Survey classification types

How we classify our survey types at Culture Amp

Jared Ellis avatar
Written by Jared Ellis
Updated over a week ago

What can I learn from this page?

How we classify our survey types at Culture Amp

Who is this guide for?

Account Admins, Survey Admins, Survey Creators

At Culture Amp we deal with hundreds of different survey types as well as numerous minor variations. To keep things simple, we think about different classifications of a survey, based on when the survey is being run, and why.

The different formats of surveys is more about the technical setup of the survey and whether the demographic data is ties to an employee or not.


Baseline surveys are probably what most people think of when they hear the term "Annual Employee Survey". We tend to avoid getting locked into the mindset of only surveying annually. Many fast-moving organizations survey their employees every quarter. Theses surveys tend to be longer and cover a broad range of topics for the organization. However, we still recommend keeping to a maximum of 50 questions. This should be able to be completed in 5-10 minutes. Irrespective of the frequency, baseline surveys are still vital. These surveys establish a strong, comprehensive understanding of the culture at your organization. The baseline survey measures employee engagement as the primary outcome. This is the thing you want to understand and improve. We call this your "index". Generally this will include questions such as "I am proud to work at ACME". In addition the survey should establish how employees feel about the major aspects of your culture. Questions are grouped around a variety of topics. Example topics include Leadership, Strategy, Company Performance, Collaboration, Learning and Development. In the baseline, you'll get a measure of all these topics. However, you will also use driver analysis to highlight hot spots. Driver analysis tells you which of your topic questions are most strongly correlated with engagement. It is calculated from your employee response data by examining the patterns between how employees responded to the engagement questions compared with how they responded to the other questions in your survey. In addition to driver analysis, you'll usually include comparisons to the relevant industry benchmarks. This lets you know where you stand against your peers and competitors. What you see as a "low score" on a particular question might not be that unusual compared to the industry as a whole. Baseline Surveys most commonly use the confidential format.


Trend surveys are a shorter survey intended to track progress on activities resulting from your baseline survey. Are you moving the needle in the way you expected? Some organizations repeat the baseline survey again, but we usually recommend going with a shorter pulse survey that targets your driver and engagement questions. This keeps employees focused on your action areas and is a useful way of reminding employees of the activities you have underway. Not only that, it reinforces your commitment to these actions and re-engages the employees in that. So, if your baseline survey was 50 questions, the trend should be 10-15 questions. Employees will be able to complete this in less than 5 minutes. In some cases you may be tempted to sample only a portion of your employees. However, we generally recommend against this as sampling can lead to results that are not representative of the organization or demographics within it. If your survey is short and targeted, your employees won't mind giving you 5 minutes to let you know how you're going anyway. Trend surveys can also be run in a "pulse" mode where all or some of your employees are surveyed on a semi-periodic basis. Trend surveys can also involve polling the employee population over time. For example, in Culture Amp we have the capability to survey every employee over a defined period such as a quarter. Within this period of time, every employee will be asked to respond. This means we can track trend. You don't see individual changes. However, if employees start to answer more or less positively, this will be reflected in the reporting. Due to their tracking nature, trend surveys use the confidential format.


The domain diagnostic is a bigger survey than the previous trend survey. We'd recommend 25 to 50 Questions, of which around half will be specific to the domain of interest, and the rest will track your drivers and engagement trend to ensure efforts are still on track. This survey should take 5 to 10 minutes to complete. Domain diagnostic surveys focus on something like Manager Effectiveness, Leadership, Values or Benefits. The sort of topics important enough to require some detailed understanding, but would be too difficult to get a full understanding of in your baseline survey. Being able to ask more questions on this topic lets you actually diagnose where an issue lies. So from a Benefits diagnostic, you won't just understand that there is an issue with benefits, you'll actually understand which Benefits are working for your employees and which ones aren't. Our classic example of a domain diagnostic is the Manager Effectiveness survey. This may not be a key driver of engagement, but as your managers are usually the real drivers of effective change in your business you want them to be self aware and readily equipped with all of the feedback that they need to really be effective. The Manager Effectiveness diagnostic will ask detailed series of question on how employees feel about their managers. This can then be used to provide detailed action planning for each manager - things like training, mentoring and cross-skilling. Depending on the content, surveys work with most survey formats.


Event-Based can also be a broad classification. These are long-running surveys that capture employee feedback at different events. The classic examples here are candidate, recruitment, training, on-boarding and exit surveys. Other examples may be events such as training courses or moving location. These surveys capture employee feedback at these "events". Unlike other survey types, these surveys tend to stay open. Instead of viewing the data as an overall, you'll look at different cohorts. For example, "how do employees on-boarding today feel as opposed to a similar batch 6 months ago?"

Did this answer your question?