Surveys are one of the best ways to useful data about your brand. Other techniques can illuminate your brand performance, of course. Things like social listening, customer service feedback, and sales data all offer different indicators of how your brand’s performing in the eyes of your target audience. However, they don’t offer the same reliability as brand surveys.
Brand surveys should be a significant part of your customer insight-gathering process. Talking to your customers directly is one of the most powerful ways to understand the health of your brand, and figure out how to take it to the next level.
Below, we’ll go through what insights you might expect from brand surveys, how you can get the most out of them, and how to give them the best chance of being answered.
What insights can you get from brand surveys?
You can gain a surprising amount of insight about your brand from surveys.
One of the most important metrics for checking your brand health is brand recognition. This is simply the number of respondents who say they’ve already heard of your brand when asked. This is a great indicator of how effective your campaigns are, as well as word of mouth and other growth channels.
Brand recall is an extension of this: it’s the same question, but unaided. The respondent is asked “which brand comes to mind when you think of…” without your brand given as a prompt. If you’re aiming to become a market leader in your chosen industry, this is the metric you want top marks in.
Insights in these two areas come from surveys sent by third parties, whether that’s a brand tracking platform or a market research organization you’ve hired for the task.
When sending surveys directly, there’s a different set of brand insights you can gain. You’ll be able to ask, for example, which associations customers have with your brand, and what feelings your brand evokes. You’ll be able to find out how congruent your latest campaign is with your perceived values, or how much trust customers have in you. You could also find out how many of your customers are willing to recommend you to a friend.
There’s quite a lot of space for creativity, and if your survey experience is well-designed, enjoyable, and respects your respondents’ time, you’re sure to get some good results.
How to ask the best questions in brand surveys
Bad data will cause you to make bad strategic calls, so it’s crucial to get it right the first time. How? Ask the right questions of the right people.
Segment your respondents
You’re not going to have much luck asking teenagers their opinions of life insurance plans, or grandparents which skateboard brands they’ve heard of (there might be some exceptions that pleasantly surprise you, but probably not on a large scale). You have to ask the right people the right questions – and that’s where audience segmentation comes into play.
Chances are you’ve already done some segmentation work if you’re responsible for a brand. You’ll have a good idea of who your product is aimed at, what messaging is appropriate for which demographic, and so on. But remember to apply this approach to your brand surveys, too. While it might take a little more effort in survey preparation, it means you won’t waste time asking irrelevant questions or even offending people with incorrect assumptions.
Any survey software worth its salt will have options for basic segmentation and customer grouping – make sure you use them.
Ask the right type of questions
This is a crucial but easily-overlooked part of survey design. And if you get it wrong, you’ll either get much fewer answers than you wanted or answers filled with misleading data. So it’s important to make surveys with your respondents’ experiences in mind.
Open-ended questions are great for certain situations. If you’re looking for ‘unaided’ opinions without respondents being influenced by your suggestions, these are the ones to use. With these, you want to use simple language, and if you want more detail, be more specific. So if you want to know what feelings your service evokes in customers, you can ask that, but if you want to know how someone feels directly after using your product, you should be specific enough to find out exactly that.
Respondents aren’t mind-readers – they won’t know what angle to take unless you do a good job in guiding them.
If you think customers might be tempted to give a really long answer, you might want to reconsider your approach. Can it be broken down into multiple questions? This might make things easier for the respondent if they don’t have to type out an essay for their answer.
In contrast, closed questions are those that have a set of predefined answers that respondents can choose from. This might be a simple ‘yes or no’ choice, or it could be a scale of answers rating something from 1 to 5. It could be a list of checkboxes, where respondents can select as many or as few as they want.
These are usually the easiest questions to answer because they require minimal effort: respondents only have to choose an answer instead of formulating their own.
