Dating Statistics 2026
Courtship has moved online faster than the dating-app business has grown. That gap is the whole story: people are meeting in social-media messages rather than on moderated dating platforms, and that is exactly where the money disappears. Every figure below is sourced and linked.
How do couples actually meet in 2026?
The last rigorous national measurement found that 39% of new heterosexual couples met online — and that was 2017. Since then the share of courtship happening online has almost certainly kept climbing, while the number of Americans paying for a dating app has fallen. Those two facts together are the argument of this page: dating moved online, but it moved past the dating apps, into ordinary social-media messages where nobody screens anybody.
The 39% figure comes from Michael Rosenfeld's How Couples Meet and Stay Together study at Stanford, published in PNAS. In 1995 the same measure was 2%. Online meeting overtook meeting through friends around 2013 and never gave the lead back. For same-sex couples the shift was faster and further: roughly 65% met online by 2017.
How has the share of couples meeting online changed since 1995?
| Year new couple formed | Share who met online | What else was true |
|---|---|---|
| 1995 | 2% | Meeting through friends was the single largest channel |
| 2013 (approx.) | Crossover point | Online overtakes friends for heterosexual couples |
| 2017 | 39% | Same-sex couples: about 65% |
| 2017 (same-sex) | ~65% | Apps reached small and dispersed communities first |
Source: Rosenfeld, Thomas & Hausen, PNAS (2019), from the HCMST survey.
Our math: going from 2% to 39% over 22 years is a compound annual growth rate of 14.5% in the share of new couples meeting online. Nothing else about how humans pair up has changed that fast in recorded demographic history — not the car, not the workplace, not the decline of church attendance.
A second-order point that most write-ups skip: the 39% is a share of new couples, not of all couples. The stock of existing relationships still overwhelmingly formed offline, because those relationships are older. That distinction produces most of the contradictory headlines you will see, and we untangle it in the methodology section below.
How many American adults are single in 2026?
Fewer than half of U.S. households — 47% — are married-couple households in 2025, down from 66% in 1975, according to the U.S. Census Bureau. One-person households now number 39.7 million, or 29% of all households, up from 20% fifty years ago.
What do the household numbers look like across fifty years?
| Measure | 1975 | 2025 | Change |
|---|---|---|---|
| Married-couple households | 66% | 47% | -19pp |
| One-person households | 20% | 29% | +9pp |
| Median age at first marriage, men | 23.5 | 30.8 | +7.3 years |
| Median age at first marriage, women | 21.1 | 28.4 | +7.3 years |
| Families with own children under 18 | 54% | 39% | -15pp |
Source: U.S. Census Bureau, Families and Living Arrangements, December 2025.
Read this carefully: the marriage-age lines for men and women rose by exactly the same 7.3 years. The gap between the sexes did not close — it simply slid forward at about 1.75 months per calendar year for half a century. And the married-couple household figure fell 19 percentage points, which sounds moderate, but in relative terms it is a 29% decline. Most coverage reports the percentage-point figure and understates the shift.
Two further Census numbers matter for anyone reading this because they are dating in their twenties: 58% of adults aged 18 to 24 still live in a parental home, as do 16% of those aged 25 to 34. A large share of the dating population is not hosting.
39.7 million Americans live alone in 2025 — 29% of all U.S. households, up from 20% in 1975.
How many people use dating apps in 2026?
Three in ten U.S. adults (30%) have ever used a dating site or app, and only 9% did so in the past year, per Pew Research Center, which surveyed 6,034 adults. Usage is concentrated sharply by age and by relationship status.
Source: Pew Research Center
Who actually uses dating platforms?
| Group | Ever used a dating site or app |
|---|---|
| Never married | 52% |
| Lesbian, gay or bisexual adults | 51% |
| Living with a partner | 46% |
| Divorced, separated or widowed | 36% |
| Men | 34% |
| Straight adults | 28% |
| Women | 27% |
| Married | 16% |
Source: Pew Research Center (2023).
