The Invisible Step Before Automation: Why Robotics Designers Keep Solving the Wrong Problem

Before you vacuum, you move toys off the floor, push chairs aside, and relocate the pile of magazines that somehow migrated from the shelf to the carpet. Before you wash clothes, you sort darks from lights, check pockets for tissues, place delicates into mesh nets, and pre-treat the curry stain on your daughter’s school shirt. Before you cook dinner, you clear the counter, put away the breakfast dishes that are still drying, and reorganise the fridge to find the vegetables you bought two days ago. 

None of this is the task itself. All of it is essential to the task being done. 

We call this the ‘pre-work problem’ — and after more than 20 years of ethnographic research into Japanese household behaviour, we believe it is the single most overlooked barrier to successful home automation and robotics adoption. 

If you’re designing autonomous systems, assistive robots, or smart home products for the Japanese market, this article is for you. Because what we’ve found, consistently, is that the gap between what technology automates and what users actually do is not a feature gap. It’s a comprehension gap. And it starts with the steps nobody sees. 

There are many great examples of technology companies that design relevant and intuitive household robots or automation systems, but there are also many that do not. The latter tend to define the problem by the visible task: vacuuming, washing, folding, cooking. The product brief says ‘automate laundry’ or ‘build a cleaning robot.’ The engineering team optimises for suction power, wash cycles, or folding speed. 

But when you sit in someone’s home and watch them actually do laundry — not in a lab, in their 65-square-metre apartment in Setagaya with two children and hanging their laundry on poles and hangers in the living room — you see something completely different. 

You see a woman shaking out each item before hanging it, because wrinkles that set during drying are nearly impossible to iron out and she learned this by watching her mother do the same thing thirty years ago. You see her spacing items precisely on the pole, because insufficient airflow means clothes won’t dry by evening and she’ll need to run the dryer, which costs money and damages fabrics. You see her checking the weather app mid-hang, because if pollen counts are high, she’ll need to bring everything indoors and reconfigure the drying setup in the bathroom. 

The washing machine did its job. The real work hasn’t even started. 

This isn’t a story about laundry. It’s a story about a fundamental design error that recurs across every category of home automation we’ve studied. 

In our cleaning research, we found that Japanese households clean with extraordinary frequency — daily or multiple times daily in many cases. Over a third clean more than once a day. But the single biggest source of frustration isn’t the cleaning itself. It’s what has to happen before cleaning can begin: moving objects out of the way. Books, toys, shoes, charging cables, the children’s school bags — all of it has to be relocated before a vacuum (robotic or otherwise) can do its job. 

Robotic vacuum owners — and ownership sits at around 40% among the households we studied — described the same frustration. The robot handles the floor. But someone still has to prepare the floor for the robot. The pre-work remains entirely manual, entirely invisible to the product team, and entirely unaddressed by the technology. 

This pattern repeats everywhere we look. In cooking: the preparation of ingredients, the clearing of surfaces, the mental calculation of what’s in the fridge and what needs using before it expires. In laundry: the sorting, the netting, the pocket-checking, the stain pre-treatment. Every automated task sits inside a larger workflow of human judgement and physical preparation that the automation doesn’t touch. 

To understand why the pre-work problem is especially acute in Japan, you need to understand something about the cultural role of household tasks. 

In our research, cleaning and laundry are not neutral chores. They are moral activities. A clean home reflects social responsibility, respect for family, and personal discipline. The desire isn’t just for a hygienic outcome — it’s for an emotionally satisfying one. Freshness, crispness, the absence of any lingering smell: these are not nice-to-haves. They are baseline expectations that carry social weight. 

This means the consequences of automation failure are disproportionately high. A robotic vacuum that misses hair in the corner isn’t just inefficient — it’s a source of genuine distress. A washing machine that leaves clothes smelling faintly damp isn’t just disappointing — it represents a failure of care. 

Japanese consumers told us repeatedly that they want the ‘thinking’ removed from routine tasks. They’re tired of deciding which programme to use, which detergent suits which fabric, whether today’s humidity means indoor or outdoor drying. But — and this is critical — they want full transparency into how those decisions are being made. They want to understand why the machine chose a particular cycle. They want the ability to override. 

This is not a contradiction. It’s a design principle: remove the cognitive load, but never remove the control. 

Across multiple studies and demographic groups, we’ve identified what we believe is the optimal design paradigm for Japanese consumers interacting with autonomous household systems. 

It’s not full automation. Japanese consumers across every generation we’ve studied resist black-box decision-making. When a system makes choices they can’t see or understand, trust erodes — often permanently. A smart speaker that gives irrelevant responses, a robot vacuum that misses visible dirt, a washing machine that can’t explain its cycle selection: each of these experiences damages the relationship between user and technology in ways that are very difficult to repair. 

