IVR surveys are a set of pre-recorded automated questions customers can answer via voice or phone keypads.
For example, they can provide businesses with an automated way of collecting valuable information, and they can also be good for eliminating response bias from customers who might otherwise hesitate to leave completely transparent feedback when talking to a live service agent.
In other words, IVR surveys can get you highly accurate information—if done right, that is.
8 IVR Survey Tips For Getting Meaningful Data
1. Stick to a fifth-grade reading level
Your survey should include simple words everyone can understand. If not, customers may think your survey is too difficult and cause them to disengage.
It’s best to aim for a fifth-grade reading level. Avoid industry-specific jargon, and ask questions that are short and precise so they can be as easy as possible to answer.
For instance, “Was our service agent kind and helpful, and did they go above and beyond?” wouldn’t work. Maybe the agent was helpful, but the question is asking too much for the customer to provide a solid answer.
The simpler and more precise questions you ask, the more straightforward answers you get.
2. Offer clear instructions
Some of your customers may not be familiar with how IVR surveys work. You have to be as clear as day when explaining to participants how to answer your questions.
For instance, make sure to offer instructions like, “You can speak freely after the beep. When you’re finished, please press zero to submit your response” after asking a question. Leave no room for guesswork to avoid instilling confusion among participants.
This is especially important for open-ended questions. Some participants may not be aware that they can provide feedback via voice, so clarifying that is critical. Also, be sure to remind respondents they have the option to repeat a question or instruction by pressing a specific number on the keypad.
3. Use a five-point scale
In short, a five-point scale gives customers the option to pick between five distinct answers. Although simple “Yes” or “No” answers can also do the trick, they often do not provide precise data.
For example, if you want to ask customers whether they’re satisfied with your customer experience, “Yes” or “No” answers won’t reveal exactly how satisfied they actually are.
Letting customers choose between multiple options will provide you with actionable data. A five-point scale strikes the sweet spot between giving customers a good variety of specific answers and not overwhelming them with too many options.
4. Label your responses properly
Make sure to label your responses as clearly as possible. In addition to five-point scales, consider answers like “Poor,” “Acceptable,” “Good,” “Very good,” and “Excellent” rather than asking customers to rate your services with numbers.
A three-out-of-five score may have different meanings depending on your respondents, but “Good” means the same thing to everyone. Clear and descriptive labels are crucial for gathering accurate data.
Also, avoid labels like “Incredibly bad” or “Outstandingly good.” They sound extreme and biased, which can influence the participant’s answers and lead to imprecise information.
5. Organize your questions accordingly
Don’t start with difficult questions right off the bat. These can be intimidating to start out with, so respondents may feel overwhelmed and drop the call before finishing.
It’s best to begin with basic inquiries and gradually ease customers into providing more complex answers to open-ended questions as they progress through the survey.
Also, make sure your inquiries flow logically. You should present questions related to one another in a natural order. This establishes continuity and helps respondents provide answers in quick succession—the faster participants go through your survey, the more likely they are to complete it.
6. Don’t drag out your survey
The longer your survey is, the more participants will drop out at some point. It’s best to keep your survey short—maybe five minutes maximum. Make sure to let participants know exactly how long the survey will take right away. This lets them set realistic expectations and prevent them from dropping off.
Additionally, consider the number of questions you’ll include within your survey’s time frame.
If you ask too many questions too fast, you’ll likely rush participants through the survey and cause them to leave inaccurate answers. If you ask too few questions too slowly, respondents will get bored, and you won’t get enough data to work with.
It’s best to write a survey script that includes both questions and answers and then set a time goal, like three minutes. You can then test the script and read it out loud against the clock to see whether you need to cut some questions out or if you still have room for more.
7. Consider branching logic
In short, branching is a condition-based survey design technique that allows you to divide your questionnaire into multiple paths—meaning you can ask respondents different questions based on their previous responses.
This is an excellent way to keep respondents engaged by asking personalized questions while skipping inquiries they may not find relevant. It’s also perfect for gathering more precise data from different customer segments.
However, just like the other tips, keep your branching short and simple. Going overboard with survey branches will eventually get respondents disengaged and cause them to drop off.
If you’re new to survey branches, you can easily set one up with tools like Voxco, for instance.
8. Test with a small audience
When you’ve got everything ready to roll, start by offering your survey to a small customer segment first. This helps identify potential survey difficulties without affecting a large portion of your customer base. It’s also useful for setting up benchmarks for further optimization.
That said, keep an eye on call metrics like survey response, completion, and abandonment rates. High abandonment rates may indicate your survey is too long or boring.
Meanwhile, a low response rate signals you should experiment with different send times. Try to identify the times of day when your customers are most likely to take the call.
Also, keep an eye out for the average survey completion time. If you set a goal for your script, compare that with your average completion time once your survey rolls out.
If customers completed the survey faster than you anticipated, you can maybe sneak in a few extra questions. But again, your survey shouldn’t take more than five minutes for optimal completion rates.
A well-thought-out IVR survey can do wonders for your company—your customer service agents can be freed up from repetitive tasks, and you get vital insights into how to make your customers happier.
If you want to set one up, the aforementioned Voxco is a great tool dedicated to collecting customer data.
However, if you wish for a more integrated solution, you can turn to some of the top VoIP providers like Ooma or RingCentral. These are excellent for setting up fully-fledged IVR systems that cover anything from surveys to CRM integrations, skill-based call routing, and more.