The average print run has decreased, but that does not mean every publisher is printing fewer copies. I look at the distribution of publishing output by print-run size in the first half of 2026 and show how a small number of high-volume editions can affect the overall picture.
The Book Industry review of the Russian book market for the first half of 2026 puts two figures side by side: the number of editions published rose by 0.9%, while the average print run fell by 13.3%. At first glance, the conclusion seems straightforward: more titles, smaller print runs.
But a lower average does not necessarily mean publishers have reduced the print run of each book across the board. The same result could occur if a few particularly high-volume editions were absent from the period’s output while the rest changed very little.
To look into this, I turned to the Russian Book Chamber’s data on the distribution of books and pamphlets by print-run size(in Russian) for the first half of 2026. The table also includes a comparison with the same period in 2025. The distribution turned out to be more interesting than the average.
Where output fell
According to the Russian Book Chamber, the total number of editions increased from 53,055 to 53,508. Total output fell from 172.54 million to 150.82 million copies—a decline of about 21.73 million. The average print run decreased from 3,252 to 2,819 copies.
| Print-run bracket | Output in H1 2025, million copies | Output in H1 2026, million copies | Change, million copies |
|---|---|---|---|
| Up to 500 copies | 5.18 | 5.14 | −0.04 |
| Over 500 to 1,000 | 3.70 | 3.50 | −0.20 |
| Over 1,000 to 5,000 | 52.59 | 53.73 | +1.13 |
| Over 5,000 to 10,000 | 22.96 | 23.17 | +0.20 |
| Over 10,000 to 50,000 | 40.80 | 34.42 | −6.37 |
| Over 50,000 to 100,000 | 6.55 | 8.06 | +1.51 |
| Over 100,000 | 40.77 | 22.80 | −17.97 |
| Total | 172.54 | 150.82 | −21.73 |
The largest decline came from editions with print runs above 100,000 copies. Their combined output fell by almost 18 million copies, while the number of editions dropped from 132 to 92.
Dividing the decline in this bracket by the decline in total output gives:
\[ \frac{17.97}{21.73}\approx 82.7\% \]In other words, about 83% of the net decline in total output came from a single print-run bracket.
Meanwhile, output in the bracket up to 500 copies barely changed. In the largest bracket by total output—over 1,000 to 5,000 copies—both the number of copies and the number of editions increased.
What the 83% contribution means
The 83% figure is this bracket’s contribution to the change in the total. We have compared two distributions and identified where most of the difference sits. We do not yet know which individual editions were printed in smaller quantities.
Nor should we read the table as though each bracket contains the same editions in both periods. An edition with 101,000 copies last year and 99,000 this year, for example, would move from the highest bracket to the one below it. All 101,000 copies would disappear from the highest bracket, even though the edition’s own output had fallen by only 2,000. The remaining 99,000 would appear in another row.
A bracket’s contribution to the net decline is also different from its share of all decreases. For example, combined output in two brackets—over 100,000 and over 10,000 to 50,000 copies—fell by 24.34 million copies, while total output fell by 21.73 million. The remaining brackets added a net 2.61 million copies, partially offsetting the decline. A share of the net decline therefore measures a contribution to the final difference, rather than a share of all decreases taken together.
To explain the causes, we would need data on individual editions: which were not published again, which moved between brackets, which appeared for the first time, and which were registered in a different period.
What counts as an average book
The distribution gives us different ways to look at the scale of publishing.
Editions with print runs of up to 1,000 copies account for about 53.6% of all registered editions, but only 5.7% of total copies.
At the other end are 92 editions with print runs above 100,000 copies. They represent less than 0.2% of editions, but about 15.1% of all copies.
So low-volume editions make up the majority by count, while a small group of high-volume editions has a substantial effect on total output.
The published average is calculated by dividing total copies by the total number of registered editions. That count also includes editions with no reported print run: there were 2,007 of them in the first half of 2026. This is another limitation of the measure. It does not automatically describe either a typical book on a particular publisher’s list or a typical order placed with a printer.
If we subtract the output of editions above 100,000 copies from each period’s total, we are left with 131.77 million copies in 2025 and 128.01 million in 2026. That is a decline of about 2.9%, compared with 12.6% for output as a whole.
This calculation shows how much the overall trend depends on the highest print-run bracket. Here, though, I am comparing the output of the remaining brackets, not the same set of books. High-volume editions are a legitimate part of publishing output, so it would be wrong to automatically treat this bracket as an outlier and exclude it from the analysis.
An edition’s total output and an order placed with a printer
The notes to the Russian Book Chamber’s table include an important detail for publishers: reprint quantities are added together.
An edition with a total of 5,000 copies might have been produced in one printing order or several. The table does not distinguish between these possibilities, but the difference matters to a publisher’s economics.
One large order and several smaller ones can produce the same number of copies, yet involve different costs, payment schedules and amounts of cash tied up in stock. A decline in the average print run in these statistics cannot be interpreted directly as a shift towards smaller production batches. That would require a history of orders placed with printers.
The statistics also cover legal deposit copies registered with the Russian State Library during the reporting period. When comparing them with internal production data, it therefore matters which date defines each dataset: registration, printing or arrival at the warehouse.
How I would check my own list
For my own list, I would split the analysis into two questions.
The first is how output by edition has changed. This requires the number of copies printed for each edition over comparable periods, with reprints added together consistently. Which books contributed to the largest increases or decreases in copies? Does the overall result depend on a few major projects?
The second is how production batches have changed. Here, the unit of analysis is an individual order placed with a printer: its quantity, cost and warehouse arrival date. How have the median order size, the range of quantities and the frequency of reprints changed? These measures would help establish whether the publisher has actually moved to smaller batches.
Mixing these levels can hide a change in strategy. A publisher might have started printing smaller batches more often, for example, while keeping the book’s total annual output unchanged.
The findings could lead to different decisions. If total output fell because of a few major projects, that alone is not a reason to revise print-run decisions across the whole list. If comparable books are being printed in smaller batches, it is worth checking whether lower stock levels offset a possible increase in production cost per copy and whether the risk of running out of stock has increased.
