The Kinks Couples Agree on Fastest

The full mutual-yes ranking, five myths the data challenges, and the quietly devastating gap between what couples accept, what they propose, and what they ask AI to imagine.

Eleven thousand couples. Real ratings. Real disagreements. Real moments of "oh thank god, you want that too."

Here's what they told us.

How we got these numbers: Field dates Oct 2024-Feb 2026. Source: anonymised in-app activity ratings from ~11,000 self-selected couples on BeMoreKinky, our BDSM communication app. "Mutual yes" means both partners independently rated the complementary activity as yes. This is a self-selected community of people who downloaded a kink app together, not a random population sample, so take the numbers as a portrait of this community, not The Universal Truth About Kink. Full methodology →

The Full Mutual-Yes Ranking

When both partners independently rate the complementary side of an activity as "yes", one rates giving, the other rates receiving, we call that a mutual yes. The ranking below covers the top kink-specific activity pairs by mutual-yes rate, excluding sex positions.

Before you start mentally composing your "well actually, what about doggy style" comment, yes, sex positions technically rank higher than any kink-specific activity. Doggy style (93.4%), missionary (89.6%), and cowgirl (87.4%) all clear the top of this list. They're excluded here because they represent baseline sexual activity rather than kink-specific behaviour. We're interested in what happens once you venture past the standard menu.

Now, the pattern in what remains. I find it genuinely fascinating. The top four activities are all sensory or anticipation-based: blindfolding, building anticipation, surprise notes, and sensory reduction. Not a single flogger, paddle, or rope in sight. These are activities that modify the experience, they remove a sense, stretch out time, or introduce unpredictability. They don't require anyone to adopt a new identity or fumble through unfamiliar language while trying to maintain eye contact.

The naming activities tell a different story. They're popular, but increasingly divisive as you move down the list. "Good girl" naming has a 72.1% mutual-yes rate but also a 13.3% conflict rate (n=1,017). By the time you reach "naughty one," you're at 48.1% mutual yes with 18.5% conflict (n=1,118). That's roughly one in five couples disagreeing, one person thrilled by the word, the other quietly cringing into their pillow.

The pattern is worth sitting with: couples agree fastest on activities that change what happens to them, and slowest on activities that change what they're called. Modifying experience is easier to align on than modifying identity. Makes sense, doesn't it? You can blindfold someone without either of you having to answer the question "who am I in this scenario?" Call someone "naughty one" and suddenly you're both negotiating something much deeper than sensation.

5 Myths the Data Challenges

I want to be clear about what this data can and can't do. It doesn't prove these myths wrong in a clinical sense. We're looking at a self-selected app community, not a population sample. But it does produce numbers that sit uncomfortably next to assumptions I've seen repeated, unchallenged, for many years.

1. Myth: BDSM is primarily about pain.

The most agreed-on kink activities are blindfolding (88.2%), anticipation (84.5%), and praise. Cuddles achieve 95%+ acceptance; verbal praise sits at 94.5%+. No pain-related activity reaches the top of the mutual-yes ranking. The most popular kink activities in this community are, essentially: take away my sight, make me wait, and tell me I'm good. Not exactly the dungeon-crawler fantasy the tabloids love to trot out.

2. Myth: Couples in kink agree on everything.

Even the lowest-conflict category in the dataset, bondage, still carries a 6.3% disagreement rate. Roleplay hits 16.9%. If you and your partner disagree on a specific activity, you are not broken. You are not incompatible. You are having a profoundly normal experience. Agreement is the norm, but mismatch is always present. That's why we built the app in the first place, because having the conversation is the bit people skip, and it's the bit that matters most.

3. Myth: Kink gets more extreme with experience.

This one's for the people who assume experienced kinksters are all swinging from chandeliers in head-to-toe latex. The biggest experience-driven shifts in our data are in non-monogamy (+13.5 points from Level 1 to Level 4) and psychological play (+12.1 points), categories defined by relational complexity, not physical intensity. Submissive naming barely moves (+0.5 points). Experience appears to expand what couples are willing to explore relationally, not how far they push physically. Depth, not extremity. Who'd have thought.

