Finally, book printing in the bookstore is becoming reality

By | February 18, 2009

For at least two decades, people have been talking about putting book printing right in the bookstore. The approach has the potential to solve a lot of problems: the bookstore (and its customers) can select from a very deep inventory, without having to have all those books (and multiple copies of many of them) on hand. Returns are essentially eliminated. There’s no warehousing, and no shipping costs. The main problems have been: creating reliable hardware and systems that are easy to use but produce high-quality books, convincing publishers to go along with the scheme, and making it sufficiently economical to buy and run the equipment.

It looks like all the pieces may finally be falling into place. On Demand Books (, with their Espresso Book Machine, seems to have come up with the right hardware. The company says the machine is printing 100 books per day in the bookstore of the University of Alberta. Their deal with LightningSource (signed in April 2008) might be the key to getting access to a huge volume of books that the Espresso could print in the stores. And, with the cost of the hardware between $50,000 and $100,000 and the consumables cost per book at roughly a penny a page, the economics are at least plausible for a fairly large bookstore.

Officially, the ten sites (including bookstores and libraries) where the Espresso is installed are beta test sites. But the results are promising. Assuming On Demand Machines can eventually replicate the Alberta success in dozens of other locations, I think this marks a change in the book printing world even more profound than the change initiated by LightningSource ten years ago, when it proved that producing single copies of books, on demand, could be profitable.

There is more on the Espresso Book Machine and its meaning for the industry at Beyond Print.

Share this post


9 thoughts on “Finally, book printing in the bookstore is becoming reality

  1. Eliot Harper

    The large bookstore chain Angus & Robertson started dropping these Espresso book machines in their Australian stores last year. They plan to have a total of 50 installed in their stores within the year. Read about it at: You can review the list of on-demand titles the bookstore offers (using Espresso) at

    There’s a good video on the machine at I’m rather curious to find out the cost per impression and duty cycles are, as they have just bolted on a couple of letter-size desktop laser printers, which, by design aren’t built for large volumes.

  2. Michael "PDF boy" Jahn

    My twins are high school seniors this year, and since they are taking AP classes, very few of the books they needed were in the school bookstore – worse, they are both weigh about 100lbs each, and you should see the size of these books. The Physics book alone is heavier than a bowling ball.

    I bore you with this point to make another point – they have to bring this back and forth. They only use a small section of this book each week. They were back-ordered. They cost a small fortune to ship – and worse, after this semester they are pointless to keep (so, we need to re-sell them)

    Even if every Sir Speedy, Kinkos or Quick printer had one of these Espresso devices, the issue is NOT the output gadget, it is the business of distribution – where is the iTunes Store for school books?

    never mind that Angela and Nicole DO NOT WANT TO CARRY these books back and forth to school (they both have Amazon Kindles) – THE BOOKS ARE NOT AVAILABLE for download for display on their Kindles or iPhones nor are they available for printing on some Espresso Book Machine, as these types of books are not available to LightningSource.

    I am not sure WHAT type of books you are speaking of – lets say we are speaking of paperback novels — these are normally Impulse sales – yes ? Such as I might buy a books at the Airport ? Who wants to wait for the silly machine to print ? Shopping is brisk business – I am browsing, I am flipping through the pages, I am buying, I am outa there. The same is true when I stop into Borders.

    Now – When I hear about a book on NPR, I go to Amazon. Even if Amazon “sent” my order to a local Kinkos, would they have a taxi drop it to my door, use USPS – or ask me to drive over and pick it up ?

    I think this sounds like Kodaks PhotoCD – wishful thinking that a stop-gap transitional technology might be a winner – which ended up more like strapping bumpers and windsheilds to a horse.

    I would rather focus on Amazons Whispernet. I think the authors – like musicians – need to pay attention to how books will transform to eBooks once the distribution model changes.

    I don’t need “on demand books” – i need on demand content.

  3. Chris

    For textbooks, especially expensive, heavy ones, digital delivery and use is a good method. But for a wide variety of other books the Espresso will serve well. I have one of the early Espresso machines at my bookstore and I can tell you that as content becomes more available it will get used more and more. There are dozens of times a day when I wish I could print a book that a customer asked for. With 100,000 titles being published every year and more than 2 million “in print” and many millions more that are “out of print” there is a vast ocean of possible content that many people want in book form. And the Espresso will be able to deliver it to them. It is really about customer service – I want to be able to supply as many of my customer’s desires as possible and the Espresso opens up possibilities beyond what was previously conceivable…

  4. Noel Ward

    As John McEnroe would say, “You GOT to be kidding me!”

    Call me cynical, but I have some doubts about how these machines will hold up as volume printed increases and when maintenance (which strikes me as being exception processing in a retail store) is not done in a timely and adequate manner. Never mind training hourly retail employees to operate and do basic maintenance on this device. Or is it devices, since there are two printers and a bindery unit?

