So you’re thinking about running an employee engagement survey? In that case, you probably already understand the importance of employee engagement, and have set out with two clear aims:

  1. To accurately measure employee engagement levels in your organisation
  2. To learn what’s impacting employee engagement so you can improve these areas of your business

An employee engagement survey is the best place to start, however the quality of your results will depend heavily on the questions that you ask, and how you ask them. Without a robust methodology, your results are likely to be unreliable and difficult to turn into meaningful action.

In this guide, we’ll explain everything you need to know in order to design an effective employee engagement survey. We’ll outline a reliable framework for measuring and understanding engagement, while highlighting some common question-writing pitfalls that you’ll definitely want to avoid.


Measuring employee engagement using eNPS

When measuring something as important as employee engagement, it’s best to use an established methodology. Fortunately a simple and reliable metric already exists for providing a snapshot of engagement in your organisation.

The Employee Net Promoter Score (eNPS) has been adopted by businesses around the world such as Apple, Rackspace and Verizon, and is based on the concept of Net Promoter Score® (NPS®).

Net Promoter Score® was developed in 2003 by Bain & Co. in order to measure customer satisfaction and loyalty. It works by asking asking the consumer a single question on a 0-10 scale: “How likely is it that you would recommend [this product/service] to a friend or colleague?”.

eNPS shifts the focus to employee engagement by asking:

”How likely are you to recommend [this organisation] as a good place to work?”

This wording encourages respondents to consider the various factors that influence their workplace experience, and the score they provide is an accurate reflection of their overall engagement level.

In recent years, the eNPS methodology has been expanded upon to also measure the outcomes of employee engagement – loyalty, belief and satisfaction – by asking three additional questions.


If you were offered the same job at another organisation, how likely is it you would stay at [this organisation]?


How likely is it you would recommend [this organisation]’s products or services to friends and family?


Overall, how satisfied are you working at [this organisation]?

It is up to you whether you choose to use the eNPS outcome questions in your survey alongside the main eNPS question. If you do, these responses must also be taken into account when calculating your company’s employee engagement score.

Calculating your organisation’s engagement score

An employee engagement score is commonly calculated on one of two different scales. The first is the -100 to 100 scale traditionally used with the NPS® methodology. The second is the 0-10 scale used when you ask the eNPS questions.

The -100 to 100 NPS® scale is primarily implemented by businesses that are already reporting other metrics – such as customer satisfaction – using the same methodology. By keeping the scale consistent across multiple applications, it is easy to compare the health of different organisational KPIs.

Other businesses find that a 0-10 scale is more intuitive and better suited to their needs. Both methods are equally valid, and there’s no reason why you can’t use a combination of the two in your organisation, each for different purposes or with different stakeholders.

How to use the NPS® scale

The first step in calculating your engagement score on the NPS® scale is to divide the responses from all your eNPS questions into three categories:

Promoters: score 9-10

These are the respondents who gave the highest scores. They are considered to be highly engaged and are loyal advocates of your organisation.

Passives: score 7-8

While these employees don’t feel overly negative about their workplace experience, they aren’t especially engaged either.

Detractors: score 0-6

These are the people who gave the lowest scores to your eNPS questions. These employees are actively disengaged.

Once you have your three buckets of responses, the next step is to calculate the percentage of overall responses within each of them. An example could be:

Promoters (52%), Passives (28%), Detractors (20%)

Finally, you calculate your organisation’s engagement score by subtracting the percentage of Detractors from the percentage of Promoters. In this case it would be 52-20 = 32.

The NPS® scale ranges from 100 (a perfect score with 100% Promoters) to -100 (a score with 100% Detractors). Anything above 0 is considered to be good, above 25 is very good, and above 50 is exceptional.

How to use a 0-10 scale

While the NPS® scale may be a little complex in its construction, calculating your organisation’s employee engagement score on a 0-10 scale couldn’t be simpler.

By using the eNPS question framework, you’ve already invited responses to your questions on a scale of 0-10. The next step is to find the average (mean). To do this, you simply sum all your responses together, and divide by the total number of responses given.

