Beyond Tier 1 and Gen Z in China

There is no single “Chinese user”. City tier, age group and income can all help researchers build a participant sample in mainland China, but none tells you enough about a person on its own. Someone in a lower-tier city may be very comfortable with mobile shopping. A young professional in Shanghai may have less money left after essential costs than an older homeowner elsewhere. Two people labelled “Gen Z” may be at very different stages of education, work and financial independence.

This matters in diary studies because they show behaviour in real situations and over time. If the sample relies on broad labels, the study may record genuine behaviour but still give a misleading picture of the people it represents.

A stronger China sample looks at five areas: place, money and household situation, age group and stage of life, digital habits and confidence, and experience with the product category. City tier and generation are useful starting points, not final answers.

China’s city tiers are useful shorthand, but there is no single permanent system. Different organisations group cities for different reasons. The same city may appear in a different tier depending on the ranking, year and method used.

One well-known example is Yicai’s city-attractiveness work. Its 2025 ranking covered 337 cities at prefecture level or above and used five measures: business activity, transport and connections, how active residents are, growth in newer industries, and future potential. In 2026, Yicai changed the public report: it focused on five urban-development trends instead of publishing an actual ranked list, although results remain available through Rising Lab’s mini-program. This change is another reason to record the source, year and method whenever a study uses city tiers. The ranking can tell researchers something useful about a place. It does not mean that everyone in the same tier has similar income, confidence, goals or digital habits.

City tier is best used as information about a place. It can suggest how many services, shops and transport options are available, how visible brands are, how developed a product category is, and which online and offline services people can use. Researchers still need to connect those conditions to each participant’s daily life.

Researchers sometimes use city tier as a shortcut for things it does not measure, such as wealth, experience, willingness to try new products or digital confidence. This can turn a useful sampling detail into a stereotype.

This is a practical problem, not just an academic one. A published study compared reactions to creative advertising in higher- and lower-tier Chinese cities. It found little difference between the groups and concluded that city tier was not a reliable way to divide people for that question. This does not mean tier never matters. It means researchers should use it only when there is a clear reason to expect place to affect behaviour. See Bilby et al., “Tiers and fears”.

Whether city tier matters also depends on the product or service being studied. Place may strongly affect grocery delivery, public charging or access to premium shops. It may matter less for a fully digital service available across the country. Before setting quotas by tier, ask which parts of the experience are likely to change because of where someone lives.

The income gap between urban and rural China is large, but that is only a starting point. According to the National Bureau of Statistics of China, median yearly disposable income per person in 2025 was RMB 51,115 for urban residents and RMB 20,711 for rural residents. Average yearly spending per person was RMB 35,869 in urban areas and RMB 20,259 in rural areas.

These figures show a large overall difference, but national averages do not tell you how much money one participant can actually use. Who lives in the household, housing costs, family support, savings, debt, job security and the number of dependants can all change what the same personal income means.

A lower city tier should not automatically be read as lower spending power. Bain and Worldpanel reported that tier-three to tier-five cities made up around 80% of China’s FMCG market growth in the first three quarters of 2025. This finding applies to a specific market and time period; it does not describe every lower-tier consumer. It does show why city rank should not be used as a simple measure of spending potential.

Source: National Bureau of Statistics of China, Households’ Income and Consumption Expenditure in 2025, published 20 January 2026. The national household survey covers about 160,000 households and reports weighted aggregate estimates. Income is shown as a median; expenditure as an average.

An income question needs a clear unit and purpose. Are you asking about personal income, household income or money left after essential costs? Is the amount monthly or yearly, before tax or after tax? Do you need to understand access to a product, sensitivity to price, or whether someone can make a purchase without asking another person?

For some studies, product ownership, the value of a recent purchase or responsibility for a household decision may be more useful than a broad income band. In others, the difference between stated income and everyday spending limits may be exactly what the diary study needs to explore.

Generation labels are easy to copy from one market to another, but people’s experiences are not. Using a Western Gen Z age range in mainland China may be convenient for a regional brief, but it can hide important local differences.

Chinese consumer research often uses groups based on birth decade or half-decade, such as 90后 (born in the 1990s), 95后 (usually 1995–1999) and 00后 (born in the 2000s). Kantar, for example, has described Chinese Gen Z as post-95 and post-00 consumers. That is one business definition, not a fixed rule. Always state the birth years used in the study. See Kantar’s China Gen Z explanation.

Birth year also does not explain someone’s current life. In 2026, the broad post-95 and post-00 group includes school and university students, people starting their careers, experienced workers, people living with parents, independent renters and parents of young children. Their use of platforms, experience with brands and financial situation can differ even when they share a generation label.

A better brief keeps the age group when it helps comparison, then adds the details that matter to the behaviour: education or employment, financial independence, who they live with, current city, how long they have lived there, and whether they make the decision being studied.

