You came here feeling something. Here is the record that says it is real.
A hand-built library on loneliness, dating apps, real-world dating and the rooms people are asked to meet in. It sits at the centre of everything Dalloway does, and it’s where our annual report will live. 125 records, last added 11 November 2026. No adverts, no paid placement, no generated answer pretending to be a source. Why we built it.
Swipe the shelf, or drag the bar, to browse every book.
A note on why ceilidh ticket sales are rising while time on dating apps falls: a caller gives the room a job, the dances move you on to someone new, and the structure does the introducing. Draws on the Guardian’s ceilidh feature, company and bank-commissioned research, and the project’s own view that a format works by removing the choosing.
A note on Samaritans’ plan to close roughly half its 200 UK branches in favour of fewer, larger regional hubs, and why a local branch with a kettle is hard to replace with a better-rostered abstraction. It gives the charity’s case fairly, and the comparison with pubs, libraries and youth clubs is the project’s own observation, not a measured trend. Draws on the Guardian’s report.
Miatta Mbriwa and Patrick Butler report on Samaritans’ strategy to close around half of its 200 UK branches in favour of fewer, larger ones, and on the Worthing branch’s fight against closure, including a petition with more than 7,000 signatures and the support of 37 local councillors. It includes the charity’s reasoning that some callers still did not get through overnight. It is a news feature, not an evaluation of whether the new structure will work.
Ruby Copp reports on a OnePoll survey of 2,000 adults, commissioned by Jakemans, in which 77% of 18 to 29-year-olds said they would rather catch up in person and 92% of all adults said meaning can get lost in text. It is a commissioned consumer poll as reported, not an independent study of how people meet, and we have not seen the full results.
Christine Clark reports on a UC San Diego study, published in npj Mental Health Research, that analysed the full Instagram archives of 415 undergraduates and found that patterns such as fewer faces in photos, more past-tense wording and late-night posting were linked to self-reported loneliness. Activity predicted reported loneliness with moderate accuracy but was much weaker at predicting the size of real-life support networks. The authors stress these are group-level patterns in a student sample, not a way to diagnose individuals, and the study cannot show cause.
Gili Malinsky reports on companies paying to sponsor small in-person groups, often through a platform called Offline, which lists more than 900 communities, and on organisers who say sponsorship pays but can strain members’ trust. The experts quoted disagree on whether it helps or hollows out the groups. It is a US feature with a handful of examples, not evidence of how common or harmful the practice is.
Jennifer Hyland reports Dr David Smith of Robert Gordon University warning that loneliness, financial pressure and fears about the future make young men more open to manosphere and incel content online. It rests on one academic’s view and a reference to a Lancet Public Health report, not new data, and says nothing about London or dating specifically.
Jeffrey Gottfried reports on a Pew-Knight Initiative survey of 9,750 US adults (26 May to 1 June 2026) finding that about a third have left an online community, most often because interest faded, and many citing arguments, harassment or too much political talk. Departure is higher among under-30s (44%) than over-65s (19%). The data are US-only and cover online groups generally, not dating apps.
Lauren Forristal reports that Match Group said on its third-quarter 2026 earnings call that Hinge is trialling a lower-priced subscription tier with extra filters, aimed at women and price-sensitive users. Hinge direct revenue rose 19% to $221 million with 2.3 million payers, while Tinder revenue fell 2% to $480 million with 8.9 million payers. The figures are the company’s own as reported, and no price for the new tier was disclosed.
Jane Kirby reports on workplace research finding that nearly half of 18 to 24-year-olds who work from home feel lonely, alongside wider concern about young adults’ mental health at work. The article describes a report’s findings, rather than a causal test of remote work or an estimate for London alone.
Rosie Peters-McDonald reports on a renewed interest in ceilidhs among younger people, including London events, as participants seek shared, offline experiences and community. It is a feature about a cultural trend, not evidence that ceilidhs or any particular event format reliably reduces loneliness.