4. Myth: Everyone in BDSM wants to be degraded.

Public humiliation and extreme degradation have acceptance rates below 3% across the entire dataset. Three percent. The community draws a sharp and consistent line between private play and public exposure. Degradation exists in the data, of course it does, but it is a niche preference, not a norm. For every person who wants to be called a filthy slut in the bedroom, there are roughly thirty-three who do not. Let's retire the assumption that kink and humiliation are synonymous, shall we?

5. Myth: "Good Girl" and "Good Boy" are equally popular.

Acceptance rates are nearly identical: 77% for "Good Girl" vs. 76% for "Good Boy." But "Good Girl" has approximately 27 times more ratings. The willingness is comparable. The demand is overwhelmingly gendered. Both genders apparently want to hear they're good, but one side of that equation is doing nearly all of the asking. The full analysis of this is honestly a bit of a rabbit hole, and it's coming in a dedicated piece on gendered kink language later this week.

The Fantasy-vs-Action Gap

Not everything couples agree on gets proposed. And not everything people fantasise about gets rated. The data reveals three distinct layers of desire: what you'd accept if asked, what you'd actually ask for, and what you imagine privately. Three layers. Three different versions of what you want. And they don't always match.

High acceptance, zero proposals

Some activities achieve near-universal acceptance but generate essentially no proposals between partners:

Read that again. Ninety-three percent acceptance. Zero proposals. These are phrases people desperately want to hear but apparently never explicitly request. They exist in a gap between desire and action, perhaps because requesting a specific phrase feels like handing your partner a script, or because people assume their partner already knows what to say. (They don't. Nobody does. That's rather the point.)

The data suggests an entire category of wants that are invisible to your partner without a structured prompt. Which is, incidentally, exactly what the BeMoreKinky app is designed to provide (I'm not being subtle about the plug here).

Most proposed activities

By contrast, the activities couples actually propose to each other tell a braver story:

The most-proposed activities mirror the most agreed-on activities. Blindfolding and anticipation dominate both lists. These are things couples feel safe asking for, perhaps because they carry low identity risk, or perhaps because their popularity is already well known enough to reduce the fear of rejection. Nobody's terrified of suggesting a blindfold. It's practically the kink equivalent of suggesting you order a takeaway.

AI scene preferences vs. proposals

The BeMoreKinky app includes an AI scene-writing tool. You describe what you're into, it writes a scene for you. The activities people ask AI to write about diverge noticeably from what they propose to their actual human partner, and honestly, the divergence is telling.

• Tickling is the number-one activity in AI scenes (124 scenes) but has only 48 proposals and 52% acceptance. People enjoy imagining it more than they feel confident requesting it. There's something about asking your partner to tickle you that apparently feels more vulnerable than asking them to tie you up. I don't make the rules.
• Bondage dominates AI scenes: "Tie wrists behind back" (87 scenes), "Wrist ties under mattress" (99 scenes), "Padded cuffs" (83 scenes). These are specific, physical, and detailed, the kind of scenario that benefits from a script. People want the choreography worked out before they attempt it live.
• Eye contact rules appear in 104 AI scenes, the second most-used AI activity, despite generating only 27 proposals between partners. There's a category of power exchange that people crave but find almost impossible to articulate out loud. "I want you to tell me when I can and can't look at you" is apparently much easier to type into an AI than to say across the dinner table.

The gap between these three layers (what people rate yes, what they propose, and what they ask AI to imagine) suggests that desire is not a single spectrum from "no" to "yes." It's at least three-dimensional: acceptance, initiative, and fantasy. A couple could have high alignment on acceptance, moderate alignment on initiative, and completely different fantasy landscapes, and still be perfectly well matched. The mismatch isn't a problem. The silence is.

More From This Research

• What 11,000 Couples Reveal About the Kinks Partners Actually Agree On - the overview of mutual-yes rates, conflict patterns, and experience effects
• Methodology and Privacy - how we collected, anonymised, and analysed this data