China is a highly connected, mobile-first internet market. CNNIC’s 57th Statistical Report on China’s Internet Development said that internet use reached 80.1% of the population by December 2025. Penetration was 85.1% in urban areas and 69.5% in rural areas.

The same report shows why access should not be confused with confidence. Internet use among people aged 60 and over reached 53.7%, which still leaves a large difference within that age group. Having access, using something often, feeling confident and using many different services are separate things.

In research, define digital confidence by what people actually do. Someone may be very comfortable with messaging, short video and mobile payments but rarely use desktop work tools, smart-home devices or a certain type of app. Another person may own several devices but still ask family members for help with unfamiliar tasks.

Ask which services people use, how often they use them, which devices they choose, whether they move between different platforms, how they solve problems, and whether they can complete the photo, video or written tasks in the study. Owning a device does not prove that someone can use it confidently.

Source: CNNIC, 57th Statistical Report on China’s Internet Development, Figure 13, data to December 2025, published 5 February 2026. Internet penetration measures access, not confidence or diary-task completion.

A useful sample does not need every possible personal detail. It needs the smallest set of details that could change the behaviour being studied. These five areas help researchers decide what to include.

Record the participant’s current city and the local details that matter to the product: region, urban or rural setting, commute, available services, climate, housing type, or distance from shops and transport. City tier can be included, but it should not replace these more useful details.

Choose the money-related measure that affects the decision. This might be personal or household income, money left after essential costs, housing costs, job security, financial independence, ability to make a purchase, or current product ownership. Set income bands that fit the people and product being studied instead of copying global bands.

Use a clearly defined birth-year group when it helps comparison, but combine it with education, employment, household and family stage. A post-00 university student and a post-00 full-time worker may share cultural references but have very different routines and limits.

Ask about real behaviour across the devices, services and tasks that matter to the study. Find out what participants use confidently, what they avoid, how they learn a new service and where they ask for help. This tells you more than simply calling someone “digitally savvy”.

Find out whether participants have never considered the category, are thinking about it, have just started using it, use it often, have stopped using it or actively avoid it. In a study that runs over time, this experience often tells you more about what a diary can capture than a broad demographic label.

“Gen Z homeowners in tier-one and tier-three cities” gives the recruiter a few labels but does not explain which behaviours the study needs to compare.

• Housing type and room size

• Current air-purifier ownership

• Reason for considering the product

• Money left after essential costs

• Confidence using connected home devices

• Record when people notice air quality

• Capture when they run the device and which settings they use

• Track how noise, cleaning and electricity use affect continued use

• See whether the purifier becomes part of daily life

Use city tier only when product access, local services or the surrounding environment form part of the research question.

“People who shop for groceries in tier-one and tier-three cities” identifies locations but misses the household and shopping details that may explain behaviour.

• Local shopping and delivery options

• Household role and caring duties

• Work schedule

• Spending limits

• Mix of in-store and online shopping

• Ask for an entry each time a shopping need arises

• Add short scheduled check-ins to compare price, speed, trust and convenience

• Compare groups without assuming that city tier or age predicts digital behaviour

Use city tier when local shops, delivery coverage or service availability may change the experience.

A sample that looks balanced may still be unhelpful. Compare city tiers only when there is a clear reason to think local conditions may change the experience.

Ask directly about income, freedom to spend or financial limits. City tier may affect costs and opportunities, but it cannot tell you what resources one participant has.

State the birth years, then recruit people at the stages of life and with the experiences that matter to the product. Do not draw conclusions from a label that the study has not clearly defined.

App-based recruitment and complicated onboarding can favour people who are already confident with digital tools. If less-confident participants matter, test the study steps, offer suitable support and watch for people who fail to join.

A diary study can explain how and why behaviour differs across a carefully chosen group of participants. It cannot tell you how common those behaviours are across China unless the project also includes a suitable quantitative study.

Do not forget the participant groups after recruitment. Keep the details needed for comparison and connect every diary entry to the person, task, time and situation. Look at change within each participant as well as patterns across the group. A diary’s value often comes from seeing an early expectation turn into a routine, not just counting how many people mentioned a theme.

The research method should also fit the question. A diary study works well when behaviour changes over time, happens in different places, or follows events that are hard to predict. Interviews may be better when researchers need to ask follow-up questions straight away. Usability testing may be better when people need to complete the same controlled task. Recap’s guides to what a diary study is and when to use one explain these choices in more detail.

Useful participant groups still depend on reliable recruitment, well-organised evidence and active fieldwork. Recap supports diary studies in mainland China through managed participant sourcing and human screening, local fieldwork support, photo, video and other diary tasks, Chinese transcription and English translation, and analysis linked back to the original evidence.

The goal is not to replace researcher judgement with a formula. It is to explain why each group is in the sample, capture behaviour in the situations that matter, and keep enough participant detail to test the team’s assumptions.

Planning a diary study with participants in mainland China? Review Recap’s platform options or schedule a demo to discuss participants, fieldwork and evidence.