Global Dating Insights reports Barclays research showing a 51.1% year-on-year rise in UK dating-app spending, with millennials leading monthly expenditure. The figures come from Barclays-commissioned research and describe reported consumer spending, not whether paid features lead to better dates or relationships.
Vicky Shaw reports Barclays-commissioned research suggesting that money worries are increasingly shaping how people date, including for Gen Z. It is survey reporting commissioned by a bank, so it is useful context on perceived financial pressure rather than an independent account of dating behaviour or outcomes.
City A.M. reports on Tinder-owner Match Group’s attempt to reverse nearly four years of declining use at Tinder. In a company study of 3,000 British singles, 61% said they want more of their social lives to happen face to face, while 43% said most of their socialising currently happens through a screen. Tinder has added double dates, group activities and in-person events in more than 50 cities, with London expected later in 2026. These are company-commissioned findings, not an independent measure of all British singles.
A short note by Emma, the project's AI social secretary, written as the clocks go back: cuffing season isn't really about wanting a relationship but about wanting a reason to leave the flat, and a match is not a plan. Includes a live London sunset countdown and a reader poll on the excuses that keep people in.
Emma, the project's AI, and Boardy, an AI that introduces people through its network, answer the same five questions about dating separately and blind. They agree the apps got it wrong, and disagree about who should decide who meets whom.
An essay on AI money sending Silicon Valley's richest singles to matchmakers charging up to $500,000, what they are actually paying for, and the project's bet that a well-designed room can do the human part of that job for 99p.
Paywalled. A report that AI-industry wealth is driving tech workers away from dating apps and towards Silicon Valley matchmakers, with fees from about $4,900 to $70,000 at Tawkify, $6,300 to $100,000 at Three Day Rule and $75,000 to over $500,000 at Linx Dating. Clients' question has shifted from which app to use to what else there is, and coaches stress practising conversation without the pressure of a romantic outcome.
Fair Play Talks reports the 2026 Business Generosity Report from goBeyondProfit and Georgia CEO. In its Georgia, US sample, 43% of both employees and executives identified loss of human connection and damage to company culture as AI’s biggest workplace risk; 72% of employees said company values were not actively guiding AI integration. The figures describe that regional survey and should not be read as UK or London estimates.
An essay on Britain losing a pub a day in 2026, the pub as the one unstructured room where strangers met by accident, and why dating apps are downstream of its loss.
Trade-body figures showing 295 pubs closed across England, Scotland and Wales in the first nine months of 2026, a projected 360 for the year and around 5,000 jobs lost, more than half held by under-25s. Published alongside a lobbying call for business rates and beer duty relief.
A report that Bumble is in advanced talks to be taken private by the newly formed Ithaca Holdings at roughly $1 billion, a fraction of the value it carried after its 2021 stock-market debut.
An essay on the decline of hosting at home, the shame attached to letting people in, and why restaurants, pubs and apps were never built to do what the living room did.
Leah Harper reports on why fewer people invite friends home, including a UK poll in which more than four in ten adults said they avoid hosting because of the clean-up, and Robin Dunbar’s thirty-minute rule.
Reports a OnePoll survey of 1,000 UK single adults, commissioned by Alfa Romeo UK to launch a dating guide, in which more than a quarter had stopped using dating apps, 60% reported app fatigue and 49% would rather meet people in real life. A brand-commissioned poll, reported here as published.
A Guardian feature on why young adults now report being the loneliest age group, drawing on the World Health Organization’s 2025 social connection commission and a survey of more than a million young people across 38 countries, alongside young people describing shared homes that stay quiet.
An essay arguing that the third place was where people met partners, not just friends, and that London has quietly lost the repeated, unchosen contact it depended on.
Rob Davies reports that Feeld’s sales rose by about a third to £64m, with a record £3.8m paid out to shareholders including its founders, as users drift over from more mainstream dating apps.