Now the part that gets left out. The dating-app business is contracting even as online meeting keeps growing. Match Group, which owns Tinder, Hinge, Match and OkCupid, reported fourth-quarter 2025 revenue of $878 million, up just 2% year on year, with Tinder payers down 8% to 8.8 million, according to CNBC's coverage of the earnings. Revenue per payer rose 5% to $17.63 — the company is charging more people less often.
What the number hides: a shrinking payer base with rising revenue per payer is not a growing market, it is a monetisation squeeze on a loyal core. Meanwhile the share of couples meeting online has not gone backwards. Something has to absorb the people who left. It is not a bar, and it is not a church group — it is Instagram, Facebook and Telegram messages, which have no dating-safety team, no profile verification and no report-a-match flow.
Pew also found what users think of the experience: 42% say online dating has made finding a long-term partner easier, 22% say harder, and 32% say it made no difference. Only 48% of U.S. adults now call online dating a safe way to meet people, down from 53% in 2019. Support for mandatory background checks on dating platforms sits at 60% overall — and at 65% among people who have never used one.
How much does dating cost in 2026?
The average all-in cost of a date in 2026 is $189, up 12.5% from $168 a year earlier, according to Bank of Montreal's 2026 Real Financial Progress Index as reported by Fortune. Annual dating spend averages $2,323.
Source: BMO Real Financial Progress Index 2026
How fast are dating costs rising by generation?
| Group | 2025 per date | 2026 per date | Change |
|---|---|---|---|
| All singles | $168 | $189 | +12.5% |
| Millennials | $191 | $252 | +31.9% |
| Gen Z | $194 | $205 | +5.7% |
Source: BMO 2026 Real Financial Progress Index, via Fortune.
The catch: millennial date spending rose 31.9% year on year, which is 2.6 times the headline 12.5% increase. Gen Z barely moved, at 5.7%. The rising cost of dating is not a general phenomenon — it is concentrated in the cohort in its thirties, the one most likely to be dating with an explicit long-term goal and most likely to treat a date as a real evening out rather than a coffee.
The independent triangulation is reassuring. Match and the Kinsey Institute's 14th annual Singles in America study, based on 5,001 U.S. singles aged 18 to 98, put average monthly dating spend at $213. BMO's $2,323 annual figure works out to $194 a month. Two surveys, different sponsors, different methods, a 10% spread — that is about as close to agreement as consumer survey data gets, and both should be read as roughly $200 a month for an active dater.
The strain shows. Forty-seven percent of single Americans told BMO that dating is no longer worth the cost. Fourteen percent now spend $300 or more on a date, up from 11% a year earlier, while another 14% spend nothing at all — the market is splitting into an expensive tier and a free one. Half of Gen Z (50%) and 40% of millennials say dating costs interfere with their financial goals, and a separate JG Wentworth survey found 86% saying money concerns had delayed or prevented dating altogether.
How common is ghosting in 2026?
Around 76% of active daters have either ghosted someone or been ghosted, according to a Forbes Health/OnePoll survey of 5,000 U.S. adults who had dated within the previous five years, published by Forbes Health. Roughly 60% report having been ghosted; 45% admit to having done it.
Who ghosts, and who gets ghosted?
| Behaviour | Men | Women |
|---|---|---|
| Admit to having ghosted someone | 44% | 47% |
| Report having been ghosted | 64% | 57% |
| Experienced ghosting specifically on a dating app | 41% of all respondents | |
| Felt upset after being ghosted | Over 50% | |
| Felt inadequate after being ghosted | 39% | |
Source: Forbes Health/OnePoll survey, n=5,000.
Ghosting is close to symmetrical by gender — women admit to it slightly more often, men report receiving it slightly more often — which quietly demolishes the folk theory that one sex does the disappearing. What is not symmetrical is the emotional aftermath, and it feeds a broader exhaustion: Forbes Health separately measured dating-app burnout at 78% of users, while Singles in America found 47% of singles feeling burned out, 54% saying modern dating leaves them drained, and 46% having taken a deliberate break to recover.
