When offered the opportunity to provide a company with feedback, I’m usually happy to participate — but only if the process is easy. Like most people these days, I’m short on time (and attention), so companies are better off making their requests for feedback short and sweet.
But how can a company find the ideal survey length? If their survey is too long, participants may not start or complete it. If it’s too short, they’ll miss opportunities to gain insights into their customer base.![→ Free Download: 5 Customer Survey Templates [Access Now]](https://no-cache.hubspot.com/cta/default/53/9d36416b-3b0d-470c-a707-269296bb8683.png)
Today, I’m going to share everything I learned about how to create the ideal survey length while talking to experts in the field.
Here are the questions I’ll dive into in this article:
I went to the experts for insight on ideal survey length, survey question count, and more.
I’m pretty fanatical about customer centricity, and I wanted to know how survey designers create an experience that meets both the needs of the business but also serves the participant — and avoids survey fatigue. So, I had lots of questions that I posed to experts, which you’ll find below.
A quick reminder: Whether you’re getting ready to build your first survey or you’re looking for ways to make longer surveys more user-friendly, you’ll want to start with the foundation: your survey structure and design.
The structure of your survey will directly impact its rate of success, from the number of questions included to their level of complexity, each question contributes to the length of time it takes to complete the survey.
Use the tips below – and helpful templates like these — to get started.
Let’s dig in.
How many questions should a survey have?
The fewer questions, the better, in most cases. In general, shorter surveys have higher completion rates and better quality data.
If your survey has too many questions, you may run into survey “satisficing,” where respondents begin answering the questions without putting much effort or thought into them. Satisficing can happen for a number of reasons, like the respondent running out of effort or the task being too difficult.
One study showed that for every additional 10 questions added to a survey, the completion rate continued to drop. This makes it important for your team to really consider what needs to be asked versus what may just be a “nice to know” question.
I asked Ting Lai, voice of the customer program manager at AuditBoard, for his opinion on the ideal number of survey questions.
His answer? “It depends.”
Lai added some more insight for me, saying, “You have to consider your audience and where they’re at in the journey, as well as the type of customer interaction taking place.”
“Transactional surveys, like NPS or support surveys, tend to be two to three questions. We try to make this sort of experience as quick and painless as possible. For surveys sent at the end of a customer journey milestone … those tend to be a little bit longer, and a general best practice for those types of surveys is 7-10 questions.”
Pro tip: Place your top priority questions earlier on in the survey so that if you do see drop-off, you’re at least obtaining key information early on.
How long should a survey be?
So, what’s the magic number? Is there a magic number? While it depends on your industry and your desired outcome, the general consensus is that the ideal number of survey questions is between 1 and 10, especially if you’re looking to get a high completion rate.
But the number of questions is only one half of the equation, as you also have to consider how much time it takes to actually answer those questions (which I’ll discuss in the next section).
You could potentially include more questions in your survey if they’re easy to answer and can still be completed within a short timeframe (or if you’re incentivizing customers to participate.)
Important point: Lai also pointed out to me that thinking about survey length in terms of number of questions can be a bit misleading, because modern survey tools allow you to create differentiated experiences based on the previous answer.
For example, if you’re running a post-onboarding survey and your customer expresses dissatisfaction or frustration, you can send them down a path that offers follow-up questions for them to provide more context on their experience (or even create an escalation path).
So, while your overall survey may have 20 questions, each survey path might only be 7 to 10 since it’s a differentiated experience.
Pro tip: In my line of work, I run a three-question post-event survey. Only the first question is required (the rating for the session), and I clearly label the second and third questions as optional so that I don’t scare participants off from at least completing the first question. Not everyone loves answering open-ended questions, and that’s okay!
How long should it take to complete a survey?
While the number of questions on your survey is important, you also need to consider how long it will take for participants to answer those questions. People are busy, attention spans are shorter than ever, and survey fatigue is real.
So, is there an ideal survey time?
I asked Amy Maret, who used to lead the Research and Thought Leadership team here at HubSpot, what she thinks about the ideal survey.
“The shorter the survey, the better,” she says.
“We‘re always looking to be more selective about the questions we ask. A shorter survey pretty much always means better data quality. Fifteen minutes is about the maximum I would want for a market survey … If you’re surveying your own customers, you‘d want to go even shorter to make the best use of the time they’re willing to give you.”
The key point? People will take the time to complete your survey, but it’s best to keep it under 15 minutes.
Recent data even suggests that you might want to go shorter than that. According to Qualtrics data, “Surveys longer than 12 minutes (and 9 minutes on mobile) start to see substantial levels of respondent break-off.”
