In a previous article, I wrote than employee surveys don’t work. You can read that one here: Why good people leave

It ruffled a few feathers, and one of the biggest questions I was asked was “but whyyyyy don’t employee surveys work?!” Sit down, my young padawans, and we shall begin.

Leading the horse

The most basic reason as to why employee surveys don’t actually work is because it’s more than likely the surveys are predicating the response. When we are asking things of each other – anything at all – there’s usually a reason, and as such we are expecting a response within a certain set of parameters. In a survey, the expected response is tightened up, as it needs to be able to fit into a nice little grid to be balanced, measured, and weighed. A closed-response question like “do you think your manager does a good job?” with a ‘yes’ or ‘no’ checkbox fails to ask why or how the employee came to that conclusion.

And yes, sometimes these types of questions might have a free-writing component to ‘add more’… but if they are truly using the free-writing component of the question to initiate meaningful action, why include the precursory question? The short answer: it’s a control mechanism. If the employee answers ‘yes’, we likely won’t have to worry about the next part. Never mind the fact that they could be an absolutely fantastic manager that also gets high and then comes into work, every single day…

There is also another interesting titbit about the way we ask questions. Most languages have evolved from other languages before them. The Modern English language is a great example; drawing on no less than a dozen other languages as it evolved from Old English, through to Middle English and into the Modern English variants we use today.

Language is experience

So with our languages and the way in which we learn, experience, and teach them, we find that everything is event-based. A young child learns the word ‘no’ very quickly as they are scolded by their parents for a particular behaviour. A more complex example is the term ‘Darwinism’; to understand the term, you first need to know who Darwin was, as well as what he did. Then you need to understand the use of the term in context. This goes for all but the simplest words. The human race has evolved in its own particular way and its languages are based on the experiences that preceded it. Try explaining the term ‘Darwinism’ to a Martian and watch their head spin!

This experience will also be used in the development and administration of any survey. It would usually start with an event, like a complaint or issue deemed worthy to warrant further investigation. Then it might become the normal thing to do, because “we’ve been doing it for years now”. However that first pull to understand what is going on will serve as the basis for all future missions to understand what’s going on. And it will evolve.

As you progress through the years, you might see feedback like one particular area of the business having a hard time, or the whole company is facing massive staff attrition. These events will likely then lead to questions such as “if you’re thinking about leaving, how soon?” with time-boxed responses such as 1 month, 6 months, and 12 months. Again, that experience is creating a predicated question with expected responses.


Managers in charge of surveying their employees will defend these questions, arguing strongly that there is always ample space for verbatim feedback, and that employees can always put in anything not covered into those spaces.

In my experience, having been involved in numerous employee feedback surveys over the years, it’s plainly untrue. And there’s a simple reason why – the volume is too great. Larger organisations, which likely need to understand the word-for-word feedback of their employees even more so have it the worst. Imagine 20,000 respondents, each with their own ways of expressing concern and how their experience has been so far. The administrator of the survey might have 10 or 15 people working on the survey if they are lucky – and in my experience it’s usually half of that. How exactly could they go through every response with a fine-toothed comb?

Answer: they cannot.

And as such, the teams working on the survey might try to cut corners. Searching for keywords or matched phrases is a good example. Another might be the length of responses – after all, if an employee has written a 1,000 word response in there, it’s got to be worth something, right?

But how do they know what to look for, or what’s really important? Prior experience perhaps, or a pre-defined set of parameters that legal or risk management has handed down. Again, these are all leading, predicated filters that are looking for a particular set of outcomes and aren’t truly asking the employee what is going on for them.

Analysts like numbers, because they can manage them with ease. When someone throws a block of text their way, it’s extremely difficult to go back – en masse – and work out what’s going on for your employees.

Asking a question like “On a scale of 0 to 10 – 10 being extremely well, how well does your manager respond to your needs in the workplace?” assumes that you can roll up every little thing that your manager does for you into a number, benchmarked across a 12 month period, and that this will somehow represent a valid response.