As dating fatigue grows, some younger users are exploring alternatives to traditional dating apps, including AI companions that offer conversation and emotional support without the uncertainty of swiping, ghosting, or mismatched expectations. Resources like AI Girlfriend.coach document this emerging trend by testing and comparing the leading AI companion platforms, helping readers understand how this new category is evolving.
One caution about ghosting statistics generally: they measure a self-reported behaviour with no agreed definition. A three-day silence after two messages is ghosting to some respondents and normal to others. Treat the 76% as a headline about a mood, not a precise incidence rate.
78% of dating-app users report burnout, and 47% of single Americans now say dating is not worth the cost.
Where do romance scams actually begin?
Forty percent of people who lost money to a romance scam said the contact started on social media, compared with 19% who said it started on a dating website or app — the Federal Trade Commission's romance-scam data spotlight. That single comparison is the spine of this page made concrete.
Source: FTC Consumer Sentinel
Reality check: money-losing romance scams begin on social media at 2.1 times the rate they begin on dating apps — even though dating apps are the venue people deliberately go to in order to date. An unexpected direct message from a stranger on Instagram or Facebook carries no expectation of romance and therefore no defensive posture. That is the vulnerability being exploited, and no dating-app safety feature reaches it.
How much is being lost, and how fast?
| Period | Reported losses | Reports | Median loss |
|---|---|---|---|
| 2023 (full year) | $1.14 billion | Not stated | $2,000 |
| 2025 (Jan-Sep) | $1.16 billion | 55,604 | $2,218 (Q3) |
| 2025 Q3 alone | $398 million | 11,200+ | $2,218 |
| Change vs same period 2024 | +22% | +22% | — |
Source: FTC Consumer Sentinel Network data, reported 2025-2026.
* 2025 annualised by CatfishFinder from the FTC nine-month figure of 1.16 billion. Source: FTC
Run the arithmetic three ways. Annualising the FTC nine-month total gives a 2025 run rate of roughly $1.55 billion, about 36% above the 2023 full-year figure. Spread across the 273 days those reports cover, $1.16 billion is $4.25 million a day, or about $2,951 every minute, around the clock. And 55,604 reports over 273 days is one new romance-scam report filed with the FTC every seven minutes.
Here is the number we find hardest to shake. The median romance-scam loss in the third quarter of 2025 was $2,218. The average American single spends $2,323 a year on dating. A typical romance scam therefore costs the victim about 95% of an entire year's dating budget — in one relationship that never existed.
The median romance-scam loss of $2,218 equals about 95% of what the average U.S. single spends on dating in a whole year.
Suspicion is already widespread, which is the encouraging part. Pew found that 52% of people who have used dating sites or apps have come across someone they believed was trying to scam them — 63% among men under 50, 47% among men 50 and over, and 44% among women. The instinct is usually right; what people lack is a way to act on it. If a profile photo, a phone number or an email address is the only thing you have, you can check whether that identity appears consistently elsewhere using our reverse phone check, an email consistency check, or the photo checker.
Age matters for severity rather than frequency. Adults aged 60 and over reported losing $2.4 billion to fraud of all kinds in 2024, up from about $600 million in 2020. Total consumer fraud losses reported to the FTC reached $12.5 billion in 2024, a 25% year-on-year rise. Our catfishing statistics page breaks the deception side of this down further.
Are Americans still getting married?
Yes — the U.S. recorded roughly 2,041,000 marriages in provisional 2023 data, a rate of 6.1 per 1,000 people, against a divorce rate of 2.4 per 1,000, per the CDC's National Center for Health Statistics. In 2022 there were 2,065,905 marriages at a rate of 6.2 per 1,000 — the first year above two million since 2019.
What do the national marriage and divorce rates look like?
| Year | Marriages | Marriage rate per 1,000 | Divorce rate per 1,000 |
|---|---|---|---|
| 2022 | 2,065,905 | 6.2 | 2.4 |
| 2023 (provisional) | ~2,041,000 | 6.1 | 2.4 |
Source: CDC/NCHS provisional national marriage and divorce statistics. Divorce coverage excludes several states.
