The back-page revival, 2004 to 2006

Between November 2004 and the middle of 2006, almost every English-language daily newspaper in the world added Sudoku to its back page. By the time the wave finished, the puzzle was running in The Times, The New York Times, The Guardian, the Sydney Morning Herald, the Toronto Star, and dozens more. A few held out for another year. Most had folded Sudoku into the standard back-page rotation alongside the crossword, the comic strip, and the bridge column.

Eighteen months is a fast wave for a daily-newspaper feature. Most back-page additions take years — the slot is finite, the editorial committee is conservative, and a new feature has to displace something that someone on the staff has a relationship with. Sudoku displaced a lot, and fast. The story of who the launch belonged to is told in who actually invented Sudoku; this piece is about the conditions on the receiving end — the newspapers, the back-page economics, and the cultural moment that made the puzzle viable in 2004 in a way it wouldn't have been in 1994.

The back page in 2003

A typical English-language daily newspaper back page in 2003 contained a crossword, a comic strip or two, a horoscope or weather column, sometimes a bridge or chess problem, and, depending on the paper, a sudoku-shaped number place or logic puzzle feature in tiny print that almost no one solved. The crossword took up most of the puzzle real estate. It had been running for forty or fifty years on most papers, was set by an in-house or syndicated cruciverbalist, and varied in difficulty by paper but not in shape.

Crossword culture had changed slowly across the late twentieth century. The cryptic-versus-standard split between British and American papers was settled. The major broadsheets had famous crossword setters; the tabloids ran simpler grids. Reader engagement was steady — crossword pages were among the most-read pages in any paper — but the engagement wasn't growing, and the cost of producing or syndicating a daily crossword was non-trivial.

This is the back page Sudoku walked into. Established, mature, and quietly under cost pressure.

What was changing in newspapers

The early 2000s were the years when newspaper economics started visibly to break. Internet news, classified-ads-to-Craigslist migration, declining circulation and the first big rounds of newsroom layoffs all hit between 2001 and 2005. Daily papers were looking for cost reductions everywhere, including in syndicated content.

A daily crossword — printed in a national paper, clued for the home market, set by a paid setter — was an unsurprisingly expensive feature in this climate. A daily Sudoku, in contrast, cost almost nothing. Wayne Gould's generator could produce thousands of difficulty-graded puzzles overnight; he licensed it to newspapers on terms that were, by published-content standards, near-trivial. The cost-per-day for a Sudoku was a fraction of the cost-per-day for a crossword, and the reader-engagement signals from the early launches suggested it was more, not less, popular.

For an editor in 2004 looking at the back-page budget, the maths was straightforward. The same back page real estate could host one new daily feature for less money than it cost to keep the second crossword running, and reader engagement might go up rather than down. That's the quiet driver underneath the wave's speed. It wasn't only that Sudoku was a great puzzle. It was that the great puzzle was also, in print-economics terms, almost free.

The contagion

Once The Times of London launched on 12 November 2004, the rest of the British nationals followed within months. The Daily Telegraph, The Guardian, The Independent, the Daily Mail — all running Sudoku by mid-2005, most of them within twelve weeks of the Times launch.

The American adoption took slightly longer because the US market is bigger and slower to copy a single competitor's move, but the trajectory was the same. The New York Post added Sudoku in April 2005. The Washington Post and the Los Angeles Times through that summer. The New York Times added it in 2005, eventually behind a paywall, but the speed of adoption surprised even Gould — he had expected the US to take a year or two longer than it did.

Australia, Canada, and English-language papers in India followed through 2005. The non-English market was happening in parallel, with each country's flagship daily picking it up on roughly the same six-to-twelve-month timeline. By the middle of 2006, not running a daily Sudoku was the surprising editorial decision; running one was the default.

What crosswords lost

The space had to come from somewhere. In most papers, Sudoku didn't replace the main crossword — it sat alongside it. What it replaced was usually the second crossword (a quick or beginner version), a third puzzle feature like Word Search, or some part of the comics or horoscope real estate.

The aggregate effect on crossword culture was modest in the short term and meaningful in the longer term. The total number of crossword setters working in commercial newspaper journalism declined through the late 2000s, partly because of the wider newsroom contraction and partly because the secondary-crossword slot that had employed many of them was now a Sudoku slot. A few celebrity setters retained their main-crossword posts; many lower-tier setters lost work. The crossword-as-craft survived in the high-end papers and dropped out of most others.

Long-time crossword editors talk about the period with a slight wistfulness. The puzzle wasn't replaced; the room around it shrank.

Why it couldn't have happened earlier or later

A useful counterfactual: would the wave have happened in 1994, with the same puzzle and the same generator? Probably not. In 1994 newspapers weren't yet under the cost pressure that made the Sudoku-economics decisive. Reader engagement on crosswords was higher relative to the eventual decline. The cultural moment around brain training hadn't arrived — Nintendo's Brain Age launched in 2005 and 2006, riding the same cognitive-fitness wave Sudoku rode. We've written about the research behind those claims, and the short version is that the marketing outran the evidence, but the marketing was real and Sudoku slotted into it cleanly.

Would the wave have happened in 2014? Probably also not, in the same shape. By the mid-2010s the back-page economy had been further compressed, and the cheap-syndicated-content slot was being filled by aggregator algorithms rather than by puzzle features. The window between newspapers had room for the puzzle and newspapers had room for nothing lasted maybe a decade; Sudoku arrived at the start of that window and rode it cleanly to global recognition.

The puzzle is a good puzzle. It's also a puzzle whose adoption story is mostly a story about print-journalism economics and a cultural moment that briefly made cheap, language-independent, vaguely-cognitive content the most valuable category of back-page feature available. Take any of those three legs away and the wave is slower, smaller, or both.

A small footnote

The funny consequence of all this is that Sudoku, which most readers think of as a digital puzzle now, owes its global recognition almost entirely to print. The newspaper wave of 2004 to 2006 built the cultural awareness that the apps and websites of the 2010s monetised. Without the back-page slot, Sudoku would still exist — it would still be a Nikoli puzzle, it would still be on the back of a few American Dell magazines, and Wayne Gould would still have his generator. But it wouldn't be the household name it became, because the household-name version is the version that ran every morning next to the comic strip for three years before any of the apps existed.

The back page made the brand. The apps just rented it.