Understanding the factors that influence employee engagement

Using eNPS will provide a reliable snapshot of employee engagement levels in your organisation. However, in order to act on these findings and improve your future score, you need to understand what’s affecting engagement in your company. That’s why it’s important to ask additional questions.

Your aim here is to ask about each aspect of an employee’s workplace experience so that you can build a picture of what’s “driving” engagement in your business. For example, employee workload, your office environment, or the level of manager support that your employees are receiving, will all influence how engaged your people feel.

Once you’ve gathered this information, you will be able to analyse your results and identify the root causes of disengagement among your teams – as well as areas in which your business is performing well.

Why you should avoid asking overly-specific questions

When it comes to identifying the causes of engagement and disengagement, many businesses make a common mistake; the biggest temptation is to ask focused questions that are specific to issues or circumstances that you might already know about.

You may believe that this will give you a headstart in honing into problem areas, but in reality overly-specific, targeted questions are bad for two reasons:

1. If you look for a problem, you will find a problem – even if it doesn’t exist.

Questions about specific events or circumstances can be “leading” in that they elicit a response from people even if they had no strong feelings about the issue, nor any inclination to raise them in the first place.

For example, you may feel that a recent merger or acquisition may be influencing engagement and morale so you ask the question “How smoothly do you think management handled our recent merger?”

This merger may have been a terribly run process and employee responses may reflect that. However, there is not necessarily any proof that the merger process itself had a negative impact on employee engagement levels.

In fact, it is the biased assumption of the person asking the question – who feels that the merger was a cause of disengagement and therefore wants to measure it – that leads to the link between engagement and the merger being drawn.

2. Specific questions aren’t benchmarkable

The other downside to overly-engineered questions is that they cannot be benchmarked, neither internally nor externally.

In the case of external benchmarking, the issue is obvious. An employee engagement survey in which all the questions are focused on the circumstances of your organisation won’t give you the opportunity to compare your performance to your industry peers.

For example, if your survey consists of questions such as “How do you feel about our latest management training sessions?” or “How do you feel about our performance in this year’s industry awards?” you will struggle to find any external data to compare your results to. This will leave you unable to tell whether the responses you receive are good or bad compared to an industry standard.

Internally, specific questions risk being more relevant to particular teams, offices or employee groups than others. Asking a question about satisfaction with commission will only be relevant to your sales teams, and won’t give you a reliable way to compare how all your departments feel about pay. Likewise, a question about a team-building retreat for your London employees might not be comparable to any activities you’ve done with your teams in other offices.

Ultimately, if your aim is to understand how events, circumstances and organisational changes are impacting employee engagement, it’s best to avoid asking questions about these changes directly – this can lead to biased conclusions and a lack of benchmarking.

Instead, the best way to uncover the information you are after is to use a consistent and universal framework, and to observe how your employees’ responses change in accordance to these events and circumstances.

So how do you build a consistent and universal framework? The answer is to ask employee engagement questions that are universally relevant to all employees – regardless of industry, organisation, role, seniority or location.

How to use an engagement driver framework

Fortunately, employee engagement isn’t a mysterious, poorly-understood field – in fact, there is a mountain of research into the subject from behavioural psychologists and management theorists.

This research has helped to identify the fundamental psychological factors that impact human motivation in the workplace. We refer to these as the “drivers” of employee engagement.

The methodology we use at Peakon uses 14 drivers based on these psychological requirements. These are Accomplishment, Autonomy, Environment, Freedom of Opinions, Goal-Setting, Growth, Manager Support, Meaningful Work, Organisational Fit, Peer Relationships, Recognition, Reward, Strategy, and Workload.

Let’s take Autonomy as an example. Research shows that when employees have the freedom to decide how they approach a task, they are more engaged and more motivated to complete it (Hackman & Oldham 1976, Ryan & Deci 1985). Therefore, it’s important for you to include a question that enables you to measure the level of autonomy your employees feel they have.

To measure Autonomy, we ask respondents to give a 0-10 score to the question:

“I feel like I am given enough freedom to decide how to do my work.”

We’ll discuss the wording of this question in more detail in the next section.

Basing your questions on research and proven psychological theories like this will help you to understand what is really influencing employee engagement in your business. Without a robust framework, it’s too easy for hunches and individual bias to influence the type of questions that you ask, which leads to unreliable findings.