Have a quality-check process
It’s also helpful to sense-check your survey questions before sending them out. Run them by colleagues or a small subset of customers before fully committing to them to see if they can be misinterpreted in any way. It’s easy for a brand manager to get tunnel vision and forget that different perspectives are available, especially across cultures, geographies, and other segments. Questions have to be clear and make sense, so get them checked first.
As well as this, don’t ask too many questions. Each additional question increases the workload for your respondents and will increase their likelihood of abandoning the survey. A high abandonment rate can really mess up your datasets, so if you want a nice set of complete answers to mine insights from, keep it short.
How to Get the Best Insights From Your Brand Surveys
Have a strategy
Running surveys should be a planned part of your insight strategy. Being reactive and throwing out some questions whenever you feel the need can be a viable solution in some circumstances. High-growth startups can thrive in chaos, and if you’re the type of company that iterates quickly and pivots on the fly, then ad-hoc monthly surveys could work for you.
But other organizations don’t work this way. Gathering new consumer data might work better on a quarterly or 6-monthly basis, where you’re better equipped to see trends in behaviors and attitudes that are more than just a temporary blip on a chart.
This is especially fitting if you run marketing campaigns that last months at a time and want to judge their impact on your brand health.
Make things analyzable
If you’re one of those companies whose marketing efforts operate on a longer timescale, there’s one simple way to make your insights more effective: ask the same questions each time.
Designing new surveys might seem a good idea in a rapidly-changing world, but data isn’t as valuable if you can’t see trends over time. From a practical perspective, it’s much easier to analyze large datasets if they’re consistent each time you send out a survey.
If you’re surveying customers on a larger scale, sifting through them manually for insights isn’t too practical. It can’t always be automated: the human perspective is important, and as a marketing manager or CMO you do need to spend some time reading through open-ended answers to get a feel for general opinion.
Sometimes, a simple CTRL-F command will do the job. If you notice a trend while scanning through answers, why not search your dataset manually and see what you can find? If there are particular words popping up, note them down alongside synonyms.
While this technique won’t give you solid data to make definitive plans, it might spark a creative idea or two that starts you on the path of an exciting new project.
But overall, you do need to keep analysis in mind when designing your surveys. Multiple-choice questions with a smaller selection of answers are easier to pick immediate trends from. Open questions can still be analyzed on a large scale in certain circumstances: sentiment analysis software can look at phrases and the tone of responses to gauge broad feelings (negative, favorable, indifferent) or associations (fun, helpful, corporate).
Whatever questions you choose, remember to consider how you’re going to analyze the answers when they eventually come flooding in.
Use mobile surveys to reach people on the move
Most modern survey software is optimized for mobile devices. These days, users can answer surveys on phones, tablets, and even e-readers. The advantage is, of course, the ubiquity of these gadgets: almost all of your customers will have one nearby at all times.
Mobile surveys aren’t too difficult to answer, especially when it comes to multiple-choice. A simple swipe and a tap are enough to do the job. That said, some people don’t like typing much on mobile devices, so it’s best not to ask open questions if you can avoid it.
If you’re using pre-built software to run your surveys, they should already be optimized for mobile. But if you’re building your own (or visually customizing existing solutions), there are some things you need to keep in mind to make things as easy as possible for your respondents:
- Keep things short and to the point. Less screen space means readers don’t want to see huge blocks of text – they’re used to skimming on mobile, not intense reading.
- Don’t offer too many options. Even though multiple-choice makes it easier, asking readers to choose through nine or ten answers slows down their response times and makes it harder.
- Design with the user in mind. You have to go by general web accessibility practices, as well as sensible visual design: readable fonts, standard colors, and an intuitive layout all make answering easier.
Without these fundamentals in place, you risk respondents abandoning the process, and you’ll miss out on some crucial insights you might not get elsewhere. In fact, mobile users are 5x more likely to abandon whatever they’re doing if the app or site they’re interacting with isn’t optimized for mobile. So it’s certainly something worth getting right the first time.