The data behind the trend

Is September actually breakup season? What two years of data say

By Matthew Firth, founder of NexSpark · Aug 20, 2026 · 6 min read

September Theory, and a lot of the “breakup season” content around it, treats September as a documented spike in relationships ending. We wanted to know if that was true before we kept repeating it. So we pulled two years of search-interest data on actual breakup-intent queries, and read the two most-cited studies on relationship and divorce timing. Neither one shows a September spike. What they show instead is more specific, and more useful than a viral calendar trick.

This page is the data behind our September Theory piece, which we edited to stop repeating the “recognised breakup window” framing once we'd run this check.

The claim we checked

“September is breakup season” shows up constantly in trend coverage, usually stated as settled fact: the summer ends, obligations lift, and couples use the gap before the holidays to end things. It's a good story. It also turns out to be one nobody seems to have actually checked against real breakup behavior before repeating it. So we checked it two ways: against real search behavior, and against the two studies people actually cite when they want a source for it.

What real breakup-intent search behavior shows

“September theory” the search term is a completely different signal from people actually searching to break up with someone. So we pulled both, in the U.S., weekly, from August 2024 through August 2026 (Google Trends, relative search interest, not raw volume). The gap between them is the whole story.

What the actual published research says

Two studies get cited most often when people want a source for “breakup season.” We read both. Neither one says September.

The 2010 study, run by David McCandless and Lee Byron for The Visual Miscellaneum, scanned Facebook status updates for the phrase “we broke up because” and charted the frequency across the year. The peaks they found were in the run-up to the December holidays and after Valentine's Day and spring break, not September. In fact, as NPR's write-up on the chart put it directly: “if you turn this graph upside down you will see that August and September are the ‘We're Together Peak’ months. After that, the trouble begins.” The most-cited breakup study on the internet found the opposite of the September claim.

The other commonly cited source is a 2016 study by University of Washington sociologists Julie Brines and Brian Serafini, presented at the American Sociological Association's annual meeting. They analyzed 14 years of actual divorce filings in Washington state (2001–2015) and found a consistent, replicated biannual spike in filings in March and August, a pattern that held up when they checked it against four other states with different demographics and economies. Their explanation: winter and summer holidays function as a “domestic ritual” calendar, family-optimism periods that couples don't want to disrupt, followed by a wave of filings once the holiday doesn't fix what was broken. September isn't in the pattern at all.

So where did “September is breakup season” come from?

Not from research. It's downstream of a real and separate thing: the 2026 “month theory” trend, where each month of the year got assigned a viral relationship narrative (June theory, July theory, August theory, and so on). September's version, which we cover in full in our September Theory piece, is really a claim about Labor Day acting as a relationship checkpoint, not a claim that September itself causes more breakups. Somewhere in the retelling, “a checkpoint people notice in September” became “September is breakup season,” and the second version is the one that stuck, repeated by enough sites that it now reads as settled. Our own chart above shows exactly when that happened: the trend term itself didn't exist in the search data until July 2025.

What's probably true instead

There's a narrower, defensible version of the claim buried inside the exaggerated one. For people who are already leaning toward ending a relationship, the specific stretch between the end of summer travel and the start of the holiday season removes some of the friction, and some of the excuses, that would otherwise delay a decision they'd already half-made. That's a real, small, individual-level effect. It is not the same as a population-wide spike, and the two studies above, plus our own search data, say the actual spikes land at different points on the calendar entirely: right before Christmas, right after Valentine's Day, and in March and August specifically.

Sources

• David McCandless & Lee Byron, Facebook status-update breakup timing chart, for The Visual Miscellaneum (2010), as covered by NPR.
• Julie Brines & Brian Serafini, University of Washington, Is divorce seasonal? UW research shows biannual spike in divorce filings, presented at the American Sociological Association annual meeting (2016).
• Google Trends, U.S. relative search interest for “how to break up with someone” and “september theory” (Aug 2024–Aug 2026), and “divorce lawyer” (2021–2026), pulled August 2026.

Figures are rounded. See each source for full methodology. Journalists and writers are welcome to cite this page and its charts, just link back so others can find the sources too.