    Moreover, this system looks almost Rube Goldbergish to me and the claims of a book every 60 seconds have to be a complete fabrication. I counted 5 seconds per sheet for printing, so it would be a darn thin “book.” To have this really work you need a production-class machine with excellent duplexing capability and a commercial-grade binding system. And that probably won’t fit in a box suitable (or affordable) for a retail store. Not yet, anyway.

    I think books on demand remains a great concept, and feeds the ridiculous consumer demand for instant gratification, but as things like Kindle become mainstream (no I don’t have one), I think on demand books is less of an answer to some imagined demand than it is just doing something because it’s possible to do it.

    But as I said, I’m being cynical.

  5. George Alexander

    Noel, you are right to be skeptical. The people at On Demand Books and their predecessor company have been working on this for at least 15 years, with little to show for it (except an impressive collection of patents) during most of that time. Still, if the machine is really churning out 100 books a day in a bookstore environment, that suggests to me a relatively mature device.

    The printing engines inside the machine are off-the-shelf office devices from either Kyocera (lower volume) or Xerox (higher volume). They presumably require the same amount of maintenance as any office printers would. The paper transport, assembly, and binding process is where On Demand Books adds its value. I have seen the transport and binding system of the prototype, and it looks quite rugged, for what that’s worth (not much).

    The speed depends on the exact printer used. “A book every 60 seconds” is certainly wrong. I have never heard anyone quote that speed, and it is clearly not part of the current spec. In the speech I linked to at, Jason Epstein claims four minutes per book for the faster (Xerox-based) configuration.

    As for “affordability and suitability,” the company has always talked in terms of sub-$100,000, which would have a very solid ROI at $3 profit per book, if the machine is fed enough book-printing jobs to average, say, 50 books per day. (Finding those jobs could be the hardest challenge of all.) I would think the machine is approximately as “suitable” for a bookstore as a copier.

    For me, the key questions have always been (1) will the bookstores be able to find enough print jobs to justify it and (2) will the machine perform adequately in a bookstore environment. It looks like we are getting close to an answer to both questions, and the initial indications are positive. That was my reason for posting the story in the first place.

    Still, I too remain skeptical. We’ve been waiting a long time for this technology.

  6. Andy McCourt

    I’m with George and Noel…nothing wrong with printing books on-demand but with this technology and at this speed and quality inside a bookstore?? Here’s a report on the first one into A&R’s bookstore in Australia as mentioned by Elliot: (Search Espresso in seach box and take the first story)
    We will not encourage booksales by turning out shoddy pieces.
    Get the quality and speed up – especially the binding – and maybe version 2.0 will be more promising. That will mean proper re-setting of type and not scanning pages as images.
    Mono books from digital engines are growing at 14% CAGR, but almost all from production-strength quality providers (such as LSI mentioned) with proper bindings.

  7. Noel Ward

    A couple more thoughts. I had the original On Demand Machine Company (of which I am assuming Expresso may be a stepchild) pick my brain several years ago and I helped connect them with some people at Xerox and elsewhere. They seemed to be folks with a great vision but unrealistic expectations. This does not seem to have changed much, although I don’t know the details of this deal or if the same folks are involved.

    This device may be their current model, but what seems lacking are the marketing and production sides of the equation. To make this truly viable and profitable for all involved, deals have to be inked with publishers, authors and cover designers for rights. The delivery mechanism for getting titles to the machines on a near immediate basis has to be worked out. Then the machine has to be sold to booksellers with service agreements. And finally the newfound ability of the bookseller to produce titles in-house on demand has to sufficiently promoted to get consumers to visit a particular store to buy a book. This could be a differentiator for a book store, but I question whether there is enough demand to make it work financially.

    I think the demand for instant gratification is vastly overrated. People rarely need a given book immediately. If my local store doesn’t have what I want they can usually get it in a day or so. If it is out of stock at the distributor, I find Amazon will list it as “In Stock” and have it digitally printed and shipped to me in a couple of days.

    There are some people who do need a book instantly, won’t care that it may be less visually appealing, and is not as durable as a book that is manufactured by more robust processes. But are there enough such readers to provide a good ROI and the income stream needed to justify investment in this device? I suspect the answer is no.

  8. Andy McCourt

    Noel said: “But are there enough such readers to provide a good ROI and the income stream needed to justify investment in this device?”

    My guess is no: the main advantage of printing books on demand is for the publisher/bookseller, not the customer; with the possible exception of out-of-print and esoteric books. HarperCollins estimates 40% of what it ‘sells’ comes back and have started an imprint – HarperStudio – that does not accept returns.

    But as Noel says, who is standing in line waiting on agitated nerves for a ‘must have’ book? If it’s that good, a million copies have been printed offset and distributed already. My view of the book-of-one site is not in retail, but in nearby zones, with established digital book printing equipment, where next day delivery is promised. This means better quality than Espresso can currently deliver. Italian firm CEM is about to release its ‘DocuConverter’ which is the bookmaking back-end for Xerox and Oce high volume, professional cut-sheet digital printers. This will make a difference, but I don’t see it in retail. Put it there if you will, but the floorspace may be better allocated to POS displays or, as Borders has done, a Gloria Jean’s coffee franchise!

Comments are closed.