It’s also not full manual control with a hundred settings. Consumers are already exhausted by the cognitive load of daily household management. They don’t want more choices. They want better defaults. 

The sweet spot is what we describe as ‘guided autonomy’: three to five well-designed presets that cover the most common scenarios, plus simple, transparent customisation for edge cases. The system makes the decision, but the user can see why and adjust if needed. Over time, the system learns their preferences — but it explains what it’s learned. 

This maps directly onto what our millennial parent participants told us about household technology: they want robots for repetitive, time-consuming, and physically exhausting tasks — but they need the ability to customise processes and verify results. ‘Doing it my way’ was a recurring theme. Not because they’re control freaks, but because the emotional stakes of household tasks are high, and trust requires understanding. 

There’s another dimension to the pre-work problem that has nothing to do with logistics and everything to do with emotion. 

Across our research, one of the most consistent findings is what we call the ‘appreciation gap.’ The people who do the bulk of household work — overwhelmingly women, whether homemakers, part-time, or full-time workers — describe a profound lack of acknowledgement from family members. The work is invisible. The effort goes unrecognised. The emotional toll accumulates. 

Many told us they crave recognition. Parents told us household management is emotionally exhausting. Cleaning was described as not just physically but ‘emotionally loaded.’ 

This has a direct design implication that most technology companies ignore: a household robot or autonomous system that simply completes a task silently is missing an opportunity. Our research suggests that a system with some form of human interaction — even something as simple as a verbal acknowledging that the person did a “good job,” that a task has been completed, or recognising the complexity of the day’s workload — creates a qualitatively different relationship with the user. 

It sounds small. It isn’t. In a culture where household labour is systematically undervalued yet held to extraordinarily high standards, the technology that notices — that treats the work as worthy of acknowledgement — has an emotional advantage that no feature specification can match. 

If you’re designing human-machine interfaces for household robotics or autonomous systems targeting the Japanese market, our research suggests five principles worth embedding into your design process: 

1. Map the full workflow, not the visible task 

Before designing for any automated task, map every step the human currently performs, including preparation, transition, and cleanup. The pre-work is where the real pain lives. Designing a laundry-folding robot without addressing sorting, netting, and hanging is designing a solution that starts at step seven of a twelve-step process. 

2. Design for emotional standards, not just functional ones 

Japanese consumers judge household outcomes against moral and emotional benchmarks, not just hygiene or efficiency metrics. Your robot doesn’t just need to clean the floor. It needs to produce a result that feels clean — visibly, olfactorily, and emotionally. Test for satisfaction, not just performance. 

3. Implement guided autonomy 

Build systems that make intelligent defaults visible and overridable. Three to five presets, clear reasoning, easy customisation relevant to Japanese users. Never hide the decision logic. Japanese consumers will tolerate imperfection far more readily than opacity. 

4. Acknowledge the work 

Build interaction patterns that recognise effort and completion. This doesn’t require sophisticated AI — it requires designers who understand that household labour is emotionally significant and systematically undervalued. 

5. Account for space constraints 

Japanese homes average 30–40% smaller than their Western equivalents, with pre-specified installation spaces for appliances and very limited storage. Any physical product that occupies floor space, requires dedicated storage, or can’t integrate into existing room layouts will face resistance regardless of how well it performs. Design for the Japanese 65-square-metre apartment, not the test facility. 

The pre-work problem is, in many ways, good news for ambitious designers. 

It means the market for household automation is dramatically underserved — not because products don’t exist, but because they’re solving the wrong layer of the problem. The company that designs an autonomous system capable of understanding and assisting with the full workflow — including the invisible preparation steps — will have a structural advantage that pure task-automation cannot match. 

It also means that ethnographic research — the kind that requires sitting in homes, watching routines unfold in real time, and listening to the frustrations people don’t think to articulate in a survey — is not a nice-to-have in your product development process. It’s the difference between designing something technically impressive and designing something people will actually trust and use. 

Because trust, in the end, isn’t built by what your robot can do. It’s built by how well your robot understands what the human was already doing before the robot arrived. 

CarterJMRN is a Japan-based market research agency, specialising in qualitative research that connects global technology companies with the real behaviours, attitudes, and cultural dynamics of Japanese consumers. Our work spans robotics, autonomous systems, household technology, aging, and caregiving — with particular depth in hard-to-access populations including rural communities, older workers, and multi-generational households. 

Want to go deeper? Look at our case studies and other blog articles: 

‘​An Immersive Safari into the Life of Older Adults in Rural Areas’ 

‘Exploring Usability and User Experience of an Autonomous Driving System’ 

‘A Cross-Cultural View of Expert UX Research in Japan’