If you’d like to learn more about the 14 drivers of employee engagement, we’ve put together a free 62-page ebook called The Psychology of Employee Engagement. This will provide plenty of insight into the engagement drivers, and will help inspire your own engagement driver questions.

However, we’re very proud of the work that our Organisational Development Science team has done in creating our full set of engagement survey questions, and we truly believe that it is the most robust engagement methodology available. By using the same question library as hundreds of other organisations worldwide, you also give yourself the ability to accurately benchmark your results against your peers. You can find all 46 Peakon engagement survey questions in the platform when you sign up for a free trial.


Now that we have established the importance of using a structured methodology to measure engagement, it’s time to start thinking about how our questions should be written to ensure useful and reliable data.

Let’s look again at the Peakon Autonomy question from before:

“I feel like I am given enough freedom to decide how to do my work.”

The wording of this question has been carefully considered. In fact, when it comes to writing your engagement questions, there are three different aspects you need to keep in mind:

  • Quantitative data: How do you ask a question so you receive good numerical data?
  • Qualitative data: How do you ask a question so that you invite additional contextual feedback?
  • Question phrasing: How do you avoid compromising your results with bias or ambiguous questions?

In this section we’ll examine each of these aspects and show you how to write your questions in an effective way.

Gathering quantitative data with consistent scales

Quantitative data will form the bedrock of your survey analysis. It will give you an easy way to compare engagement in teams, departments, or offices, so that you can identify areas of the business that require improvement, or sources of best practice.

You’ll also want to compare engagement driver scores to each other so you are able to see in which areas of the employee experience your company is performing well or poorly. The easiest way to get this result is to use a clear and consistent numerical scale with every question. We recommend a scale of 0-10.

Some businesses may be used to asking questions on a Likert scale, which ranges from 1-5, or from “strongly disagree” to “strongly agree”. This, however, doesn’t give the same level of granularity.

For example, small improvements in scores over time, such as a shift from a score of 6 to a 7, will be shown on a 0-10 scale, but may not register on a 5-point Likert.

If you are already using the eNPS questions to measure engagement, it is important that you continue to use the same 0-10 scale with all your engagement driver questions. Switching between scales within one survey will confuse respondents and damage the integrity of your data.

Getting additional context with employee comments

Qualitative data can provide additional context to help you understand why your engagement scores are changing.

Employee comments will often point to specific issues that you can address in order to improve your overall engagement score. They can also provide an opportunity for you to start an open discussion with employees around a particular topic. This might allow you to allay concerns or provide updates on an issue that employees are consistently raising.

To achieve this, you should ensure that every question you ask has an optional comment box for employees to leave additional feedback. It’s important to keep it optional so that your team only raise issues that they feel strongly enough about to discuss voluntarily – otherwise it can be hard to discern the significance of the issue, and you may end up with more comments than you can handle.

While it’s possible to introduce new policies and initiatives based on quantitative engagement and driver scores alone, qualitative comments give you a more accurate and efficient way to identify the exact causes of disengagement.

You might also consider asking open-ended questions, separate from your engagement driver questions, which simply invite employee feedback and suggestions without requiring a score.

For example, you could ask:

“If you had a magic wand, what’s the one thing you would change about [your organisation]?”

This provides the opportunity for employees to be a little more imaginative, but also to offer detailed, practical information that you can act on immediately.

Question writing tips to avoid biased results

The phrasing of your questions is an essential part of designing an employee engagement survey. If your questions are confusing or guide employees towards giving a particular answer, it can skew your results dramatically. When they’re phrased correctly, your data will more objective and actionable – and it will increase the chance of your employees responding. All this can be achieved by following some very simple rules.

Use clear, unambiguous language

Using too many technical terms in your questions might confuse your employees, which can lower response rates and result in inaccurate answers based on a poor understanding of the question. Awareness of technical terms, acronyms and internal language can vary across an organisation, so make sure to keep your questions as straightforward as possible.