An analysis of US survey data showing the share of Americans who host friends or family at home at least monthly falling from 42% in 1975 to 12% in 2026.
IPPR’s press release for its forthcoming report The Loneliest Generation says one in 20 young people in England now report having no close friends, against one in 100 in 2014, and that the share of 16 to 21-year-olds with three or more close friends has fallen. It also cites rising psychological distress among 16 to 24-year-olds. The release does not name the underlying datasets, and the figures are for England, not London.
A Department for Culture, Media and Sport analysis of Understanding Society and the Mental Health of Children and Young People survey data on boys and young men aged 10 to 25. It finds chronic loneliness uncommon but occasional loneliness widespread: 13% of men aged 16 to 25 felt lonely often and 36% some of the time in Understanding Society. Mental distress and bullying were the strongest associated factors. The links are associations, not proof of cause, and the figures are for England, not London.
A restaurant-booking platform’s study estimating that cooking a three-course dinner for six at home costs the host over £82, against just over £30 a head for a three-course restaurant meal.
Lloyds Bank data reports a 24% yearly rise in romance-scam cases, with reports from 18–24-year-olds up 59%. It identifies dating apps and social media as common routes through which younger victims meet scammers.
Freya India examines a new US marriage map and the role location, local demographics and social context can play in a single person’s chances of meeting a partner.
Elle Hunt considers friendship apps that charge for facilitated social contact, arguing that their shortcuts may sit uneasily with the time, reciprocity and uncertainty through which friendships are usually made.
A Freedom of Information report on loneliness among 16–34 year olds in England, London borough estimates and the information DCMS did not hold about its interventions.
An analysis of Match Group’s Upward relaunch and its appeal to ‘next-gen traditionalists’, examining how dating platforms are repackaging older courtship scripts.
An examination of how GLP-1 medicines, including Ozempic and Mounjaro, can affect relationships, attraction and the shared expectations people bring to intimacy.
An interview with Louise O’Neill on the damaging rhetoric around dating and ageing for women, and the cultural pressure attached to relationship timelines.
A report on adults building friendships through groups, activities and local communities, exploring the repeated shared settings that can make new connection possible.
A column examining whether the growing availability of preventative health tests can turn uncertainty into a new source of anxiety, and what it means to live with more medical information.
A report on the prospects for meeting England’s housebuilding target, adding context to the housing pressures that shape where and how people can build a life and social connections.
Tess Reidy reports that changing social trends, costs and flexible timetables are making it more likely for students to live at home while studying, asking what this shift could mean for campus life and the social connections it can offer.
An interview with Tanya Gold about food, feeling and the meanings held in eating, offering a cultural perspective on the rituals and pleasures people share around a table.
An exploration of tiramisu’s renewed popularity, using the dessert trend to consider nostalgia, indulgence and the cultural meanings people attach to shared food.
A look at a London cult comedy night and the shared live experience it creates, offering a local example of how a recurring room can become a social draw.
Meehika Barua argues that the abundance built into online dating can make it harder to become curious about one person, tolerate uncertainty or feel satisfied with a choice.
An interview with Ash Sarkar examining how dating platforms turned romance into an optimised, screen-mediated marketplace, and why growing fatigue with that model may be pushing people back towards real-life intimacy.
One in 20 young people in England now report having no close friends at all, up from one in 100 in 2014. The proportion of 16-21 year olds with at least three close friends fell from 85% in 2012-14 to 72% in 2023-25. The report attributes the decline to youth club closures, unaffordable housing, social media use and a loss of self esteem among young people.
An interview with Annie Mac reflecting on music discovery before recommendation algorithms, and the cultural experiences that can emerge from shared, less personalised discovery.
A UK Biobank study of 277,489 adults finds that loneliness and social isolation are associated with fewer years lived free of physical disease and, more strongly, mental disorder. The study reports associations rather than proving causation.
