this is that whole idea, I keep on saying constantly that, you know, your metadata is your most important non people resource. It is the thing that as a publisher, you know, it—obviously you pay your employees and you have some sort of system in place for making sure they get their PTO and they get insurance and all these other things we deal with on an employee basis. But if we were to do the same kind of thing with our metadata, we would put a lot more emphasis on it in the publishing world. We would have systems and processes and, you know, people engage with it, it would be the kind of thing that gets more attention, but unfortunately, it just doesn’t get that nearly enough.

Thad McIlroy 

Yeah. I think when—the publishers I’ve consulted to I say to them, you know, this needs to be understood by the President, it certainly needs to be understood by the Sales Director, it certainly needs to be understood by the marketing manager, it absolutely needs to be understood by the PR manager. And in so many cases, you know, one of my little sporting activities is I go into the job boards, and I look at descriptions of the marketing manager, no mention of metadata, sales manager, no mention of metadata, and you think how are you going to hire a sales manager who does not understand the role of metadata in the overall sales and marketing package? It’s to me that’s ludicrous.

Joshua Tallent 

Yeah, it really is. And publishers that don’t have that are then missing out on all the opportunities that come with better quality data for their PR team and for their—you know, if you don’t have that upper management knowing enough about it to know that it’s important and to be able to direct their staff in, hey, go research this issue or go find—let’s do some digging on how we can handle these issues we’re seeing in our data. You don’t have to know everything. It’s not like, you know, those strategic people need to be thinking about how do I—how do I change the HTML code? And this, you know, that’s not necessary on their level, but definitely on the strategic level thinking through how that’s going to impact. And that comes to data. Right? So when we’re thinking about, you know, from a strategic perspective, as a publisher, how do I figure out what’s important? How do I even know where to start? What would you say to a publisher who’s trying to figure that part out? They know, it’s important, they’ve been convinced by our little conversation here that, hey, I, I need to be thinking about this. What would you say first and foremost, for someone in that sales manager position, that PR, manager position? What do they do first, to kind of just get a baseline understanding of what’s important or all that detail?

Thad McIlroy 

Now, let me take it and go a little bit in a different direction with that. There’s so much that’s important. Right? So, you know, we’ve been through the years of them coming to an understanding of the basics, you know, the basic fields are reasonably well understood. You know, it’s not surprising to tell someone that reviews, you know, posted on Amazon are going to have a positive impact. I think part of the—that reminds me that you say that someone “Reviews have a positive impact in metadata.” And they say, That’s not metadata, those reviews, and you say no, everything that describes the book, everything that amplifies the message in the book, every tweet, is a piece of metadata. And I argue to them that, in fact, the book itself is a, you know, meta instance of the metadata of itself, it really gets their eyes crossing, but it is every word in that book, you know, every paragraph in that book is another opportunity via Google Books, for example, you know, people will, in searches get to a book on Google Books because of the content within there. So, you know, there’s a fundamental sense and the—at that point, I have to take them to the next level. And you say to them, that there are these fields that are less obvious. So how are we going to approach that? Well, we have to approach it experimentally, you know, you have to commit to finding out what works best for you. And that means setting up a number of tests and trying, you know—isolating which books you wish to impact, and then trying different methods, because there’s, you know, lots of different costs and degree of difficulty involved.

Joshua Tallent 

Yeah, and also, for each publisher, it might be different just based on what you publish, and, and what market you really care about. If you’re a publisher that doesn’t care about Amazon, you’re going to take a very different approach to your metadata than someone who is only selling or, you know, 95% of their sales happen on Amazon. It’s a very different. Yeah, and the same thing for if I’m a fiction publisher, exclusively fiction, or if you’re a, you know, an education publisher, it’s a really different approach to data for those kinds of those kinds of differences.

Thad McIlroy 

It sure is, that’s a really good point. Bringing up fiction, you know, in my, let’s say, metadata practice, I just skip right over fiction. There are things that can be done, absolutely, but if I’m trying to get you know—if a publisher only does fiction, obviously, we have to come to terms with it, but most publishers that publish fiction have a lot of nonfiction as well. And I say to them, you know, we’ll get to the fiction. There are things we can do but the nonfiction is 10 times as malleable and subject to influence by enhancement of metadata. And the other thing is to this distinction between frontlist and backlist: frontlist has a very different metadata status because there’s active things going—you know, reviews are appearing, whereas with the backlist reviews are no longer appearing. So you’ve got different kinds of data that are impacting the frontlist versus the backlist so my emphasis to publishers to say, let’s go deep into the backlist. Let’s find the books that still have activity and we’re going to move that activity forward. If you’ve got a book that has no activity, we’re going to ignore that book. It’s not worth trying to bring it back from the dead.

