Methodology

How Cleria produces defensible brand-tracking data.

How Cleria produces defensible brand-tracking data: recruitment, quotas, quality checks, representativeness, sample sizes, error margins, and AI data use.

How Cleria fields research

Quality control starts before the first answer.

Each study starts with the population being studied, then applies recruitment, screening, quotas, category lockout rules, and quality checks around that design.

Sampling & recruitment

Cleria uses its global panel and vetted recruitment sources, with quotas, screening, and quality checks on every study.

Cleria runs brand tracking worldwide, with surveys localized into each market's language — translation is handled for you.
Recruitment is designed around the market, category, and study design rather than a one-size-fits-all respondent pool.
Where vetted recruitment sources are used, respondents still go through Cleria screening and quality checks before responses are accepted.
Samples are designed to represent the population being studied, whether that is the general population or a defined target group.

Category lockout

Inside the Cleria panel, respondents are prevented from answering other surveys in the same category for 12 months.

The lockout reduces awareness creep from repeatedly asking the same people about the same category.
Because Cleria operates its own global panel, it can enforce this category lockout inside the panel.
When outside vetted sources are used, Cleria still applies screening and quality controls before accepting responses.

Fraud and respondent quality

Cleria combines respondent history with security questions on every survey to detect fraud, inattentive responses, and inconsistent answers.

Respondent history helps spot patterns that one-off recruitment models may miss.
Security and control questions run on every survey before responses are accepted into reporting.
You act on data screened for fraud and inattentive responses, not raw panel output.

Representativeness

Samples are designed to be representative of the population being studied, not always the general population by default.

The population may be the general population or a specific target group defined for the client.
Quotas guide sample composition against the study design.
Reports should make clear when weighting or other methodological choices shape the final read.

Sample sizes and error margins

Every wave starts at 2,000 respondents and can scale up to 50,000 respondents in most markets when deeper reads are needed.

The 2,000 respondent minimum is designed to keep results reliable and margins of error usable.
Dashboards and reports show sample sizes and error margins alongside results.
Larger samples help support deeper market, segment, or competitor reads when the study design requires them.

AI data policy

Your data stays private to your team. Client-specific data is not shared with other clients or used to train AI models.

The assistant helps teams explore their own Cleria data inside the product experience.
Client-specific data remains private to the client team.
The assistant works only within your own Cleria data, never across other clients.

What every report should show

A number needs context before it becomes insight.

Cleria reporting makes the study design visible enough to understand the read, the filters, and the level of confidence behind each result.

Sample size

Show the respondent count behind each wave or view, including the population being studied.

Fieldwork dates

Make clear when the data was collected, especially for fast-moving categories or campaigns.

Market and category

Define the geography, category frame, and target population before interpreting the result.

Competitor set

List the brands or competitors included so benchmarks are read in the right context.

Filters and segments

Show the filters used for each read, including demographic, socioeconomic, market, or custom cuts.

Error margins

Show error margins alongside results so teams know how much confidence to place in each read.

Quotas and weighting

Document quotas and note when weighting or other methodological choices shape the final read.

Wave & cadence

Note which wave a read comes from and how the study repeats over time, so trends are read in context.

AI and your data

The assistant works inside your Cleria data.

The assistant makes the dashboard easier to explore, without changing the privacy boundary around your data.

Your data stays private to your team.

Cleria does not share client-specific data with other clients.

Specific client data is not used to train AI models.

Wondering what this looks like?

See how we show your brand tracking data in your dashboard.

Explore the Brand Tracker 360