How couples meet now: what the data says about dating apps

DATINGLast updated: · 7 min read · 1 tag

Online is now the most common way couples meet in the U.S., yet dating apps are losing paying users. Here is what the best available data does and does not say.

🏷️ Page tags (1):
📌 Key takeaways
  • In a Stanford-led U.S. survey, about 39% of heterosexual couples who met in 2017 met online, up from about 22% in 2009.
  • In a 2022 Pew survey, 30% of U.S. adults had ever used a dating site or app, and one in ten partnered adults met their partner that way.
  • Match Group, which owns Tinder and Hinge, reported about 5% fewer paying users in 2025 than in 2024.

The short answer

Online is now the most common way for couples in the United States to meet. In the best nationally representative survey on the question, about 39% of heterosexual couples who met in 2017 did so online. That is one country and one survey, and it is several years old. Since then, the paying audience of the biggest dating-app company has been shrinking.

So both things are true: meeting online is normal, and many people are tired of the apps. Here is what the data actually shows, and where it stops.

How couples met, according to a Stanford-led survey

The most cited study on this is "Disintermediating your friends" by Michael Rosenfeld, Reuben Thomas and Sonia Hausen, published in PNAS in 2019. It uses the "How Couples Meet and Stay Together" survey, fielded in 2017 among a nationally representative sample of American adults by the survey firm GfK.

The key finding: for heterosexual couples, meeting online overtook meeting through friends around 2013. In the authors' words, meeting online became the most popular way couples meet in the U.S.

Meeting through family, church, the neighbourhood and school have all declined sharply since 1940, according to the paper.

  • About 39% of heterosexual couples who met in 2017 met online.
  • For couples who met in 2009, the share was about 22%.
  • About 65% of same-sex couples who met in 2017 met online.

What changed: phones, and fewer introductions by friends

The authors tie the rise to two technology shifts: the graphical web around 1995, and smartphones after 2007. They report that phone apps accounted for at least half of the growth in meeting online between 2010 and 2017.

The second finding is quieter but just as interesting. Online meeting did not simply add to friend introductions; it displaced them. The paper says growth in meeting online "strongly displaced" meeting through friends, and that among couples who met online, the share introduced through third persons has declined.

One detail worth knowing: the apparent rise in couples meeting at bars and restaurants after 2010 is, according to the authors, due entirely to couples who met online and then had their first in-person meeting at such a venue. Without them, bars and restaurants were a declining way to meet after 1995.

How many people use dating apps?

A Pew Research Center survey of 6,034 U.S. adults, fielded July 5–17, 2022 and published in February 2023, gives a snapshot.

Those two sources measure different things. The Stanford study asks how couples met, by the year they met. Pew asks all partnered adults, including people who met decades ago, which is why its 10% looks so much lower than 39%. They are not in conflict.

  • Three in ten U.S. adults (30%) have ever used a dating site or app, the same share as in 2019.
  • About half of adults under 30 (53%) have; among adults 50 to 64 it is 20%, and among those 65 and older, 13%.
  • One in ten partnered adults met their current partner through a dating site or app (married, living with a partner or in a committed relationship).
  • 53% of users say their experiences have been very or somewhat positive, and 46% say negative.

Is the app business shrinking?

Match Group owns Tinder and Hinge, among other brands. In its results release of 3 February 2026, it reported an average of about 14.2 million "payers" for the full year 2025, down 5% from about 14.9 million in 2024. The company defines a payer as a unique user at a brand level in a given month from whom it earned revenue, averaged over the period.

Read that carefully. It counts paying users, not everyone who uses a free tier. Because the company runs several brands and cannot identify people across them, the same person can be counted twice. And it covers one company, worldwide. It tells you about Match Group's paying audience, not about how many people date online overall.

The same release says technology should help people connect "and then step away, into the real world." That is a company statement of intent, not evidence of a trend, but it shows where the industry's own language is heading.

