In 2003, it cost ScrollWorks roughly $14,000 to print 3,000 copies of a trade paperback. That included setup, plates, paper, binding, and shipping to our warehouse. The per-unit cost was about $4.67, which meant we needed to price the book at $15.95 or higher to have any chance of breaking even after distribution fees, returns, and the author’s royalty. We printed 3,000 because that was the minimum run that made economic sense. Printing fewer copies would have driven the per-unit cost even higher. Printing more would have been reckless, because most books from a small press do not sell 3,000 copies.

This was the math that governed small press publishing for decades. It was simple math, and it was brutal. You had to guess how many copies you would sell, commit to printing that many (or more, because of minimum run sizes), and then hope your guess was right. If you printed too many, the unsold copies sat in your warehouse eating storage fees. If you printed too few and the book found an audience, you had to wait weeks for a reprint while the book was unavailable and the momentum was dying. Either way, you were making a high-stakes bet with incomplete information.

Digital printing changed all of that. And I do not think the publishing industry has fully reckoned with how much it changed, or how profound the implications are for small publishers like us.

Let me explain the difference between traditional offset printing and digital printing, because it matters. Offset printing, the method that dominated book manufacturing for most of the 20th century, works by creating physical plates that transfer ink to paper. Making those plates is expensive and time-consuming, but once they exist, printing each additional copy is very cheap. This creates a classic economy of scale: the more copies you print, the lower the per-unit cost. A run of 5,000 copies might cost $3.50 per unit, while a run of 500 copies might cost $8 per unit. The setup cost is spread across the entire run, so bigger runs are more economical.

Digital printing eliminates the plate-making step. The book is printed directly from a digital file, essentially the same technology as a very sophisticated office printer. There is no expensive setup, which means the per-unit cost is roughly the same whether you print 10 copies or 1,000 copies. The cost per unit is higher than offset at large volumes (digital printing cannot match the per-unit economics of a 10,000-copy offset run) but dramatically lower at small volumes. And, critically, there is no minimum run size. You can print one copy at a time.

This sounds like a technical detail, but it fundamentally altered the risk calculus of small press publishing. Let me walk through exactly how.

Before digital printing, acquiring a book was a financial commitment that started at several thousand dollars and scaled from there. The moment you decided to publish a book, you were on the hook for a print run, and that meant spending real money before a single copy was sold. This created a conservative bias in acquisitions. We could not afford to take a chance on a book unless we were reasonably confident it would sell enough copies to cover the print run. “Reasonably confident” is doing a lot of work in that sentence, because nobody is ever truly confident about book sales. But we tried to be, and that meant we passed on books that we loved but could not justify financially.

I think about those books sometimes. The manuscripts we turned down not because they were not good enough, but because we did not think we could sell enough copies to cover the printing costs. Some of those books were published by other houses and did fine. Some were never published at all. And some of them, I suspect, would have found their audience if we had been willing to take the risk. The economics of offset printing made us more cautious than our editorial instincts wanted to be.

Digital printing removed that barrier. Today, we can publish a book with an initial print run of 200 copies. If it does not sell, our financial exposure is minimal. If it starts to sell, we can reprint in small batches, matching supply to actual demand rather than guessing in advance. The risk of each individual publishing decision has dropped by an order of magnitude, and that has made us braver as publishers.

We have published at least a dozen books in the past five years that we would not have published in the offset-only era. Not because they are lesser books, but because their commercial prospects were uncertain enough that committing to a 3,000-copy offset run would have been irresponsible. With digital printing, we could take the chance. Some of those books found their audience and went on to sell thousands of copies through successive small print runs. Some sold modestly and that was fine. A couple barely sold at all, and our financial loss was small enough to absorb.

This shift has had a real effect on the diversity of books being published. When every book requires a significant upfront investment, publishers naturally gravitate toward books with the broadest possible appeal, because those are the ones most likely to recoup the investment. Niche books, experimental books, books aimed at small but passionate audiences, these are harder to justify under offset economics. Digital printing makes them viable. A book that will sell 500 copies over five years is not a commercial failure if you printed 200 copies at a time and your total investment was modest. Under offset economics, that same book would have been a disaster.

Now, I should be honest about the downsides of digital printing, because there are some. The first is quality. Digital printing has improved enormously over the past decade, but it still does not match offset for certain types of books. Art books, photography books, and any book that requires precise color reproduction are still better served by offset printing. The difference has narrowed, and for most text-heavy books, the average reader would not notice any quality difference. But for visually intensive titles, offset remains superior.

The second downside is per-unit cost at scale. If you know a book is going to sell 10,000 copies, offset printing is still significantly cheaper per unit. Digital printing makes sense for uncertain quantities and small runs, but once a book has proven demand, switching to offset for larger reprints saves real money. At ScrollWorks, we often use a hybrid approach: digital printing for the initial run, then offset for reprints once we have a better sense of demand. This gives us the flexibility of digital at the start and the cost efficiency of offset once the uncertainty is resolved.

