Does Being Less Selective Get You More Matches? We Checked 2,006 Users

Every dating advice column says the same thing: stop liking everyone, be selective, quality over quantity. We had the data to check whether that advice survives contact with reality, so we checked. We grouped 2,006 SoulMatcher users who had made at least 50 swipe decisions by how large a share of the profiles they saw they said yes to, and followed each group through to matches and real conversations.

The short answer is uncomfortable for the advice columns: the people who like a larger share of profiles end up with more matches and more conversations per profile they see, not fewer. The more interesting answer is why. Each individual like is worth less than the one before, conversion falls the whole way down, and yet for most of the range the extra volume more than makes up for it. Only among the most generous men does the final yield stop rising.

How selective people actually are

Before the results, the baseline. Among users with 50 or more decisions, the median woman liked 6.4% of the profiles she saw. The median man liked 23.2%. That gap is consistent with what we found across all 417,785 swipes in our earlier study of like economics, and it is the reason men and women need separate tables below. The same like rate means something different depending on which side of the market you are on.

Men: volume buys matches, at a falling price

Men, by share of profiles liked	People	Avg. matches	Matches per like	Matches per 100 seen (95% CI)	Conversations per 100 seen (95% CI)
Under 10%	187	0.8	10.2%	0.42 [0.24, 0.64]	0.08 [0.04, 0.12]
10 to 20%	147	1.7	6.1%	0.88 [0.68, 1.10]	0.23 [0.16, 0.33]
20 to 40%	204	3.2	5.3%	1.51 [1.12, 2.03]	0.42 [0.27, 0.63]
40 to 60%	85	6.1	5.1%	2.46 [1.75, 3.19]	0.99 [0.55, 1.44]
Over 60%	103	5.4	3.3%	2.49 [1.56, 3.55]	0.97 [0.53, 1.48]

The last two columns are the fairest comparison, because they put every group on the same footing: what did each hundred profiles a man looked at actually produce? The bracketed ranges are 95% confidence intervals from a bootstrap over users, so you can see how firm each number is.

Read also: Why First Dates Feel Awkward Even When There's Real Chemistry

Read that way, liking more works, and it keeps working further than the advice would predict. Men who liked fewer than one profile in ten got 0.42 matches per 100 profiles seen; men who liked 40 to 60% got 2.46, and their confidence intervals do not overlap, so that rise is real rather than noise. Conversations per 100 views climbed the same way, from 0.08 to 0.99.

Where it stops is higher than the round number the advice would pick. The gains are still large going into the 40 to 60% group; only above 60% do they level off. Men who liked more than 60% of profiles got 2.49 matches per 100 views against 2.46 for the 40 to 60% group, a gap of 0.03 whose bootstrap confidence interval runs from about minus 1.2 to plus 1.3 per 100 and sits squarely on zero. So the curve appears to flatten somewhere in the high like-rate range; our five buckets cannot pin the exact point, and we do not claim one.

The efficiency column runs the other way throughout. A like from the most selective men converted into a match 10.2% of the time; from the most generous, 3.3%. Every step up in volume cut the value of each individual like. What kept the final yield climbing was that the extra likes more than covered that falling efficiency, until the very top, where they no longer did.

Read also: Can You Guess Your Own Psychological Profile? We Checked 7,013 People

One detail worth flagging: men in the 40 to 60% group also turned matches into conversations at the highest rate. I have a guess about why, and it stays a guess; it is in the caveats below.

Women: the same shape, with a higher floor

Women, by share of profiles liked	People	Avg. matches	Matches per like	Matches per 100 seen (95% CI)	Conversations per 100 seen (95% CI)
Under 10%	824	0.8	11.6%	0.45 [0.38, 0.53]	0.16 [0.13, 0.19]
10 to 20%	278	2.5	11.2%	1.57 [1.36, 1.81]	0.47 [0.39, 0.56]
20 to 40%	132	3.2	8.0%	2.04 [1.56, 2.54]	0.68 [0.52, 0.85]

Two thirds of women sit in the most selective group, which is the market structure we already knew. Within it, the pattern matches the men's: more likes, more matches per 100 views, lower conversion per like. The difference is the floor. A woman's like kept converting at over 11% up to a 20% like rate, and only then dropped to 8%.

Another 46 women had like rates above 40%, but that group fell below our minimum reporting size of 50 people, so it is not shown as its own row.

