Dating App Usage Data: Methodology and Free Download
Methodology, data dictionary and free CC BY 4.0 download for the PartnerCheck Dating Trust Index — aggregated, anonymized relationship search trend data.
What this dataset is
The Dating Trust Index is a set of aggregate counts describing how people find and use PartnerCheck, a photo-based check of public environmental signals. It covers monthly visit volume, visits by country, device class, hour of day (UTC), referrer class, and the most-visited public content pages, plus the total number of completed photo checks. It exists so that journalists, researchers and students writing about online trust, dating behavior or the shift to AI-mediated search can cite original numbers instead of recycled ones.
What this dataset is not
It is not a record of individuals, and it cannot be turned into one. It contains no rows — only totals. It contains nothing about the people our users search for, nothing about search outcomes, and no scan content of any kind. It is also not a representative survey: it measures the visitors of one website, and those visitors are self-selected. Read the headline findings with that limitation in mind.
Data dictionary
| Field | Type | Definition |
|---|---|---|
| byMonth | month → count | All recorded visits per calendar month (UTC), including automated traffic. |
| byCountry | country → count | Human visits by coarse country, resolved from IP at request time. Cells under 50 suppressed. |
| byDevice | class → count | Human visits by device class: mobile, desktop, tablet or unknown. |
| byHourUtc | hour → count | Human visits by hour of day, always UTC. Convert to a local timezone before drawing conclusions. |
| bySource | class → count | Human visits by referrer class: Google, ChatGPT, Perplexity, Bing, TikTok, Direct and Other. |
| topContent | path → count | Most-visited public content pages. Internal app routes (checkout, upload, results) are excluded by design. |
| totals.photoChecks | integer | Total completed photo checks in the period. Count only — no content, no outcomes. |
Privacy safeguards
The dataset is produced by a server-side aggregation function with hard rules, not by exporting a table. The rules are:
- Aggregates only. The function returns counts. Raw analytics events are never readable through any public endpoint.
- Minimum cell size of 50. Any group with fewer than 50 visits — a small country, an unusual hour, a rare source — is dropped entirely, so no cell can describe a handful of people.
- No identifiers exist in the pipeline output. IP addresses are stored only as one-way hashes for abuse prevention and are never read by this aggregation. Session IDs, user IDs, emails, user agents, cities and regions are never selected.
- No scan content. Uploaded photos are deleted automatically within 36 hours and are never part of any statistic. The dataset knows that a check happened, never what it found.
- Bot filtering. Automated traffic is flagged and excluded from every human-behavior breakdown (country, device, hour, source, content).
Live snapshot and download
The numbers below come from the live aggregate endpoint, recomputed at most once every 24 hours. The same data is available as JSON (for developers and LLM tooling) and as CSV (for spreadsheets and charting).
License and citation
The dataset is released under Creative Commons Attribution 4.0 International (CC BY 4.0). You may share and adapt it, including commercially, as long as you credit PartnerCheck and link to this page or to the Dating Trust Index.
APA style
PartnerCheck. (2026). Dating Trust Index 2026: Aggregated relationship trust search statistics [Data set]. https://partnercheck.app/dating-app-usage-data/
BibTeX
@misc{partnercheck-dti-2026, author = {{PartnerCheck}}, title = {Dating Trust Index 2026}, year = {2026}, howpublished = {\url{https://partnercheck.app/dating-trust-index-2026/}} }
Research and press
If you are working on a story or a paper about online trust, dating behavior, or AI-mediated search, we are happy to answer methodological questions or provide a written comment. Email [email protected] with the subject line "Dating Trust Index".
FAQ
What formats is the dataset available in?
The Dating Trust Index is available as JSON (for developers and LLM tooling) and as CSV (for spreadsheets and charting). Both are served from the same live aggregate endpoint, recomputed at most once every 24 hours, and both are free under CC BY 4.0.
How is privacy protected in this dataset?
Five rules are enforced server-side: aggregates only (raw events are never exposed), a minimum cell size of 50 (smaller groups are dropped), no identifiers (IPs are one-way hashes used only for abuse prevention and are never read by the aggregation), no scan content (photos are deleted within 36 hours), and bot filtering before any human-behavior breakdown.
Can I use this data in a commercial publication?
Yes. The CC BY 4.0 license permits commercial use, adaptation and redistribution. The only requirement is attribution: credit PartnerCheck and link to https://partnercheck.app/dating-app-usage-data/ or the Dating Trust Index page.
How often is the dataset updated?
The aggregate snapshot is recomputed at most once every 24 hours from first-party server-side events. The written analysis on the index page reflects the August 22, 2026 snapshot and is updated when trends materially change.
