If you want to know how your customers feel about your business and what you can do to improve it, ask them using an ecommerce survey. There is no better group of subjects to get insights from than from people who patronize what you do and offer.
Carefully-crafted website survey questions are reliable, honest and quick ways to know what matters most to your customers. It tells you what’s lacking in your site, what areas of your site you should improve on and what’s getting in the way of your conversions. From your ad campaigns and new products to your site navigation and CTAs, these surveys can provide a wealth of marketing information you may not get from other marketing tools and data.
Get useful and immediate insights and know what’s going on in your customers’ minds with these nine best practices for conducting ecommerce surveys.
Clearly define your objective
Before conducting your survey, have a clearly defined objective in place. By beginning with the end in mind, you can be sure you will arrive at conclusive results.
The objective of your survey can take on many forms:
- to know how satisfied or dissatisfied your customers are
- to discover what they want or expect future products to be like
- or to know what you can do to improve their experience of your products and service
Be specific with the goals you have in mind. Avoid goals that can be answered with open-ended statements because answers may be all over the map. When responses are too broad, it will be challenging to translate them into actionable insights later on.
It also helps when you make your survey’s objective very explicit. When respondents realize that they are given the power to initiate change in your business, it can motivate them to come up with honest and meaningful responses. You may not even need to spend on incentives and freebies just to know what your customers’ needs and desires are if you are clear about your objective.
Remind yourself the objective of your survey too. It will help you stay focused on the website survey questions you need to ask and the answers you can expect to receive. Some responses you will receive from your survey may not be straightforward. When this happens, you need to analyze it and establish connections from other responses. If you can’t, then your survey will be pointless. Don’t lose your concentration in your survey’s goals. Otherwise, your customers will lose theirs too.
Make your survey accessible
Avoid having pop-ups for your ecommerce website questionnaire as 90% of consumers are not fond of it, according to a study by Ignite Feedback. It interrupts them from a task they are doing at the moment.
Since ecommerce surveys are addressed to customers anyway, a good location for it would be just below your order confirmation / thank you page. By this time, they have successfully checked out, done placing their orders and are only waiting for their items to be shipped.
Another way to send your survey is through a post-purchase email. Also by this time, your customers have already purchased from you and you want know what their experience was like with your service and product.
It is up to you to embed your survey in the email or link your customer to your survey page. How it is designed will also depend on the survey software you wish to choose, which we will discuss more later on.
While these two options are targeted for people who actually purchased, you can position your survey differently depending on how you qualify your other respondents. Will you be surveying random visitors on your site? Will it be people who are considering to buy from you? This helps you ask the right questions in the right location on your site.
To know whether a random visitor is on their way to actually buying something from you, check the average time on site and pageviews metrics on your Google Analytics data. People who are just above the average engagement are people who are likely to purchase.
Remember that regardless of the qualification of the people you want to survey, the accessibility of your survey is essential. An accessible survey means higher volume of responses.
Set the expectations
When sending out your ecommerce survey via email, keep the invitation to answer it short and sweet.
Make sure that your potential respondent knows what the purpose of the survey is, how long will it take them to answer and how many questions there will be. Laying down the expectations for your survey is essential to better engagement among your respondents.
If you are embedding the survey in an email, make sure the survey loads completely and respondents won’t experience any usability issues with it. In the same way, if you are sending your respondents to a different page to access your survey, have the link to it as noticeable and easy to click on as possible.
Be creative and engaging with your buttons too. Vary your CTAs to ‘Take our 2-minute survey’ or ‘Yes, take me to the survey!’ and see which copy people respond to more.
Use urgency triggers too in your survey invitation. Set fairly short deadlines for the submissions and follow up with reminders emails after that.
Ask clear and straightforward questions
Only 16% of web users do more than scan for keywords and sentences when reading information online, according to a study by Nielsen. This means if you are sending out an ecommerce survey, you have to keep it short with clear and straightforward questions. It’s ironic that the more questions you ask, the fewer respondents and replies you often get. Keep your ecommerce website questionnaire focused at all times.
