I built an app based on Gottman's research. Here's what I learned about relationships.

After months of reading John Gottman's research, I tried to turn 40 years of relationship science into software. The process changed how I think about love, technology, and the strange space between them.

A couple of years ago, I found myself doing something odd. I was reading academic papers about marriage at 1 AM. Not because my own relationship was in trouble, but because I kept stumbling on the same name in every article about what makes relationships actually work: John Gottman.

Gottman and his colleagues at the University of Washington have spent over four decades studying couples. They've observed thousands of them in what the media nicknamed the "Love Lab" -- a research apartment wired with cameras and biometric sensors where couples just... live for a weekend. The researchers watch how they talk, fight, eat breakfast, and make small requests of each other. From those observations, they can predict whether a couple will divorce with roughly 90% accuracy.

Gottman, J. M., & Silver, N. (1999). The Seven Principles for Making Marriage Work. Harmony Books.

That number floored me. Not because prediction is inherently impressive, but because of what the predictions are based on. It's not the big things. It's not how often you fight or how compatible your life goals are. It's the small things. The tiny, forgettable moments that happen twenty times a day.

The surprising thing about what matters

Here's what I didn't expect to find in the research: the strongest predictor of relationship success isn't communication style during conflict, although that matters. It's what Gottman calls "bids for connection" -- those small moments when one person reaches out for attention, affection, or acknowledgment, and the other person either responds or doesn't.

"Hey, look at this bird outside." That's a bid. "I had a rough day." That's a bid. Even a sigh from the next room can be a bid. In the couples who stayed happily together, partners responded to these bids 86% of the time. In couples who eventually divorced, that number was 33%.

Gottman, J. M., & DeClaire, J. (2001). The Relationship Cure. Harmony Books.

What struck me wasn't the statistic itself. It was the implication: relationships don't fail in dramatic explosions. They erode, slowly, through a thousand missed small moments. The person who stops saying "how was your day?" and meaning it. The text that goes unreturned not once, but as a pattern. The gradual withdrawal of attention that nobody notices until the damage is done.

Why I chose 5 relationships, not 500

When I started thinking about building something, my first instinct was wrong. I thought about social networks, about connecting people at scale. But Gottman's work kept pointing me in the opposite direction. And then I found Robin Dunbar's research.

Dunbar, an evolutionary psychologist at Oxford, has shown that humans can maintain roughly 150 meaningful social connections. But within that number is a much more important structure: about 5 people in your innermost circle, 15 in the next layer, 50 after that. Your closest 5 relationships account for a disproportionate share of your emotional wellbeing.

Dunbar, R. I. M. (2010). How Many Friends Does One Person Need? Harvard University Press.

The problem is, those 5 people are the ones we're most likely to neglect. We assume they'll be there. We invest our relational energy in newer, more uncertain connections -- a new coworker, a potential friend, a social media audience -- while the people who matter most get the leftovers of our attention.

So I made a decision that felt counterintuitive for a tech product: Tend would be about your closest relationships only. Five to ten people. Not your network. Not your followers. The humans who actually shape your daily experience of being alive.

Translating research into software

The hardest part of building Tend wasn't technical. It was philosophical. Relationships are deeply human. They resist optimization. They aren't problems to be solved with push notifications.

I spent a lot of time thinking about what an app can and can't do in this space. Here's where I landed:

What an app can do: It can remind you. Not in a nagging way, but as a gentle pattern interrupt. When you haven't reached out to your brother in three weeks, a quiet nudge can be the difference between a relationship that stays warm and one that slowly fades. It can surface the research -- give you a framework for understanding why something felt off in a conversation. It can help you notice patterns you'd otherwise miss.

What an app can't do: It can't replace vulnerability. It can't have the hard conversation for you. It can't make you care about someone you don't care about. And it absolutely shouldn't try to gamify love. I've seen relationship apps that give you "points" for being a good partner. That misses the entire point. The moment you're doing something for points instead of for the person, you've lost the thread.

What I actually learned

Building Tend changed me, which I wasn't expecting. Reading Gottman's work carefully, not skimming it for product insights but actually sitting with it, shifted something in how I show up in my own relationships. A few things I keep coming back to:

Small things often. This is Gottman's mantra, and it sounds like a greeting card until you internalize it. Grand gestures are wonderful, but they're a terrible relationship strategy. The couples who thrive aren't the ones who plan elaborate anniversary trips. They're the ones who consistently turn toward each other in the mundane moments. Consistency beats intensity, every time.

Maintenance isn't a dirty word. We maintain our cars, our homes, our health. But we treat relationship maintenance as a sign that something is wrong. It's not. It's a sign that you understand how relationships actually work. They need tending -- attention, warmth, repair after the inevitable small hurts.

Attention is the fundamental unit of love. Not money. Not time, exactly. Attention. The quality of your presence when you're with someone. Whether you look up from your phone. Whether you remember the small thing they mentioned last Tuesday. Attention is the currency, and most of us are spending it in the wrong places.

An honest assessment

I don't think Tend will save anyone's marriage. I don't think any app can. What I do think is that there's a gap between knowing what the research says and actually doing it. We've all read that we should "be more present" or "communicate better." The gap isn't knowledge. It's practice.

That's the modest thing I'm trying to build: a tool that helps you practice. That gently surfaces the right insight at the right moment. That helps you notice the bid you almost missed. Not a replacement for the hard work of loving someone well, but a small companion for the journey.

If Gottman taught me one thing, it's that small things done consistently are the whole game. I'm trying to build something that reflects that truth. Whether I've succeeded, I'll let you decide.