”I understand how my OKRs support the Strategic Alignments”


”I understand how my work supports the goals of my team and department”

Only ask one question at a time

Trying to gather data on different aspects of engagement in one question can lead to muddied responses. For example, “my manager gives me meaningful tasks and always rewards good performance” covers three drivers (Meaningful Work, Recognition, and Reward). A better approach would be to keep these as separate questions.


”My manager gives me meaningful tasks and always rewards good performance”


”The work I do is meaningful to me”

Avoid leading questions

Positive or negative language can subconsciously guide your employee towards giving a specific answer. For example, ”I see a path for me to advance my career in our fast-growing organisation.” is biased and fast-growth might imply many development opportunities. Biased questions like this may serve to corroborate hypotheses, but the data won’t necessarily reflect reality.


”I see a path for me to advance my career in our fast-growing organisation”


”I see a path for me to advance my career in our organisation”

Use the first person and reference behaviours

First-person phrasing encourages reflection by helping the employee to recall real-world examples of their workplace experience. For example: ”I can count on my co-workers to help out when needed” encourages them to reflect upon their peer relationships and give a full, considered response.


”There are reliable and helpful co-workers within the organisation”


”I can count on my co-workers to help out when needed”

Avoid mandatory questions

Finally, always include an option to skip a question. This will help to keep your data clean, as those who don’t understand a question, or for whom it doesn’t apply, won’t be forced to give a random answer. If the same question is skipped repeatedly, this could be a sign of ambiguous phrasing, poor internal communication, or a deeper organisational issue. You may want to provide employees with ways to indicate why they chose to skip a question to help you understand further.


When you use a robust, theory-based framework to create an employee engagement survey, you end up with reliable results that make it easy for you to benchmark, and for real action to be taken.

However, there are three caveats to the advice we have given so far: If you want to guarantee accuracy in your results, and if you are serious about improving employee engagement in the long term, you must be consistent with your questions, ensure employee anonymity in your surveys, and also run them frequently.

Without consistency, you can’t measure change

The issue is simple: If you choose to edit and update your engagement survey questions often, it makes it impossible for you to measure how your employee engagement and engagement driver scores have changed over time.

Even minor question changes, such as swapping out a word for a synonym, can cause a variation in how people answer.

It’s important to resist the temptation to make edits, and remember that consistency is key. It’s more valuable to observe the changes in engagement scores, than the actual scores themselves.

A score of 8.2/10, for example, does not tell you as much as observing an increase from 7.7/10 to 8.2/10 over a six month period. This would show you that your engagement initiatives are working and help you to prove a return on investment. This is only accurate when your engagement questions remain consistent.

Anonymity encourages honest feedback and high response rates

Employees are naturally cautious when answering surveys, and understandably so. In order to get the most useful feedback, you are relying on your team to be completely honest and open about their feelings towards your organisation, leadership, their managers and their peers. Not all of this feedback will be positive.

If employees feel unsafe providing honest answers, for fear of negative repercussions, then they will hold back information – or worse – choose not to participate in the survey at all.

If you want to achieve high response rates, with honest responses, then you must ensure and communicate to your team that your survey is anonymous.

Regular surveys allow you to continuously improve engagement

Finally, while many businesses are used to running employee engagement surveys and initiatives annually, we would advise a much more frequent approach.

There are a handful of issues associated with an annual approach to employee engagement – namely a lack of insight from your data, and an inability to iterate and adapt your initiatives.

When you conduct a single survey once a year, you receive a quick glimpse at the state of engagement in your organisation. This glimpse has the tendency to be inaccurate, since it only reflects a single point in time, and is not representative of engagement over a broader timeframe. It also makes it very difficult to tell if engagement is currently increasing or decreasing.

When you run an engagement survey more frequently, such as monthly or quarterly, you give yourself the ability to identify engagement trends. This is especially useful for understanding the performance of your initiatives, and allowing you to adapt them accordingly. With an annual survey, you have to wait 12 months before you learn whether or not any improvement to engagement was made.

Best of all is a weekly “pulse survey” schedule, in which a small subset of your survey questions are sent to your team each week. While this may seem daunting to perform manually, there are many employee engagement platforms, such as Peakon, that are especially designed to make survey distribution, analysis and action-planning a walk in the park.