Our guide to good surveying for arts, culture and heritage organisations

Effective audience / visitor research is about much more than counting participants. It helps you understand who is engaging - and who is not - while providing valuable insight into audience experiences, motivations, and needs.

Here's our guide to doing it well.

First, consider the following questions:

• What decisions will this data help inform?
• What information do you actually need?
• What resources do you have?
• How will the data be collected?
• What questions need to be asked?

Taking time to think these through will help ensure that research remains focused, practical and is not too much of a drain on valuable time and energy. 

• You need measurable, quantifiable results
• You can reach a sufficient number of people
• You are able to gather responses from a sample that represents your wider audience
• There is no simpler or more efficient way of obtaining the information required.

Otherwise, alternative approaches such as interviews, focus groups or observations may produce more useful insights.

Once objectives are clear, you can begin practical planning. You'll want to consider:

How?
Options may include face-to-face interviews, online surveys, self-completion questionnaires, or a combination of methods. 

When?
Surveying can take place before, during, or after an event or programme. Some organisations may benefit from repeating surveys periodically or adopting a longitudinal approach to track changes over time.

Who?
The people collecting data also play a significant role in research quality. Staff and volunteers should be appropriately trained, understand the survey process, and apply methods consistently. 

Data management
Good planning should also address data management responsibilities, including compliance with GDPR, secure storage of information, and processes for sharing and analysing results. 

• Every question should serve a clear purpose. If information is not needed, it should not be collected. 
• Use plain language and avoid jargon. 
• Neutral, non-leading questions
• A logical question order that reflects the visitor journey. More personal or sensitive questions are usually best placed towards the end of the survey.
• All questions are answerable and provide opportunities for respondents to opt out of personal questions if they prefer. 
• Collecting full postcodes opens up opportunities for further analysis (eg Audience Spectrum segmentation)

Importantly, surveys should always be tested before being launched. 

You'll want to ensure that the people who participate in your research accurately reflect the wider population being studied (in this case the ‘population’ is all your visitors / audience).

A good sample should be large enough to provide confidence in the results while also reflecting the diversity of the audience. It should cover the full programme rather than focusing on a limited number of events or activities.
 

How many responses should you collect?

Two concepts are particularly important when determining sample size:

Confidence level refers to how certain we can be that survey findings reflect the wider population.

Margin of error indicates how much results may differ from the true population value.

A commonly accepted standard is a 95% confidence level with a margin of error of plus or minus 5%. So as a general rule, around 380 completed surveys are often sufficient to achieve this level of accuracy for many audience studies, regardless of population size.

For more on sampling, read our Guide > 

Research becomes skewed when some groups are overrepresented while others are excluded or underrepresented. This can happen for several reasons.

Limited staff availability may result in surveys being conducted only on certain days or at certain times. Research methods may unintentionally favour some visitors over others. Surveyors may also unconsciously approach people who appear friendly, approachable, or similar to themselves.

Several practical strategies can help minimise this risk:

• Collect data across a variety of days and times.
• Increase surveying activity during busier periods.
• Consider where surveying takes place – is it quiet? Is there somewhere to sit?
• Use random selection methods, such as approaching every second visitor.
• Employ face-to-face surveying where appropriate to reduce barriers to participation.
• Actively encourage a wide range of people to take part.

A good interviewer approaches participants confidently, explains the purpose of the research clearly, and obtains informed consent before beginning.

Consistency is essential; questions should be asked exactly as written, without alteration or interpretation. 
Interviewers should remain neutral, avoid expressing personal opinions, and reassure participants that their responses will remain confidential.

Participants should be given sufficient time to consider their answers, and interviewers should remain patient, polite, and appreciative throughout the interaction.

For more on recruiting, training and managing fieldworkers, read our Guide > 

Surveying is only valuable if the findings are analysed and used effectively.

Consider who will be reading the results and what decisions the findings are intended to support. Reports should provide context by explaining the methodology, data collection dates, sample size, and margin of error. Findings should be presented clearly and linked back to organisational objectives, targets, or key performance indicators wherever possible.

Visual communication can make results more engaging and accessible. Charts, maps, infographics, and summaries often help stakeholders understand findings more quickly than tables of data alone.

Most importantly, research findings should be revisited and used – don’t let your hard work languish in a drawer somewhere! 

• Webinar recording: Good data collection practice >
• Sample size calculator > ​
• GDPR guidance >​
• Market Research Society Code of Conduct > 

Survey design and analysis, fieldworker training, bespoke research and more. Get in touch to discuss your needs.