What can go wrong

Pew found that Americans are split on whether online dating is a safe way to meet people: 48% of adults called it very or somewhat safe in 2022, down from 53% in 2019. Among users, 38% had received unsolicited sexually explicit messages or images, 30% had experienced continued contact after saying they were not interested, and 52% said they had come across someone they thought was trying to scam them.

The U.S. Federal Trade Commission's advice on romance scams is blunt: never send money or gifts to a sweetheart you have not met in person, and be wary of someone who pushes you to move off the platform or says they cannot meet. It also suggests a reverse image search of the profile picture and talking to someone you trust. For practical steps on a first meeting, read our safe meeting guide in the Dating section of this blog.

What the data cannot tell you

Who In, a site for finding people nearby for walks, coffee and conversation, is in preparation and not yet launched. We mention it only because the open question in this data, how people meet in person without swiping, is the one it is built around.

  • It is mostly American. The Stanford and Pew figures describe U.S. adults. We found no equivalent, comparable figure for the world as a whole, so we are not giving one.
  • It is not current. The "how we met" data is for couples who met up to 2017; Pew fielded in 2022. Habits since then may differ.
  • It counts couples, not friendships. Nobody tracks how people make friends or find a walking partner with the same rigor, so none of these numbers directly apply to meeting people for coffee or sport.

Data at a glance

Each figure with its source, year, place and sample.

  • Heterosexual couples who met online: about 39% — Rosenfeld, Thomas & Hausen (PNAS); couples who met in 2017, U.S., HCMST 2017.
  • Same-sex couples who met online: about 65% — same study; couples who met in 2017, U.S.
  • Heterosexual couples who met online, earlier: about 22% — same study; couples who met in 2009, U.S.
  • Adults who ever used a dating site or app: 30% — Pew Research Center; 6,034 U.S. adults, July 2022.
  • Partnered adults who met their partner on a dating site or app: about 10% — Pew Research Center; same survey.
  • Match Group average payers: about 14.2 million, down 5% — Match Group results release; full year 2025, worldwide, one company.
COMMON QUESTIONS · FAQ

Frequently Asked Questions

What is the most common way couples meet?

In the 2017 U.S. survey analysed by Rosenfeld and colleagues, meeting online was the most common way for heterosexual couples, at about 39% of couples who met that year. This is U.S. data only.

Do dating apps lead to lasting relationships?

The Stanford authors report that earlier follow-up research found neither breakup rates nor relationship quality were influenced by how a couple met. Pew found that one in ten partnered U.S. adults met their partner on a dating site or app.

Are dating apps declining?

One company's paying-user count fell 5% in 2025 (Match Group, worldwide). We did not find a source that measures overall dating-app use across countries, so we cannot say whether use as a whole is falling.

Is it safe to meet someone from an app?

There is no guarantee either way. Pew found Americans split on it, and the FTC advises never sending money to someone you have not met. Read our safe meeting guide before a first meeting.

Sources and review method

Opened and read the Pew Research Center short read (2 Feb 2023), the Stanford-hosted author manuscript of Rosenfeld, Thomas & Hausen (PNAS 2019), Match Group's full-year 2025 and 2024 results releases and the FTC romance-scam page; every figure was checked against the opened text. Sources that could not be opened (PNAS site, PMC, Ofcom) were not used.

Last checked: November 4, 2026. If you spot a mistake, report a correction.

DISCOVER YOUR AREA

Browse profiles of people who share your interests

Pick your area and explore profiles by interests such as 💻 Working from a cafe or library or ☕ Coffee & chat. A linked social account shows that the account was checked; it does not verify identity or that someone lives in the chosen area.

Feedback · hello@whoin.app

We use the records needed for sign-in, security and the features you choose. Ads load only on blog pages, and only when you allow it. They are not shown on city, district or account pages.

Google’s consent screen could not load right now. Ads stay off; you can try again later.

Cookie policy · Privacy