A point of view worth stating plainly, because it appears on nearly every competing dating-statistics page: the claim that "half of all marriages end in divorce" is not what the current rate data shows. A divorce rate of 2.4 per 1,000 against a marriage rate of 6.1 per 1,000 is a ratio of roughly one divorce for every 2.5 marriages in the same year, and crude rates measured on different cohorts cannot be divided to produce a lifetime divorce probability anyway. Anyone who divides one by the other and publishes the answer is doing arithmetic on incompatible denominators.
Attitudes have shifted alongside the demography. Singles in America found 60% of singles say they believe in love at first sight, up 30 percentage points since 2014, and 26% now use AI in some part of their dating life — a 333% year-on-year jump. Later marriage and fewer married households have not produced a less romantic population. They have produced a more cautious, more mediated one.
Why do dating statistics disagree with each other?
Because "how couples meet" studies survey completely different populations — new couples, all partnered adults, or married couples only — and the same underlying reality produces answers from 10% to 60% depending on which one you pick. This is the single biggest source of confusion in dating data, and it is worth ten minutes of your attention if you intend to cite any of these numbers.
What are the three denominators, and why do they diverge?
- New couples (flow). Rosenfeld's HCMST asks currently-partnered people when and how they met, then looks at those who got together recently. This is what produces the 39% figure for 2017. It measures the leading edge of behaviour change.
- All partnered adults (stock). Pew asks every partnered adult how they met their current partner. Answer: 10% met through a dating site or app. This includes couples who met in 1978. It measures the accumulated past.
- Recent, young or LGB subsets. Pew's same survey finds 20% of partnered adults under 30 and 24% of partnered LGB adults met their partner online. Narrow the population, and the number moves toward the flow figure.
Both numbers are true: Pew's 10% and Stanford's 39% are both correct and differ by a factor of 3.9. Neither is wrong; they are measuring stock versus flow. A page that quotes one without naming the denominator is misleading its readers, and that describes most of the dating-statistics pages currently ranking for this query.
There is a second trap specific to the Stanford data, and we would rather lose a headline than repeat it. You will see "60% of couples now meet online" attributed to HCMST's 2022 wave. Be careful. The 2017 wave surveyed 2,997 adults; the 2020 wave (2,107 respondents) and the 2022 wave (1,722 respondents) are follow-ups of those same 2017 participants, documented on the ICPSR study record and the Stanford HCMST data page. A panel that re-interviews the same people cannot measure how couples formed in 2022 met, because the couples formed in 2022 are largely not in it. The last clean cross-sectional measurement of new-couple formation remains 2017, at 39%.
Three more disagreements worth knowing about:
- Scam losses are reported losses, not actual losses. FTC figures count what consumers chose to report. Romance fraud carries unusual shame, so the true total is higher by an unknown margin. Every FTC number on this page is a floor.
- Full-year versus part-year totals. The $1.16 billion figure covers January to September 2025. Comparing it directly with a full-year 2023 figure of $1.14 billion would be a units error. We annualised it explicitly and flagged it.
- Survey sponsorship. Singles in America is funded by Match Group, which sells dating subscriptions. Its cost figure ($213/month) and BMO's ($194/month, derived) agree within 10%, which is why we trust the range — but a single sponsored survey should never be the only source for a number.
What does this mean if you are dating right now?
The practical translation of everything above: the venue where you are most likely to meet someone is now also the venue with the least verification. Dating apps are imperfect, but they have safety teams, reporting flows and some identity friction. A direct message from a stranger on a social platform has none of that, and it is where 40% of money-losing romance scams begin.
- If someone contacted you first, unprompted, on social media, you are in the highest-risk category in the FTC's data — before anything else about them is even known.
- If they will not video call, treat that as data, not shyness. Photo, name and phone number should all point at the same consistent person.
