The Economy of a Like: What 417,785 SoulMatcher Swipes Reveal About Matches and Conversations

We analysed 417,785 swipe decisions on SoulMatcher. Women liked about 1 in 11 profiles, men about 1 in 3. Across the full dataset, one reciprocal pair formed for roughly every 174 swipes, one active match remained for every 206, and one two-way conversation emerged for roughly every 622.

That is the whole article in one paragraph. The rest is where those numbers come from, what they hide, and the one result that surprised me most: a precise answer to the question "my like got no response yet; what are my odds now?"

The funnel, end to end

Stage	Count
Swipe decisions	417,785
Likes (unique person-to-person)	73,782
Reciprocal like pairs (both said yes)	2,396
Still matched at measurement date	2,029
Pairs with at least one human message	1,095
Pairs with a two-way conversation	672

Two details in this table deserve a pause.

First, the gap between 2,396 and 2,029: 15.3% of reciprocal pairs were no longer active matches at the measurement date. Even mutual matches do not all remain active.

Second, the funnel keeps leaking even after mutual interest. Of the 2,029 reciprocal pairs that were still matched at measurement, only 672, about one in three, had reached a two-way conversation. Getting a mutual yes is only one stage. Turning it into an actual exchange is another.

Read also: Why So Many Conversations Die on Dating Apps: Understanding Modern Dating Patterns

Women say yes to 1 profile in 11. Men say yes to 1 in 3

Who is swiping	Swipes	Likes given	Like rate
Women	244,826	22,918	9.4%
Men	168,828	51,258	30.4%

Women like fewer than one profile in ten; men like almost one in three. Researchers have observed a similar gender gap in liking behaviour on Tinder, suggesting that the pattern is not unique to SoulMatcher (Tyson et al., 2016).

Since a match needs a "yes" from both sides, the scarcer "yes" does most of the deciding. That shows up directly in the next result.

Women's likes were more than twice as likely to become mutual

Two ways to measure it, and they agree.

Across all unique likes, 10.5% of women's likes belonged to pairs that eventually became mutual, compared with 4.7% of men's likes. But that figure includes likes that were the second half of a pair, answering a like already received. So we also isolated pure initiations: first likes, sent when the other person had not yet liked back.

Read also: Being Chosen vs Being Settled For: How to Tell the Difference

Initiator	First likes sent	Reciprocated within 30 days	Conversion
Women	20,858	921	4.4%
Men	49,310	1,141	2.3%

When a woman opens, one first like in 23 becomes a match within a month. When a man opens, one in 43. The two-to-one ratio survives the stricter test; only the absolute numbers shrink, which tells you how much of everyone's reciprocity actually comes from answering rather than initiating.

Your odds, day by day: the survival table

This is the number I most wanted to publish, because the intuitive versions of it that circulate in dating advice are wrong, ours included until we computed it properly.

We took every first like with a full observation window and asked: what is the chance it gets a reciprocal like within 30 days, given how long it has already waited?

Situation	Chance of becoming mutual within 30 days
Just sent	3.0%
Still unanswered after 24 hours	1.2%
Still unanswered after 72 hours	0.8%
Still unanswered after 7 days	0.5%

One silent day cuts the odds by more than half. A silent week cuts them six-fold. The like is not dead, but it is not resting either; it is expiring.

Read also: Choice Overload in Modern Dating: How Too Many Matches Can Harm Decision Making

For the minority of likes that do get reciprocated, timing looks like this: the median wait for the answering like was 18.7 hours, 53% arrived within a day, and 27% took more than a week. Slow reciprocity exists, it is just rare relative to everything still pending.

Attention is concentrated, and exposure explains only part of it

Who receives the likes? We ranked all 9,649 profiles that were ever swiped on by the number of likes they received.

Group	Share of all likes received	Share of all swipe views received
Top 10% most-liked profiles	54.5%	37.1%
Top 20% most-liked profiles	73.0%	53.2%

Part of the concentration comes from exposure: the most-liked profiles were also shown more often. But exposure alone does not explain the gap; the top 10% received 54.5% of likes while accounting for 37.1% of swipe views.

To separate the two further we computed each profile's like rate, likes divided by times shown, for the 2,421 profiles shown at least 50 times. The top decile by like rate gets liked 45.6% of the time it is seen. The average profile: 17%. The bottom decile: 0.7%. Being shown is not the same as being chosen.

Read also: Why Deep Conversations Matter in a Relationship and Why So Many Couples Struggle to Have Them

Why some profiles convert views into likes so much better is a question our data does not answer; audience composition, gender, age, geography, activity period, ranking and profile quality are all tangled together in it. We measured the concentration; we are not claiming a single cause.

What follows from these numbers

Do not treat a pending like as having constant odds. In our data, the probability of reciprocity within 30 days fell sharply with every period of silence: from 3.0% at send to 1.2% after one day and 0.5% after a week. Whatever checking the app hourly does to your nerves, the informative signal arrives mostly in the first day.

Women's likes were associated with the highest conversion in the dataset, on both measures: pair membership (10.5% against 4.7%) and pure initiation (4.4% against 2.3%). Our data does not say why any individual should or should not like first; In our sample, women's first likes were reciprocated about twice as often as men's.

Matches are not the scarce resource; conversations are. A third of active matches reach a real exchange, and that is after two rounds of mutual selection. Some context on why conversations die on dating apps is in our earlier piece, and if the empty message box after a match is what stops you, our opening lines that go somewhere were written for that exact moment of doubt.

Read also: How SoulMatcher’s “Matcher Marketplace” Could Revolutionize Matchmaking

Expect the market to be slower than the interface. Half of reciprocal likes take longer than 18 hours; a quarter take more than a week. The app feels instant, people are not.

This funnel also explains the product hypothesis behind SoulMatcher's compatibility score: giving people more information before they decide whether to like someone may help make that decision more deliberate. Whether it actually reduces attrition is a separate question for us to test.

Methodology and definitions

Sample: 417,785 swipe decisions recorded on SoulMatcher through 24 August 2026. Results are reported only as aggregated statistics; no individual profile or swipe history was manually reviewed.

Definitions used throughout:

• Swipe decision: one recorded like-or-pass action on a shown profile.
• Like: a unique positive decision from one person about another. Repeated likes in the same direction (977 raw rows) were collapsed to the earliest one, leaving 73,782 unique likes.
• First like (initiation): a like sent while no reciprocal like existed yet.
• Reciprocal like pair: two people who each liked the other, whenever that happened; 2,396 pairs.
• Match: a reciprocal pair that still existed as a match at the measurement date; 2,029. A pair can stop being an active match for several reasons, including one person unmatching, account deletion or moderation, and our data does not distinguish between them.
• Two-way conversation: a match where both people sent at least one human message. Automatic system messages that the app creates on matching were excluded; without that exclusion, messaging statistics inflate drastically.

Survival figures use a fixed 30-day observation window. For each point, we calculated the share of first likes still unreciprocated after X hours that received a reciprocal like within 30 days of the original like. Only likes sent at least 30 days before measurement are included, so every like has a complete window.

Gender splits cover users who specified gender; the non-binary group (4,131 swipes) is included in totals but is below our minimum group size for separate rates. Percentages are rounded.

This article is informational and reports aggregate, de-identified patterns of user behaviour. It is not advice about any individual person or relationship.