Read also: What Makes the Perfect Dating Profile? 70% of App Users Say This Trait Gives Them the Ick

Two forces pulling in opposite directions

It is worth being explicit about the arithmetic, because it is the real finding. Matches per 100 profiles seen is just the number of likes per 100 profiles multiplied by the share of those likes that come back. Those two terms move in opposite directions. As someone likes a larger share of profiles, each individual like converts less often: the efficiency column falls. But they are sending enough extra likes to more than offset that decline, so the final number of matches per 100 views still climbs. The flattening at the top is simply the point where the rising volume and the falling efficiency cancel out; past it, more likes no longer buy more matches. "Be more selective" is advice about the efficiency term alone. It is true about that term and wrong about the total.

What this does and does not show

This is the part of the article I care most about getting right, because the temptation to overclaim here is enormous.

It shows that, across the people who use SoulMatcher, higher like rates go together with more matches and conversations per profile seen, with the gains largest through the 40 to 60% group for men and levelling off only in the highest like-rate range; and that conversion per like falls throughout. Differences in the number of profiles viewed do not explain the pattern, because the per-100-view figures standardise for exposure volume. They do not, however, control for which profiles each user was shown, or for other differences between the groups: heavy and light likers may sit in different cities, age ranges or times of day, may be shown different people, and may be more or less sought-after themselves.

Read also: Why Am I Not Getting Matches on Dating Apps? 9 Reasons and What to Fix First

It does not show that any individual who starts liking more will get more matches. People who like generously may differ from people who like sparingly in ways we did not measure: how often they open the app, how they write, what their profile looks like, how quickly they reply. The fact that high-volume men also converted matches into conversations at the highest rate hints that engagement, not just volume, is part of the picture. Our data cannot separate the two.

It does not support the popular advice that being more selective is a route to more matches. In this sample the more selective groups had fewer matches per 100 profiles viewed, not more, even though each of their likes converted more efficiently. Selectivity buys efficiency per like, and perhaps better-fitting matches. In these numbers it does not buy more of them.

What I would take from it

If you like fewer than one profile in ten and see very little happening, the arithmetic helps explain why. In the most selective groups, a single conversation was the exception rather than the rule: 92.5% of the most selective men and 81.2% of the most selective women reached no two-way conversation at all in the window we measured. Seven likes at a 10% conversion rate is less than one expected match, and a match is still several steps short of a conversation.

Read also: Does Your Zodiac Sign Predict Who You Match With? We Checked 70,759 First Likes

If you like almost everyone, you are past the point where it helps. Above about 60%, men in our data got no more matches or conversations per profile seen than men in the 40 to 60% group, while each like was worth a third of a selective one. The matches that do come still face the attrition we described in the funnel: only about a third of matches become conversations, and the opening message still has to happen. Our collection of opening lines exists because the match is never the finish line.

The interesting zone is the middle. For women especially, the 10 to 20% group kept almost all of the per-like efficiency of the most selective group, 11.2% against 11.6%, while getting about three times the matches per 100 views; for men the trade-off is steeper, but the direction is the same. Whether that is a strategy or just a description of people who are naturally moderate, our data cannot say; it is, at minimum, where the numbers stop punishing you in either direction.

If you want a way to be selective that does not rely on the photo, the compatibility score exists so that a "yes" can carry more information than a swipe. Whether that changes these conversion rates is something we intend to measure rather than assume.

Methodology and definitions

Sample: 2,006 SoulMatcher users who specified male or female gender and had at least 50 recorded swipe decisions (1,280 women and 726 men). Six additional non-binary users met the activity threshold but were excluded from the gender-stratified analysis because the group was below our minimum reporting size of 50. Data aggregated on 31 August 2026. Groups with fewer than 50 people are not reported; the two highest male groups have 85 and 103 people and should be read with that in mind.

• Like rate: the share of a user's swipe decisions that were likes.
• Like: a unique positive decision about another person; repeated likes in the same direction were collapsed to the earliest.
• Match: a pair in which both people liked each other.
• Conversation: a match in which both people sent at least one human message; automatic system messages created on matching are excluded.
• Matches per like: total matches divided by total likes within a group, so heavy likers weigh more than light ones.
• Per 100 profiles seen: total matches (or conversations) in a group divided by total swipe decisions, times 100. This is the exposure-adjusted comparison.
• Confidence intervals: the bracketed ranges on the per-100 figures are 95% intervals from a bootstrap that resamples users within each group 3,000 times, so the uncertainty shown reflects differences between people rather than between individual swipes.

All figures are group-level aggregates; no individual profile or swipe history was manually reviewed. The analysis is descriptive: it compares groups of people who already behave differently and does not test what happens when one person changes their behaviour.

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