Joshua Tallent 

Yeah, and for anyone who’s listening for the first time to the podcast, or or haven’t had a chance to listen to some of the more recent ones. I would recommend you go back and listen to the last podcast episode when I talked to Todd Satterstern about the idea of frontlist versus backlist and where you know how to build out that that frontlist so you can get to the backlist, and some of the other episodes where I’ve talked about similar topics. The backlist—and Thad you actually have a really awesome article that I’ll link in the show notes about the backlist and how important that is, and kind of digging into some of the details on how backlist works. I agree 100%. I think—I keep saying this, you know, 69% of all book sales according to NPD in 2021, were from the backlist. And we have a real need in the industry to rethink this idea that every, you know, 95% of our work happens in the frontlist. You know, if your publishing company is getting so much revenue from that backlist, and so many units are being sold in the back list, it makes sense to focus more attention on that and try to figure out how to increase that. Even a small increase across 100 backlist titles can make a big impact on your bottom line.

Thad McIlroy 

I always say that to people, you know, you—run the math, a 10% increase, if you’re going to put it across 100 titles, you’re going to get an ROI. Absolutely. And that’s—10% that that’s a cinch for us metadata folks. You know, I can do that one in my sleep. My target is, you know, can I move it by 30%? Can I move it by 40%? That’s the fun challenge for me. And I believe I can, you know, with a robust program, but with even, you know, quite minimal changes, I think, you know, an afternoon spent with the BISACs and keywords is going to get you your 10% right away.

Joshua Tallent 

Yeah. Let’s talk a little bit about some of these specific metadata areas that you focus on when you’re when you’re talking to a client about, you know, metadata and fixing metadata issue. So BISAC codes, what are some of the recommendations that you tell people about getting subject codes on their on their product data?

Thad McIlroy 

Number one, accept that it’s complicated. You’ve got 6,000—upwards of 6,000 choices in the BISAC directory. And so you know, there, you want to look at that BISAC listing and become familiar with the degree of granularity that is available on the BISAC universe. It can be tremendously granular for certain topics and other ones, you know, there’s only a top level heading. So you know, you can—you have to focus on trying to find the ultimate degree of granularity, you have to accept the fact it may not be available. How do you choose the BISACs? I think the easiest first step is to get the comparables: Find the titles that you feel you compete with most directly for any particular backlist title—and for sure, you’re competing with someone, and probably you’re competing with four or five, you know, other titles—and look at what they’ve done with your BISACs. And look at what Amazon has done to change their BISACs as well. I think, you know, the proviso there is that because publishers, a lot of publishers have difficulty—you’re looking kind of for consensus, when you go and look at other publishers’ assignment or BISAC codes. In many cases, they don’t have three codes, and when they do have three codes some of them are ludicrous. So don’t treat that as gospel. It’s just—it’s like a conversation with those other books as to what the best BISACs might be.

Joshua Tallent 

Yeah, that’s good. And there’s also the issue of what codes are—what subject categories are available on the retail websites. Amazon doesn’t follow BISAC, right? They’ve got their own little thing going on. So as a publisher, looking not just at, you know, at the BISAC codes that were provided by those other publishers for those comp titles, but also looking at where those titles show up on Amazon might be beneficial as well.

Thad McIlroy 

Great point. Yeah.

Joshua Tallent 

And then when you’re thinking about the value of you know, as a publisher, digging into BISAC, one of the things I will recommend as well to publishers is they look at those categories where there aren’t enough subcategories in them—enough sub topics in them. And if there’s enough value, enough benefit in having some sort of subcategories in that you can reach out to the BISG Subject Codes Committee and request that that be added as a subject category in the next update to BISAC, because BISAC is updated every year. They’re gonna look into the value of that across the industry, not just for one publisher. But if you get enough support behind adding new BISAC categories that can actually help a lot. You know, if you’re digging into, you know, publishing books in a certain category, you’re not seeing enough depth in that list, it is possible to expand that.

Thad McIlroy 

Yeah, I think that’s a good point. Of course, it takes a little while because they do it every summer, that new list comes out. [It actually usually comes out every December] I think the requirement I saw at one point is you need to find 100 titles within a proposed new category. That’s not a lot. And I agree, you know, if you’re a successful backlist publisher, you have a subject focus or multiple subject focuses. And if you don’t, you’re not a very good publisher. So you have these multiple subject focus areas that you know that you really try and plumb more successfully than your competition. And as you’re saying, you know, if you find out that you’re—there isn’t an available classification for some of your key titles, yep, you’ve got to go to the BISAC committee.

Joshua Tallent 

So what about keywords? We talk about keywords all the time in the metadata world. What would you recommend for publishers who are trying to figure out what to do to really engage their keywords?

