Our Research
Real data from real relationships.
No assumptions.
We study what couples actually do. Not what they say they do, not what therapists prescribe. The observable patterns that predict connection, drift, repair, and growth.
How we work
Observe
We collect behavioral data from real couples using our consumer platform. Every data point is consent-tagged and anonymized at the point of collection.
Classify
Raw signals are categorized into five data classes: behavioral, linguistic, self-report, relational, and outcome. Each class has its own handling rules and consent requirements.
Analyze
We look for patterns, not prescriptions. Our analysis identifies what couples actually do differently, not what experts think they should do.
Publish
Findings are published with full transparency about sample size, methodology, confidence intervals, and limitations. No cherry-picking. No overstating.
Areas of study
Our research is structured around the DPR MAP framework. Six measurable qualities of any relationship, plus four named conditions that predict the slow erosion of a bond.
The six corners we study
Three corners describe how a bond is built. Three describe how it is kept. Together they define the shape of a relationship at any given moment.
Desire
The force that pulls two people toward each other. Not only physical. Desire is wanting to be chosen, wanting to be seen, wanting to matter to someone specific.
Key signals: initiation patterns, anticipation behaviors, vulnerability frequency, curiosity about partner's inner life
Presence
The quality of attention you bring to the person next to you. Presence is measurable. It shows up in eye contact duration, device-free time, and the depth of daily conversation.
Key signals: device-free interaction time, response latency, active listening markers, shared silence comfort
Rhythm
The daily patterns that form the skeleton of a relationship. Morning routines, evening rituals, weekly check-ins. Rhythm is what keeps the connection alive between the big moments.
Key signals: routine consistency, ritual maintenance, shared activity frequency, synchrony in daily patterns
Maintenance
The daily care between partners. The unglamorous, repeated acts. Dishes done without a scoreboard. A text in the middle of the day. Love expressed as upkeep, not applause.
Key signals: unsolicited care acts, mental load distribution, balance of small tasks, gratitude expression frequency
Awareness
The attention partners bring to each other's inner world. Awareness catches the shift before it becomes the fight. It is the early-warning layer of a relationship.
Key signals: accurate read of partner's mood, knowledge of current stressors, anticipation of needs, repair-attempt detection
Protection
The bond's defense system. From inside, repairing rupture before it sets. From outside, noticing when work, family, or the world is taking too much, and doing something about it.
Key signals: rupture-repair latency, boundary-setting frequency, defense of shared time, external pressure response
The four conditions we research
Four named states predict the slow erosion of a bond. Each one shows up before couples have words for what is happening.
Drift
The gradual, often imperceptible distancing between partners. Drift doesn't announce itself. It accumulates in the silences, the routines, the moments you stop noticing each other.
Key signals: decreased check-in frequency, parallel screen time increase, conversation depth reduction
Quiet
Emotional suppression between partners. Things felt but not said. The unsaid that builds pressure underneath ordinary days. Quiet is what drift sounds like before either partner has words for it.
Key signals: reduction in self-disclosure, monosyllabic responses, withholding of stress events, lower emotion-word density
Later
Chronic postponement of the conversations a relationship needs. The "we will talk about it later" that becomes never. Later is not avoidance. It is the slow forfeit of repair windows.
Key signals: unresolved-topic recurrence, postponement language frequency, repair-attempt rejection, postponement-to-confrontation lag
Signal Loss
Disconnection normalization. The day a partner stops noticing something is missing. Signal Loss is the most expensive condition because by the time it shows up, the cost is already paid.
Key signals: flattened satisfaction self-reports, lowered partner curiosity, normalized distance, baseline shift in expectations
What we measure
Every data point we collect falls into one of four categories, each with its own consent requirements and handling protocols.
Behavioral Signals
- Completion rates and consistency
- Timing patterns and engagement windows
- Repair velocity after conflict
- Streak maintenance and drop-off points
Linguistic Signals
- Sentiment shifts over time
- We-language vs. I-language ratios
- Blame marker frequency
- Emotional valence in self-reports
Self-Report Signals
- Assessment completions and scores
- Attachment style distributions
- Drift awareness scores
- Relationship satisfaction tracking
Relational Signals
- Partner synchrony patterns
- Asymmetry in engagement
- Repair-to-rupture ratios
- Mutual check-in frequency