Every day, there is a crisis that breaks out in some office in the world which ultimately leads to misunderstandings between employers and employees. Somewhere down the line, there has been a breakdown of communication between employers and employees, and we have to do something about it.

We must do something about it, because let’s face it: employee engagement numbers are dismal. Only 13% of employees worldwide, according to Gallup, are engaged at work.

HR professionals have long advocated for employee surveys to be used in gauging employee satisfaction and employee engagement, but we know that it’s not enough. Employee surveys are often employed in order to generate a meaningful outcome after they’ve been executed, but too often we don’t use them in making sound decisions that affect employees.

How then should we execute employee surveys in order to make sure that the data generated is used to make these sound decisions? Here are three ways:

Set clear goals for the survey.

Employee surveys oftentimes are made to account for as broad a scope as possible, trying to solve every single problem. The most effective surveys, however, try to solve only one set of problems first so that it would be completely resolved before moving on to the next one.

Set clear expectations for how the survey should be used.

Employee surveys are designed to gauge employee engagement, but there is always the likelihood that it can be gamed if employees believe that the data will be of no use to the organization. Companies that take surveys seriously will more likely than not see employees taking the survey seriously as well, giving more meaningful data.

Engage employees and note externalities while using the data to effect meaningful changes.

Data can’t be used in a vacuum when making corporate decisions, but whether we like it or not, it happens. Companies often take survey conclusions for granted, with the results being the only things that they should address when in fact the reasons for why their office environments are dysfunctional may be found outside the survey, and often only when talking to employees. When taking all of this into account, better solutions can be devised.

Employee surveys are often taken for granted, and are usually given to employees for the sake of giving them, rather than for meaningfully wanting to seek change. But this is flat out wrong. Surveys exist to help bring clarity to what exactly is going on in an organization so that appropriate action can be taken. It only follows then that employee survey data should be used for that purpose.

Employee engagement is a two-way street–employers and employees should trust one another to do the right thing. That includes making sure that employee surveys are used effectively so that everyone is happy.