Thad McIlroy 

That’s another tough one, right? At first glance is so simple, right? Just think of some of the words that describe this book and put them in the keyword field, you’ve got up to 250 characters. Don’t repeat words, is a fairly obvious recommendation, but they do, right? The keyword stuffing is shameful to me. There’s a little utility, you probably use the BookNet utility so you can see what other people are doing with keywords. And that’s important, again, in the same way that you want to look at the BISACs, you want to see what other people are doing with keywords. But Pete McCarthy was the first one who alerted me to the notion that your keywords appear throughout the metadata that’s on, let’s say, Amazon, so you know, the title is keywords. And more significantly, the description is keywords, the author bio is keywords. And so don’t waste the—when you get to the keyword field, recognize that you’ve already used some prioritized words in those descriptions, and try to go very carefully and see what else is missing. You have to be thinking through what are the main levers that drive a sale of a book and make sure you’ve got keywords associated with those levers.

Joshua Tallent 

That’s an interesting thought. I know that Pete talks about the keywords in the description, and that the keywords are actually indexed there. I have never seen that actually happen: The keywords in a book description are—words in a book description are indexed in search on Amazon. So it’s one of those interesting things. Chris Sim from Kadaxis has done a ton of research on this, and he swears that there’s no indexing of the book description, that the keyword field itself in your ONIX file is where Amazon is looking for that kind of search information. That’s an interesting topic that I think, obviously we can’t come to, we can’t decide on necessarily, but I’d be interested to hear from the audience, if you’ve done any research yourself about your own titles, to see what kind of difference you might have using keywords in your book description, you know, versus putting them into the keywords field. That’d be interesting to see some data or some experience on that. So keywords and BISAC and a lot of other fields that you put into your data have, in some ways to do with discoverability and have in other ways to do with conversion. Can you talk about your thoughts on the differences between discoverability, and findability, and conversion and these kinds of issues?

Thad McIlroy 

Absolutely. Findability—let’s start with findability and discoverability. A lot of people—the word discover and discoverability has been tightly associated with metadata for a long time. I think people have a kind of vague idea of what they mean by discoverability. Because it really does assume a zero state, right? That someone’s coming in not looking for your book and discovering it. If they’re looking for your book, they’re finding it. You have metadata that makes it easy to find the book that people are looking for. So, certain metadata, you know that the more rigorous kind of clerical metadata, tends to be the stuff that makes a book findable. And that’s absolutely, you know, essential, because a certain number of people are indeed looking for that book. But when you go into the deep backlist, in many, many cases, it’s more important to have discoverability. So people are going into Amazon and saying, you know, I want a book on Greek cooking, let’s say. Of course, there’s a gazillion of those on Amazon, it’d be very hard to optimize your book against the competition. But let’s assume there were only 100 books on Greek cooking, but you want to be the one that gets discovered ahead of the other people. And you don’t have to be number one, you know, if you’re discovered in the top three to five, that will have an impact on sales. And indeed, you know, don’t feel you’ve failed if you don’t get the number one position. But you have to think okay, what will people be searching for, for which my book is the answer? And then at that point, we get to conversion. So let’s say we’ve—you know, either they’ve come in looking for your specific book and found it, or they’re looking for a book in that subject area, and they’ve discovered your book. Now what happens? You’re really at that point, like the clerk in the bookstore. They say, is this book any good? Is this book the one I’m looking for? Have you got other books on the same topic? And your metadata has to answer all of those questions, because there’s no clerk standing there, making the sale. The metadata has to convert from discover/find into at click, buy.

Joshua Tallent 

Yeah. And those elements that make a big difference on the Convert side, are going to be things like, does it have good images? Does it have Look Inside, right? Those kinds of things that make a difference for that person who’s coming there and doesn’t know anything about the book, they’re going to look for a book description, an excerpt, you know, they’re going to try to see is this really the book that I want? So it’s important to have those elements there, too, not just the keywords.

Thad McIlroy 

Absolutely. Yeah. You want a grand and deluxe listing, you want people to be delighted by the listing as much as you know, they understand the book is delightful, as delightful as the images and the words that they’re encountering in your listing.

Joshua Tallent 

Yeah. That’s great. Well, Thad I think we’re out of time. I appreciate you coming on the podcast today. Would you tell our listeners where they can learn more about the work that you’re doing and read the things that you write?

Thad McIlroy 

My own website is thefutureofpublishing.com, as it sounds. I do a lot of writing in Publishers Weekly, as well. So, search my name on PW and you’ll find all the recent articles: the article in the backlist, the article—I did a fun article on all the best books that I think anyone who’s interested in publishing should have on their bookshelf that appeared relatively recently. So, there’s some fun things on PW as well.

Joshua Tallent 

Yeah, that’s great. And again, I highly recommend that everyone go read all of that stuff. I think he’s got an interesting perspective and definitely an important voice in the industry. So Thad thanks a lot for joining me, I appreciate you coming on the podcast.

Thad McIlroy 

Thanks for asking me and having me onboard Joshua.

Joshua Tallent 

So that’s it for this episode of The BookSmarts Podcast. If you like what you’ve heard, please rate and review us in Apple Podcasts and Spotify and everywhere else that you listen. And if you have questions, or you have thoughts or comments about the show—feedback, feel free to email me at joshua@firebrandtech.com. Thanks for joining us and getting smarter about your books.