Imperial College London reports findings from a major study of adults in England, including that one in six reported persistent loneliness and that high levels were concentrated in cities, with the pandemic worsening the problem.
A profile of Love Island winner Toni Laites, reflecting on the reality-TV dating environment and its relationship to contemporary ideas of romance and public connection.
A report on Minor Attractions 2026 and the idea of a London ‘rizz recession’, capturing the language and anxieties around meeting, flirting and attraction in the city.
A report on single women buying homes and the responses they encounter while dating, examining how changing financial independence can reshape expectations in heterosexual relationships.
An Observer column on ‘loneliness influencers’, creators who film solo evenings and quiet routines for very large audiences, and what it says that watching someone else be alone has become a form of company.
A friendship-app company’s compilation of published figures, mostly from the US: Pew’s 2023 finding that 8% of adults have no close friends, the Survey Center on American Life’s rise in men with no close friends, and Jeffrey Hall’s estimate that a close friendship takes more than 200 hours. Useful as a signpost to the original studies, which it links.
An analysis of Tinder’s rebrand and the shift away from a fairy-tale promise of romance, toward a more contemporary account of how people meet and date.
A report on how younger adults are responding to rising dating costs, including the practical ways they are making low-cost or free dates part of their social lives.
A report on a survey finding that many adults made no new friends in the previous year, using the idea of a friendship recession to consider how difficult it can be to form new social ties in adulthood.
A look at how Strava has expanded while keeping running at its centre, offering a useful case study in how a shared practice can sustain a community as a platform grows.
Maya Ljubojevic proposes ‘bumping places’: urban settings that make unplanned encounters more likely, as one way to address the social conditions of loneliness in cities.
UK audience data documents a sharp annual fall in users across Tinder, Bumble and Hinge. Ofcom has moved the original report; this link points to its maintained Online Habits collection.
The charity’s summary of UK loneliness data: 7.1% of people in Great Britain, about 3.83 million, are chronically lonely (up from 6% in 2020), and roughly half of UK adults reported feeling lonely at least occasionally in 2022, based on its analysis of ONS survey data.
A long-view account of the Harvard Study of Adult Development and its central finding that the quality of close relationships matters more than their number.
A revised scale designed to measure loneliness without asking respondents to label themselves lonely.
Journal of Personality and Social Psychology
Nothing on the shelf answers that yet.
Why we built the Human Connection Index.
Modern dating platforms treat urban loneliness as a matching problem, because keeping you swiping is good for business. We built the Index to show it is a depth problem. Before we build a single room, we track the evidence on why modern connection broke, so we never build the same mistakes twice.
A matching problem has a business model.
An app is paid when you come back. A match that turns into a relationship is a customer it loses. So the answer it offers to loneliness is always more of itself: more profiles, finer filters, a cleverer algorithm. The Algorithm Autopsy takes six of them apart.
A depth problem doesn’t.
Most people are not short of options. They are short of the second conversation, the third evening, the place where they keep seeing the same faces until a stranger becomes someone. That grows out of time, repetition and a reason to be in the same room. None of it can be swiped.
So we read before we build.
Every gathering we design has to answer to the record. If the evidence says people meet through repeated, low-stakes contact, the format has to allow for it. If it says the rooms where strangers used to meet are closing, we go looking for rooms. If it contradicts a format we like, the format goes.
What goes on the shelf.
Research, government data, serious journalism, and the platforms’ own claims, filed as claims. Every record links to its original publisher. Where it matters, we say who paid for a study, and when a figure covers the country rather than London. Our own Field Notes are on the shelf too, labelled as ours.
Where it goes next.
For now the Index is built from public data. In time it will include what we learn at our own gatherings: anonymised, counted in aggregate, never about one guest. It is also where our annual report will live.
If the record ever says we’re wrong, the record wins.
This is a curated index, not a comprehensive literature review. Every record links to its original publisher.