The quick comparison: current use peaks among adults under 30
SSRS found that 10% of adults ages 18–29 currently used online dating in January 2026. The figure was 8% for adults ages 30–49 and 2% for adults age 50 or older. Each percentage uses all U.S. adults in that age group, not only single adults, app users, or people seeking a relationship.
The two-point difference between the younger groups may be within survey uncertainty. The full survey's margin of error was plus or minus 2.5 percentage points, and age groups have more uncertainty. The clear result is that current use was higher below age 50 than above it.
Past use tells a different age story
Ever use was 51% for ages 18–29 and 53% for ages 30–49. This does not mean that adults in their thirties and forties are more active today. They have had more years to try a service, and lifetime use still counts them after they stop.
For adults age 50 or older, ever use was 20%. Current use was 2%. Some former users may now have a partner. Others may have paused or left apps. The gap does not show why they left, how long they stayed, or whether the experience succeeded.
Pew’s 2022 age bands provide a useful older baseline
Pew found ever-use among 53% of adults ages 18–29, 37% of ages 30–49, 20% of ages 50–64, and 13% of adults 65 and older in July 2022. Its finer split shows that adoption was lower among the oldest adults, but the 2022 values should not be combined mechanically with the 2026 SSRS series.
The surveys used different samples, dates, and questionnaires. SSRS grouped everyone 50 and older together; Pew separated ages 50–64 and 65+. A change from 37% to 53% for ages 30–49 is notable, but it is not a clean trend estimate unless the questions and methods match and survey uncertainty is tested.
Age reflects life stage and years of experience
People who are 45 in 2026 came of age before swipe apps but have had years to encounter online dating. People who are 22 entered adulthood with apps already normalized but have had fewer years to accumulate an “ever used” response. Older adults may have first partnered before online dating existed and may return after divorce or bereavement.
These factors can pull the rates in different directions. A one-time survey measures people of different ages at one moment; it does not follow one age group from 20 to 60. Saying use “falls as people age” can confuse age with generation, partnership changes, and past availability.
Tinder, Bumble, and Hinge are more common among younger users
Among 18–29-year-olds who had used any online dating service, SSRS found that 74% had used Tinder, 49% Bumble, and 44% Hinge. These are the three largest platform estimates published for the group. Because people surveyed could have used several services, the percentages overlap and cannot be added.
The counted group is especially important. Seventy-four percent does not mean nearly three-quarters of all U.S. adults ages 18–29 use Tinder now. It means nearly three-quarters of the group with any online-dating history had tried Tinder at some point. This separates a platform reach statistic from the share of the whole population.
Adults ages 30–49 have a more mixed platform history
Among people ages 30–49 who had used online dating, 46% had used Tinder, 38% Plenty of Fish, and 33% Facebook Dating. Bumble and Hinge remain relevant but did not occupy the same top-three order published for the youngest group. A single national “best app” list erases this age-specific history.
Historical reach still does not prove current local depth. A service can have many former users in an age group but fewer active profiles today. It can also be strong nationally and sparse in one city. Platform choice should begin with age and intention fit, then be validated by the visible local pool before paying.
Older users report a different platform mix
Among people ages 50 and older who had used online dating, Plenty of Fish was the most commonly tried measured service at 36%, followed by Match at 33% and Facebook Dating at 28%. Tinder did not lead this age group. That is a better clue for product research than projecting the under-30 ranking onto every adult.
It is not proof that any one service is best for people over 50. The survey does not report city-level active profiles, profile quality, response rates, relationship results by app, or the cost required to communicate. Those practical checks belong beside the group data, not beneath a ranking unsupported by it.
Current use is not the same as a company's active-account count
A person surveyed may interpret “currently use” as having an account, opening an app occasionally, or actively seeking dates. A company may instead count monthly active users, daily active users, logged-in accounts, or paid subscribers. Those measures can differ even when each is labeled correctly.
A person can also use multiple apps, so adding platform audiences counts some people more than once. Survey-based current use is useful because it represents the national adult population, including nonusers. It is not a replacement for a company's checked account data when the question is platform operations or market share.
National age rates cannot size your local pool
Multiplying a city’s age population by a national current-use rate can create a rough scenario, but it is not a verified app-user count. Local adoption may differ, and the population base includes partnered people, incompatible orientations, people outside a chosen radius, users on other services, and accounts hidden by each side’s preferences.
Additional filters add more uncertainty. If each filter comes from a different data set, multiplying estimates assumes links between traits that may not hold. Our dating-pool calculator uses person-level public data where possible and labels modeled variables, but app participation remains a separate unknown without platform access.
Small age differences need uncertainty, not rankings
The SSRS study surveyed 2,012 adults and reported a full-sample margin of error of plus or minus 2.5 percentage points. An estimate for one age group is based on fewer people and has a wider range of uncertainty. Therefore, 51% and 53% should be described as similar estimates, not a decisive age-group victory.
Large contrasts are more secure. The distance between roughly half of adults under 50 having ever used online dating and one-fifth of adults 50 and older is unlikely to be explained by rounding alone. Good reporting distinguishes robust patterns from leaderboard-style differences too small for the design to resolve.
Use one survey for a chart and another for context
A clean 2026 age chart should use SSRS for every bar so the question, field period, and age bands match. Pew can sit beside it as an earlier benchmark and supply its 50–64 and 65+ detail. It should not contribute one bar to a chart where another organization supplies the rest.
The same rule applies to platform data. Compare platform shares within the SSRS table of people who had used online dating, not against company downloads, web traffic, or subscriber counts. Consistent measures make a smaller chart more useful than a large mix of numbers that do not match.
Age data should change how someone evaluates a paid app
The platform-by-age results can guide a free trial order, but they do not justify an immediate subscription. Start with the services most commonly tried by the relevant age group, create a complete profile, and inspect several sessions for local relevance, freshness, and repetition. A national age fit is only a reason to look, not proof that the local pool is deep.
Payment is easier to evaluate after the constraint is visible. If enough relevant profiles appear but a useful communication, filtering, or review feature is locked, compare the total charge and renewal terms. If the free pool is sparse or mostly irrelevant, a visibility boost cannot manufacture compatible local supply.
What age data can—and cannot—tell an individual
The data can show that online dating is mainstream among adults under 50, that current use is much smaller than lifetime adoption, and that platform histories differ by age. It can help a reader select services to inspect and avoid mistaking a young-audience statistic for a population-wide fact.
It cannot calculate a personal probability of matching, receiving a reply, going on a date, or finding a partner. Those outcomes depend on location, compatible supply, presentation, preferences, behavior, safety, and chance. Age is one segmentation variable, not a destiny score.
What full group does each percentage or rate use?
| Population age group | Used at some point, SSRS 2026 | Use now, SSRS 2026 | Used at some point, Pew 2022 |
|---|---|---|---|
| 18–29 | 51% | 10% | 53% |
| 30–49 | 53% | 8% | 37% |
| 50 and older | 20% | 2% | 20% ages 50–64; 13% ages 65+ |
What does “denominator” mean? The denominator is the whole group used to calculate a percentage or rate.
How to read it: The surveys used different dates, groups of people, and age ranges. Pew's two older groups cannot be combined directly into the SSRS 50-and-older estimate without adjusting for each group's size.
Download this table (CSV) ↓