The Imagined User

The services exist. This is what the reports mean when they describe the available provision for lonely and socially isolated adults in New South Wales. The services exist. They are funded. They have coordinators and referral pathways and outcome measures and websites with photographs of people eating lunch together at long tables in community halls.

The provision is documented. The documentation is extensive. The gap, according to the documentation, is in awareness—people who need the services do not always know the services exist.

I knew the services existed. I spent some time looking at what they offered.

The person the services appeared to have been designed for was over sixty-five, mobile, in possession of a car, comfortable in group settings, interested in structured social activity, without significant dietary complexity, and within reasonable distance of a community hall or similar venue where lunch is served on a weekday. This person is not described explicitly in the service documentation. They are implied by the format of what is offered—the lunch outings, the group excursions, the social mornings, the activity clubs that meet at locations not served by public transport at times that assume the participant has a flexible daytime schedule and a vehicle to reach the venue.

The imagined user is not a conspiracy. They are a design outcome—the person who emerged from the accumulation of decisions about what kinds of activity are fundable, measurable, organisable, and low-risk enough to insure. Lunch is fundable. Lunch is measurable—attendance can be counted, frequency can be reported, the number of meals served can stand as evidence of programme reach. Lunch is organisable into a fixed time and location with a clear beginning and end. Lunch is low-risk in the sense that eating together is a socially legible activity that does not require specialist facilitation and does not produce the kind of liability that more individually tailored support might generate.

Lunch works for the system. Whether lunch works for the person the system was established to serve is a question the system has not been primarily designed to ask.

 

The pathway to the lunch, even for the imagined user, begins with the GP. Several of the main befriending and social connection services in New South Wales require a referral from a general practitioner before access is granted. The referral requirement does two things simultaneously. It routes the issue into a medical frame, which makes it fundable through health commissioning channels rather than through social care or community development budgets that may not exist. And it controls the volume of demand, ensuring that the service does not have to manage self-referral from anyone who encounters it and considers themselves eligible.

The practical consequence, in a health system where bulk billing has become the exception rather than the rule in most parts of Sydney, is that accessing a referral to a social connection service may cost as much as one hundred and fifty dollars. The consultation may not produce the referral, because the GP may determine that loneliness or social isolation does not constitute a medical issue within the scope of the consultation, which is both correct and unhelpful.

The consultation may produce a different outcome—a prescription, a recommendation for a different service, a suggestion to return if the situation does not improve—that substitutes a clinical response for the social one the person was seeking. The medication addresses the symptom the GP can clinically manage. The condition that produced the symptom is not within the GP’s referral pathway to address.

The pathway is: pay, translate your situation into the language the clinical frame requires, hope the translation produces a referral, access the service if it does. At each stage there is a point at which the pathway may close—the cost of the consultation, the GP’s assessment that the issue is not clinical, the absence of an appropriate service within the referral network, the mismatch between the service that is available and the situation of the person who has been referred to it.

The person who has paid for the consultation and received the referral and arrived at the service and found that the service is a fortnightly lunch in a location forty minutes away by car, at a time they are not available, for a group format they find draining rather than sustaining, has been processed by the pathway without being reached by it.

 

The services rarely name loneliness directly. This is not an oversight. Loneliness is difficult to operationalise—it is subjective, variable, resistant to the kind of standardised assessment that funders require—and it is therefore difficult to measure as an outcome.

What can be measured is participation: the number of people who attended, the frequency of attendance, the retention rate over the programme period. Participation can stand in for connection in the reporting document. The report does not have to demonstrate that connection occurred. It has to demonstrate that participation occurred, and participation can be demonstrated by the attendance register.

The language of the service reflects this substitution. Loneliness becomes social isolation. Social isolation becomes reduced participation. Reduced participation becomes a need for engagement activities. Engagement activities are what the lunch provides. The lunch addresses the operationalised version of the problem rather than the problem itself, which is the specific felt absence of connection that the person who considered attending the lunch was experiencing when they navigated the pathway.

The word loneliness does not appear in most service descriptions because naming loneliness would require the service to demonstrate that it addresses loneliness, which would require measuring whether connection occurred, which is much harder than measuring whether attendance occurred. The language protects the model from the question of whether the model is producing what it claims to produce.

 

The filter stack is the structure that emerges from the accumulation of these design choices. The imagined user passes through the filters without noticing them—is over sixty-five, drives, is comfortable in groups, is available on weekday afternoons, eats without complexity, lives within a manageable distance of the venue. The filters were not designed to exclude anyone. They were designed to define what the service can provide within its operational constraints. The exclusion is the byproduct of the design, not its intention.

The person who does not match the imagined user encounters the filters at multiple points. The GP filter excludes the person whose situation does not fit the clinical frame or who cannot afford the consultation. The mobility filter excludes the person without a car in an area not served by public transport at the relevant times. The group filter excludes the person for whom group social settings produce anxiety or exhaustion rather than connection. The dietary filter excludes the person whose eating requires more accommodation than the standard lunch format provides. The distance filter excludes the person for whom forty minutes each way represents a commitment that the available reward does not justify.

None of these filters is unreasonable in isolation. The GP referral exists to manage demand and fund the service. The transport assumption exists because the service is located where it is located for reasons of venue availability and cost. The group format exists because groups are easier to organise and report than one-to-one arrangements. The meal format exists because shared eating is a culturally legible social activity with clear logistical parameters. Each filter has a reason. The accumulation of reasons produces a service that is accessible to a relatively narrow band of people, presented as though it addresses a condition that affects a much wider one.

 

The quiet risk in this presentation is the person who reads the service description, identifies that they do not fit the model, and concludes that the problem is themselves. The service is what help looks like. Help looks like the fortnightly lunch. The fortnightly lunch is not accessible to them, or is not suitable for them, or does not address the specific quality of what they are experiencing. If help looks like the fortnightly lunch and the fortnightly lunch is not for them, then perhaps they are not the kind of person who can be helped. Perhaps the fault is in their preferences, their situation, their unwillingness to try the thing that is available.

This is the inversion that the filter stack produces. The person has not been excluded by their own inadequacy. They have been excluded by the mismatch between their situation and the imagined user the system was optimised for. But the system does not present itself as optimised for a specific user. It presents itself as the available help. The gap between the presentation and the reality produces, for the person who cannot access what is presented as universal, the experience of personal failure.

The experience of personal failure is its own form of isolation, distinct from the structural isolation that produced the need for the service in the first place and layered on top of it by the encounter with a system that has classified them as outside its design without saying so.

 

The system works well for the person it has already imagined. For that person, the pathway is navigable, the service is appropriate, the lunch is a genuine occasion for the incidental contact that might, over time, produce something more. The imagined user is a real person and the service provides them with something real.

The person who is not the imagined user is not mentioned in the service description. They are not mentioned in the report that documents the service’s reach. They are not in the attendance register. They are counted, if they are counted at all, in the survey data that shows significant numbers of people reporting chronic loneliness despite the existence of the services that were established to address it.

The survey does not explain why they are not in the attendance register.

The service does not ask.

The report records the attendance.

The imagined user attended.