In that example, management interprets the information as they please. No longer are they actually asking for a valid response, but are using the information to answer their own developed questions. An average, over a cycle period, to determine whether something has or hasn’t gone wrong.

Asking the critical questions

Many managers will try to combat this gap in the response by framing what they term to be ‘critical questions’. These questions are supposed to tick off the big items on their lists like feeling safe at work, combatting bullying, or inappropriate behaviour. The biggest problem with asking a critical question is that someone has to determine if it’s critical… and it’s usually not the employee answering it.

Bullying in the workplace is a scourge, and it happens more often than any of us would like to admit. A snide comment, an employee left off an invite, a lunch that only includes people in the workplace clique. All of these things can, and have, ended in major lawsuits all around the world. As such, the Legal and HR departments will try to micro-manage the policies. Including examples of how they define bullying is probably one of the worst ways of alienating an employee that is experiencing an issue that isn’t defined on a list.

The organisation will never know, because they are too busy trying to cover their own arses in lists of legal definitions to mitigate their own risks that they never stop to think that less control, less micro-management, and less bloody lists might help accommodate more of their employees. And yes, it also might mean more work for the HR team, who would then have to get involved in more workplace issues. Wouldn’t that be a good thing?

But therein lies another rub; engagement is wrongly perceived as an ‘HR issue’.

In reality, engagement is everyone’s issue, including, and especially, the leadership of the people that are asking for survey responses. And that responsibility is quite easily absolved, justified by the fact that the questions are asked on a survey, with the blessing of Legal and HR. So, the difficult questions might go unasked by managers, and instead left to that once or twice-yearly survey check-in.

What to do with all that data

Here’s an interesting conundrum. Let’s say you have an organisation that decides they are going to somehow read through every piece of verbatim feedback supplied in response to their survey. A number of things are likely to occur.

Firstly, management will start sifting through the data, and it is quite likely that a vocal minority will begin standing out. Something will have happened to a select group of people, which may or may not be horrendous, and management’s instinctive reaction will be a knee jerk one. Fix the vocal problem, quick!

Except that this problem might not be ‘the problem’. The vocal minority experiencing a particular issue will always seem larger than the issues faced by the silent minority, and we see examples of this in politics all the time. A few protesters become vocal on a particular issue, leading to a change in policy… when there are likely different, larger fish to fry to get to the real root cause of a problem, which also likely affects a great deal more people.

The second thing likely to come of any in-depth investigation of issues is… nothing. Yep, you read that correctly. Zilch.

You see, change is difficult… and if the management have uncovered a really big, serious issue that impacts a large proportion of their business and would require a serious change, they will typically weigh that change against their own nominal measures. If the problem is big enough, it will be senior management’s problem, in which case, their measure is the almighty dollar. “Got to keep those share prices trending upwards, after all… can’t spend all our money on operational change programs… unless we are going to get sued!”


Questionable anonymity

This is the last point to make, and is a concern raised by so many employees to me in the past. As an employee, how do you know your survey responses on (possibly serious) concerns are truly anonymous? After all, most large corporates nowadays have the capability to monitor your communications, read your emails, and in some cases, listen to your phone calls. Not only this, but that link they sent you to click on and complete the survey – it is a one-time use, unique link… because of course we wouldn’t want a person taking the survey more than once now, would we?

In this day and age of digital-everything, if someone really wanted to find out who supplied a particular anonymous response, it wouldn’t actually be that hard.

Let’s be honest here; if your employees feel the need to submit an anonymous ‘problem’, you have a bigger problem, and it’s not one that any kind of survey is going to help fix. Tackling a problem like a lack of employee trust in your organisation will take an organisational shift of earth-shattering proportions. It is not impossible, but is certainly an uphill battle.

So, no, employee surveys do not work. What to do instead? How about making the time to engage, genuinely, and empathetically of their concerns. Start a dialogue, one-on-one… What you find out might surprise you. Start with something that can’t be shoved into a matrix!

Something like… how are you feeling, today? 😊

Leave a Reply

Your email address will not be published. Required fields are marked *