Ultimately, the questions you will ask will depend on your survey’s objective. As there are no ‘right’ questions to ask, keep in mind the following reminders on what NOT to do when creating your ecommerce website questionnaire.
When using multiple choices, do not complicate the options. Remember that people only scan information they see online. Complicated multiple choices take time to read and require people to think more especially if the choices do not match with how they feel. Construct your choices in single words or short phrases.
Do not use words like ‘suggestions’, ‘complaints’ and ‘ideas’. Just provide a free-form ‘Any Comments’ text box where respondents can choose what they want to say. It can be complicated if you are requesting for specific things. It may reduce the effort and the value of that comment.
Avoid open-ended questions as much as possible. Similar to the previous point, it may seem like too much effort for some people to write their insights. Unless your respondents are super motivated either positively or negatively, not everyone will take time to fill your survey out as comprehensively as you want them to.
Do not have wordy questions. Wordy questionnaires can be boring and can lead to a lapse in the overall quality of responses. Keep questions condensed and stick to at least ten website survey questions. In the graph below, the shorter the survey is, respondents are less likely to abandon it.
Choose wisely the response scales to be used
Aside from the language and the logical order of your questions, the response scales you use also affects the answers you will get from your survey.
Response scales are ways of getting immediate insights from your respondents. There are three types of response scales.
Dichotomous response scales present only two choices. The choices can be true or false, yes or no, agree or disagree. There is practically zero chance for nuance and no neutrality is provided.
Rating scales often start with the introduction “On a scale of 1 to 5, how would you rate your …?” Rating scales can be three-point, five-point or seven-point scales. The wider the scale, the bigger the variance in the outcomes. Therefore, the 1 to 5 scale or the Likert Scale is widely recommended where 1 is negative and 5 is positive.
When using rating scales, always match the numerical rating with its value. Switching a response of 5 as negative and 1 as positive can be confusing for your respondents.
Semantic differential response scales measure attitudinal responses and are used mostly in specialist surveys to gather answers and interpret data based on connotative meanings. It uses clearly opposite words and can be marked or unmarked.
So, what response scale should you use? It all depends on the data you want to see.
Although dichotomous scales offer precise data results, it does not give any insight if you are improving satisfaction and experiences for your customers. A Likert Scale may be more recommended in measuring attitudes and behavior, but a smaller range is best to limit any sort of nuance in the answers.
Control your biases
A degree of bias always creeps in any survey process from the identification of objectives and website survey questions to data collection and analysis.
In a qualitative research such as an ecommerce survey, one of the most common biases you need to control is confirmation bias. This is the tendency to interpret information that confirms your preconceptions and ignore information that challenges it.
Another thing to control is the backfire effect. This is when your preconceptions are challenged with new information and you alter your opinions to incorporate the new info into your mind.
Be careful also of self-selection bias where respondents volunteer themselves to participate in your survey. This likely happens when surveys have incentives in return. This may fail to factor in real customer motivations in the feedback you will receive.
Always keep your biases in check and have the following tips in mind:
Avoid leading questions. Don’t use words that will sway your respondent to one side of the argument. Instead of asking “how soon were we able to get in touch with your concern?”, a better formulation is “how would you describe our customer service?” The use of the word ‘soon’ creates the idea that you are prompt when dealing with customers when you may not appear the same way to your customers. Always check the neutrality of the words you are using.
Randomize response options. Primacy or the tendency of respondents to select the first reasonable choice instead of reading all available options and selecting the best choice happens a lot in surveys. This form of bias is hard to detect and greatly affects how data is collected and interpreted.
Although randomizing options in your survey won’t fix the effects of primacy and how your respondents engage with your survey, this can guide you design better surveys and improve the quality of your data.
Work on your survey’s structure. How your survey questions are ordered and revealed, number of questions, logic, length, and the introduction and conclusion can influence your respondents’ engagement in your survey.
Designing your survey well not only includes the layout and formatting, but also studying how modifying each component of your survey will affect your respondents’ participation.
Be careful with your styling and coloring. Remember that people respond differently to colors and imagery. It can stimulate participation or induce fatigue.
With every color scheme, font, layout and logo that you wish to add to your survey, pretest it to see if it is leaning towards one demographic or not. At the very least, keep the style and color of your survey professional and inviting.