- If money enters the conversation, the median outcome in FTC data is a $2,218 loss — roughly a year of what you would otherwise spend dating.
- If you already suspect something, you are probably right: 52% of dating-platform users have hit someone they believed was a scammer. Run the photo, phone or username through a consistency check before you invest more time.
- If the person is asking you to move to WhatsApp or Telegram immediately, that is the documented pattern in FTC reporting, not a coincidence.
The spine, restated now that the data has carried it: courtship kept migrating online while the moderated part of online dating shrank. The meeting is happening in the messages, not on the platforms — and the messages are unverified by design. That is not an argument for dating offline. It is an argument for doing five minutes of consistency-checking before you invest five months.
Frequently asked questions
What percentage of couples meet online in 2026?
There is no rigorous 2026 measurement. The last clean cross-sectional figure is 39% of new heterosexual couples in 2017, from Stanford's HCMST study published in PNAS, up from 2% in 1995. Figures above 50% circulating for 2022 come from a follow-up panel of the same 2017 respondents and cannot measure new couples. Pew's separate 10% figure counts all partnered adults, including couples formed decades ago.
How much does the average date cost in 2026?
The average all-in date cost $189 in 2026, up 12.5% from $168 in 2025, according to BMO's Real Financial Progress Index. Millennials spend the most at $252 per date, Gen Z $205. Annual dating spend averages $2,323, or roughly $194 a month, which lines up within 10% of the $213 monthly figure from Match and the Kinsey Institute.
How many Americans are single or unmarried?
Married-couple households fell to 47% of all U.S. households in 2025, from 66% in 1975, per the Census Bureau. There are 39.7 million one-person households, 29% of the total. Median age at first marriage reached 30.8 for men and 28.4 for women, both up 7.3 years since 1975.
Is ghosting really that common?
About 76% of active daters have ghosted someone or been ghosted, per a Forbes Health/OnePoll survey of 5,000 U.S. daters. Roughly 60% have been ghosted and 45% admit doing it, with men and women ghosting at near-identical rates (44% and 47%). Note that ghosting has no standard definition across surveys, so treat the figure as directional.
Are romance scams more common on dating apps or social media?
Social media, by a wide margin. FTC data shows 40% of people who lost money to a romance scam said contact began on social media, versus 19% on a dating site or app — 2.1 times the rate. Scammers typically move the conversation quickly to WhatsApp, Telegram or another private messaging app.
How much money is lost to romance scams?
U.S. consumers reported $1.16 billion in romance-scam losses across the first nine months of 2025 from 55,604 reports, up 22% on the same period in 2024. That works out to about $2,951 lost every minute and one report filed every seven minutes. Full-year 2023 losses were $1.14 billion with a median loss of $2,000; the Q3 2025 median was $2,218.
Do half of all marriages really end in divorce?
No, and the arithmetic behind that claim is broken. CDC data shows a marriage rate of 6.1 per 1,000 and a divorce rate of 2.4 per 1,000 in provisional 2023 figures. Dividing one crude rate by the other compares different cohorts in the same year and cannot produce a lifetime divorce probability.
Sources
- Rosenfeld, Thomas & Hausen, PNAS (2019) — Disintermediating your friends
- Pew Research Center — Key findings about online dating in the U.S.
- U.S. Census Bureau — Families and Living Arrangements, 2025
- CDC / National Center for Health Statistics — Marriage and Divorce
- FTC Data Spotlight — Romance scammers favorite lies exposed
- FTC — Consumer Sentinel Network Data Book 2024
- FTC 2025 romance-scam figures, as reported
- Match & The Kinsey Institute — Singles in America 2025
- Fortune — BMO 2026 Real Financial Progress Index on dating costs
- Forbes Health / OnePoll — Modern dating and mental health
- CNBC — Match Group Q4 2025 earnings
- ICPSR — HCMST 2017, 2020, 2022 study record
Related reading on this site: catfishing statistics and online dating statistics.
Think Someone Might Be Catfishing You?
Run a free check to verify their identity in seconds.
Free Catfish Check