How to Create a Successful Survey

Running a survey is one of the most accessible and immediate forms of research available to cultural organisations. They can be incredibly powerful, giving you data from both current and potential audience members. But, a great survey requires a surprising amount of thought to get right.

We’ve been advising on and running surveys for cultural organisations since 2004, so here is a deeper look into what makes surveys successful, and what can hold you back from getting those useful responses and insight.

A survey isn’t always the answer

Before you begin, it’s always good to ask yourself if a survey is the best way to find out what you need. Surveys work well when you have a clear idea of what information you need. For example, you want to measure audience reaction to a specific event, or collect marketing preferences from potential audience members. If you want to explore a topic in-depth and you’re not sure what you might find out, then some focus-groups could be the way to go, instead of a survey.

Before designing your survey look at some of the information you might already have. Have you done surveys in the past? If so, have these been properly analysed, or are they sitting in a cupboard somewhere? Take a look at any box office data or social media insights on your audiences and see how these can inform and complement your survey.

Figure out what you want your survey to do

Ask most people what the first step is in designing a survey, and brainstorming questions will be the answer. In reality, this is skipping ahead a bit too queasily (admittedly, to the most fun part).

Too often, surveys are sent out for the sake of doing a survey, to have ‘something’ to give to funders, and to pay lip service to listening to your audiences. This is why it’s so important to align your survey with your objectives. Why are you doing the survey? What do you hope to gain?

By linking your survey questions to your objectives, you can tell a clear story about the impact your event, festival, or cultural venue had on your audience. Think about how each survey question addresses each of your objectives. It helps to have a basic analysis plan (e.g., objective 1 relates to questions 1, 2, and 3).

Define the audience for your survey

Pinpointing your objectives will naturally lead to defining a target audience for your survey. Your audience might be determined by demographics (young people in the city suburbs), by their relationship to you (season ticket holders, one-off attenders), or a combination of both.

To save time later on, it’s a good idea to think about how you’re going to look at the data after the survey is completed. Are you interested in looking at the results by different groups of people? If so, make sure to include a question that identifies them. (e.g., have a question with age ranges, ask if they’re a returning customer, or ask for a postcode if you want to segment the data by how far they’re travelling.)

Designing Your Survey Questions

They key to asking good questions is to make sure they’re relevant. For every question you add – ask yourself what you’re going to do with the resulting answers?

Try to make the survey experience as easy as possible. Keep language clear and simple, and avoid questions that are too big in scope e.g. “who should our audiences be?” Keep your survey as short as you can - nobody likes a 30 minute survey.

Avoid asking leading questions like “tell us why you loved X”. You’ll end up with mostly positive answers which won’t help you make useful changes.

If you’re asking sensitive and personal questions, be sure to explain why you’re asking them and what you’ll do with the resulting data. We’ve seen some dodgy questions over the years. For example, an arts venue asking about prior convictions, and surveys asking about sexual orientation when the organisation had no plans to do anything with this data.

Carrying Out the Survey

Comment cards are good for smaller organisations to hear what audiences thought about your events but you’ll need to factor in the extra time you’ll spend inputting the answers to analyse them.

In-person surveys are usually run by staff and volunteers, however that’s very time-intensive and requires training.

Running surveys online using sites like Survey Monkey works well for longer or more complex surveys, and is great for when you need results quickly.

At one event series we evaluated, we provided free photo booths that attendees could use and choose then to get their photos emailed to them. This was a fun way to collect the email addresses for the post-event survey.

Above all, keep it in context. Don’t do in-person surveys at the end of an event with alcohol, instead, send a link online after the event to bookers. If you’re unsure about which type of survey to run, you can always contact us for advice.

How many survey responses do you need?

To calculate how many survey responses you need, we recommend using a sample size calculator such as this one from Surveymonkey: www.surveymonkey.com/mp/sample-size-calculator/.

You just enter a few figures: first, enter your population size – this is the total number of people you want your survey to represent. If you’re conducting a survey among festival attendees, you can enter in the total number of tickets sold for your festival. Or, if you’re conducting a survey among people who have attended your events in 2018, enter your total number of attendees in 2018.

Next, we recommend keeping the next two numbers set at 95% confidence level and a 5% margin of error – these are both industry standard figures, and ensure that your data will be as statistically robust as possible. Once you enter in the last number, Surveymonkey will generate a target number of surveys you should aim for. So one example would be if you had 1000 attendees at your event, and you enter in a 95% confidence level and a 5% margin of error, you will need to get 278 responses. See? Easy peasy!

Case Study: Getting a good response rate

In 2017, we worked with EastSide Arts Festival to improve their post-festival response rate by over 500%, just by making a few tweaks. Here are some tried and tested tips for getting people to fill out an online survey.

• Offer a prize – a £40 Amazon voucher usually works well.
• Tell your audiences why you’re doing the survey – this will make them feel better about taking part and helping you out.
• Optimise the time of day you send out survey links.
• Ask others to share your survey on social and through their newsletters.
• Use the link customisation feature in Survey Monkey to track where you’re getting your survey responses from.

Remember...

Surveys can be as simple or as complex as you need them to be. As long as you keep them focused on your objectives, and tailored to your audience, you’ll massively increase the usefulness of your research.

Get Some Help with Your Survey

If you need some more personalised help, we can design your survey, help get it out there, and then analyse the responses. We've run surveys for Earagail Arts festival, the John Hewitt Society, and C.S. Lewis Festival. Get in touch to find out more.