Use a reliable survey software
There are plenty of survey software available.
Surveymonkey allows anyone to quickly and easily create online surveys. You can choose between a free or a pro account.
Qualtrics also offers the ease of Surveymonkey. It comes with real-time reporting helping you make better business decisions.
SurveyGizmo has easy-to-use and flexible tools so you can create the kind of survey you have in mind. You can use its 40+ survey question types as a guide and its theme designer to make your surveys look professionally made.
Google Forms are great to use too as you can embed it directly on your website page or email or simply send the link to your respondents. You can customize it too with your photo or logo for branding purposes. Google Forms is free.
Wufoo is a general-purpose form builder with a cloud storage database. It is easy to customize and widely recommended for the flexibility of its form design.
You can also use some of the more expensive and powerful survey software like iPerceptions and Clicktools. Consider using the basic software first and push the limits of its benefits before moving on to these more sophisticated options.
Check out this link too for a more comprehensive list of survey software choices.
Optimize for mobile devices
As the fair portion of consumers today move from a mobile-first to a mobile-only outlook, it is important that your surveys are easy to fill out on a mobile device. When selecting your survey software, check if their surveys are responsive in smartphones and tablets.
By designing your ecommerce survey with mobile devices in mind, this can help you retain more respondents and and give you valuable insights especially if some of your website survey questions are related to mobile shopping experiences.
When optimizing your surveys, keep in mind that how it appears on a mobile screen will directly affect its completion rate. Check out these best practices for mobile surveys and apply them during the design phase:
- Set expectations and have the link or CTA button to your survey in a distinct part of your email.
- Use multiple choice question types. This way, your mobile respondents will only need to check the answer boxes instead of supplying the answers in a text format.
- Limit the number of choices so all options are available on the screen at the same time.
- Refrain from using long descriptions.
- If you are using response scales, have the positive options at the top and the negative options at the bottom to align with your respondents’ expectations.
- Avoid having free form ‘Any Comments’ boxes. Tiny screens can negatively impact survey completion rates on mobile.
- Consider having multiple short survey pages than one long survey page. Unlimited scrolling can affect the usability of your survey.
- Avoid having images on the survey. Depending on your respondent’s mobile download speed, your images may load slowly on their device.
Analyze and dig into your Analytics
Survey work is not over after data collection. In fact, analysis of survey data is where the actual work begins.
If your survey software of choice has a real-time reporting dashboard or analytics tool, use it to attach more meaning to your survey data. But more importantly, you should be able to dig into your Google Analytics account and do your work once you have your survey data with you.
Sometimes, data from Google Analytics may not be as straightforward as we expect it to be. The responses and comments we receive from our survey can guide us better as to where we should look in our analytics tool and see problems that need to be fixed.
Using survey data, identify the area or page in your site where there are reported problems. If you see a pattern, work it out immediately. Using your Google Analytics, know how many of your site users are affected by it. Compare it with relevant conversion rates from the people affected by the issue and those who are not.
Always calculate the percentage of your site users affected by a reported or a seemingly common issue. When there seems to be a deluge of negative comments, it can be disheartening for any marketer to accept it and easily believe that the issue is significant when in fact it isn’t.
While you have to stay on guard when it comes to negative data, ecommerce surveys are also filled with positive remarks from your respondents. Especially when you know that you have done your job in ensuring a fantastic shopping experience on your site and you take comments constructively and use it to improve the usability of your online store, the praises and affirmation you receive will be instrumental in identifying your strengths as a brand.
When done regularly and with both positive and negative feedbacks taken constructively, ecommerce surveys can help you build the kind of customer loyalty that generates buzz around your business and attracts other equally happy referrals.
As a recap, keep in mind these 9 best practices:
- Have a clearly define survey objective
- Ensure its accessibility
- Set the expectations
- Keep questions straightforward
- Choose the right response scales based on your survey objective
- Control your biases
- Use the survey software ideal to your objective
- Optimize your survey for mobile respondents
- And analyze your Analytics data
Do you regularly survey your customers? What other survey practices have been giving you amazing outcomes?