That being said, it’s worth noting that the length of your survey may vary based on the type of information you’re looking for.
For example, when I run NPS surveys, they only consist of one question and takes less than a minute to complete. In contrast, a research survey is likely going to be longer and take more time to complete.
As Mike Christopher, customer experience manager at AuditBoard pointed out to me recently, if your participants are engaged with your brand or have opted into the experience, you can likely get away with a slightly longer survey.
“When I’m building out a survey,” Mike says, “the length will depend on what the goal of the survey is.”
“If you’re engaging a specific curated panel, or running a research survey, I think you have a bit more leeway to make it longer because these people have opted into it. They’re more likely to accept a longer survey because they’re engaged and interested.”
Pro tip: Be transparent with participants about how long the survey will take. You can do this by including the approximate time to completion when you send out your survey. Software like Hubspot’s Customer Feedback Software provides an easy way to send surveys and collate customer insights.
How do you create surveys that feel customer-centric?
When building your survey, start with the needs of the team you’re working with. Be thoughtful about what questions really need to be asked and remember that more complex questions take more time to complete.
Additionally, different question types may require more time to complete than others. For example, multiple choice questions might take less time to complete than having a matrix or open-ended questions.
I asked Lai how he and his team go about prioritizing the customer with their surveys, and here’s what he said:
“When I came into this industry, I always felt like surveys were kind of lacking, because while you’re providing your opinion as a participant, it’s not necessarily helping you with your current experience. It felt very one-sided.”
Lai said that adding contextual questions led to better responses.
“If the customer has a negative response to a question, we can send them down a path that asks follow-up questions for us to better understand (and then later respond) to their issue. I believe surveys can be a way of building trust with your customers.”
In fact, research shows that personalization in surveys can result in higher engagement and more granular insights. You can do this by creating micro-segmentation or creating different branches and paths for customers based on their responses.
Lai also told me that he and his team will test out their surveys to get a feel for the participant experience. This includes testing it out at the point of interaction. For example, Lai might complete an in-app experience or create a support ticket, then take the resulting survey that follows.
I personally really like this approach to testing as it feels really customer-centric and allows you to ask yourself, “Would I fill this survey out if it were offered to me?”
Pro tips:
- Make sure you’re thinking about your survey participants (and not just your company) when you’re building out a survey.
- Find ways to make sure your participant feels heard and supported and create an experience that feels more collaborative than one-sided.
- Stay away from repetitive questions, test the survey for technical issues, and be mindful of privacy concerns (e.g., too much personal information being requested).
How do you find the right survey length?
I asked Maxwell Iskiev, a senior market research analyst for the HubSpot blog, how he finds the ideal survey length.
“The ideal length of a survey depends on the purpose behind it,” he shares. “For example, my surveys on broad topics like Marketing Trends will be longer, covering many aspects of the industry.”
“From there, I might notice interesting themes to follow up on, like data privacy, or the impact of the recession on marketers. I’ll then write shorter, more focused surveys around 5-10 questions diving deeper into a specific topic of interest.”
Maret seconds the idea of shorter, deep-dive surveys:
“[Keeping surveys short] might mean breaking up what could be a longer survey into multiple, smaller, sequential surveys with the same audience.”
“An added benefit to this approach is that it allows you to tailor your follow-up surveys based on what you learn from the previous ones, and pivot more quickly if priorities change — which they often do.”
Here’s what this means for you:
- Start with the purpose of your survey, then build your questions.
- If your topic is straightforward and specific, keep it simple. More comprehensive subjects will likely require slightly longer surveys or a sequence of questionnaires.
- If your survey is heavy on open-ended questions, less is more. If there are a majority of multiple choice or rating scale questions, you can probably get away with some extras.
Pro tip: Don’t forget to measure and improve. Track the survey completion rate and make adjustments if it’s lower than what you want it to be.
Start With the End in Mind
In addition to the length of the survey and number of questions, it’s important to start with your end goal in mind.
As Christopher told me, “You need to understand the goal of your survey and then work backwards from there. Identify the point of your survey, the goals you’re trying to achieve, and make sure you understand your audience. Then, you can decide what questions to ask.”
Survey length matters, but there’s no one-size-fits-all rule for how long a survey should be. The length of your survey will ultimately depend on survey type, question format, and survey frequency.
Based on my research, I’d recommend using a 10/10 rule:
Create a survey that has 10 questions or fewer and takes no more than 10 minutes to complete.
When in doubt, put yourself in your customers‘ shoes. If you wouldn’t spend the time to take your own survey, chances are they won’t either.
Editors’ note: This article was originally published in August 2018 and has since been updated for comprehensiveness.