The third downside is the psychology of scarcity. This might sound strange, but when offset printing forced publishers to commit to a specific number of copies, it created a form of discipline. You had to think carefully about demand, plan your marketing around the print run, and make every copy count. Digital printing, with its “print what you need” flexibility, can lead to a more casual attitude toward sales and marketing. Why hustle to sell books when you can always print more? I have noticed this tendency in myself and have had to consciously resist it. The ease of digital reprinting can make you complacent, and complacency is the enemy of effective publishing.

Let me talk about print-on-demand specifically, because it is the most radical extension of digital printing and it has its own set of implications. Print-on-demand (POD) means exactly what it sounds like: a book is printed only when a customer orders it. The order goes to a POD facility, a single copy is printed and bound, and it ships directly to the customer. There is no warehouse, no inventory, no unsold copies. Every book that is printed has already been purchased.

POD eliminates inventory risk entirely. You never have unsold copies sitting in a warehouse. You never have to write off unsold stock. You never face the painful decision of whether to pulp 2,000 unsold copies of a book that did not find its audience. From a pure risk perspective, POD is the ideal model. Publish a book, make it available, and if nobody buys it, your financial loss is limited to the editorial and design costs. The printing cost is zero because no printing happens until a sale occurs.

We use POD for our backlist titles and for books where demand is unpredictable. It keeps books available indefinitely at no carrying cost. A book published five years ago that sells one copy a month is still generating revenue through POD, whereas under the old model, it would have gone out of print long ago because maintaining warehouse inventory for a slow-selling title does not make financial sense.

The concept of “out of print” has essentially been eliminated by POD. Any book with a digital file can remain available for purchase indefinitely. This is a profound change for authors, who used to face the indignity of seeing their books become unavailable within a few years of publication. It is also a profound change for readers, who can now find and buy books that would have been impossible to obtain under the old system.

But POD has limitations that are worth understanding. The per-unit cost is the highest of any printing method, because every book is produced individually. There are no volume efficiencies at all. For a trade paperback, POD might cost $5-7 per unit, compared to $3-4 for a short digital run and $2-3 for a large offset run. This means the retail price needs to be higher, or the publisher’s margin needs to be thinner. For slow-selling titles, the higher per-unit cost is acceptable because there is no alternative. For faster-selling titles, it makes more sense to switch to digital or offset runs.

There is also a quality consideration with POD. While the printing quality is good and improving, the binding quality can be inconsistent. I have seen POD copies of the same book where the spine glue was perfect on one and slightly off-center on another. This variability is a consequence of producing books one at a time rather than in batches where quality control can be applied across the entire run. For most readers, the difference is negligible. But for a publisher who cares about the physical quality of their products, it can be frustrating.

Let me zoom out and talk about what all of this means for the economics of small press publishing in general. The shift from offset-only to a mixed model of offset, digital, and POD has fundamentally changed the financial profile of a small press. Under the old model, publishing was capital-intensive. You needed significant upfront investment for each title, and your cash was tied up in physical inventory that might or might not sell. Under the new model, publishing is much less capital-intensive. Your upfront costs are primarily editorial and design, which are largely fixed regardless of how many copies you print. The variable costs (printing and shipping) scale with actual demand.

This has lowered the barrier to entry for small publishers. It is now possible to start a small press with much less capital than it would have required twenty years ago. Whether this is entirely positive is debatable. More publishers means more competition for readers’ attention, and some of the new entrants lack the editorial standards and professionalism that the capital barrier used to enforce. But on balance, I think democratizing access to publishing is a good thing. More voices, more perspectives, more books that might not have existed under the old economics.

At ScrollWorks, digital printing has allowed us to build a catalog that is more diverse and more interesting than it would have been otherwise. Titles like The Cartographer’s Dilemma and The Last Archive were acquired with the knowledge that digital printing gave us the flexibility to manage the financial risk of books with uncertain commercial prospects. Both of those books found their readers, and both justified the bet we made on them. Under the old economics, I am not sure we would have made those bets at all.

The revolution in printing technology is not finished. Advances in digital printing continue to close the quality gap with offset, reduce per-unit costs, and expand the range of formats and finishes available. I expect that within ten years, the distinction between digital and offset will be largely irrelevant for most books. We will simply print books, using whatever technology produces the best result at the best price for the specific quantity we need. The technology will become invisible, which is exactly what technology should do. What will remain visible is the books themselves, and the fact that more of them exist, reaching more diverse audiences, because the economics of printing finally caught up with the ambitions of publishers who always wanted to take more chances.