Ten Biggest Mistakes Made When Conducting Employee Surveys

Very often, companies conduct employee surveys because it is merely part of the HR’s KPI’s. This overlooks the fact that a properly conducted employee survey can not only pre-empt employee disquiet and demotivation, but can further corporate directives and strategy.
Our experience in working with major MNC’s and Singapore public sector organizations allow us to understand industry best practice. On the other hand, we have also gained an insight into some of the more damaging mindsets in executing an effective employee diagnostic tool

Mistake 1: “The employees are not complaining so why conduct an employee survey?”

If we use the medical analogy, it would suggest that since you are well today, you will not drop dead from a heart attack tomorrow. Progressive companies conduct employee surveys regularly every 12 to 24 months. This reveals areas where employees are dissatisfied and also reinforces that the organization cares enough to conduct a pro-active survey, regardless of visible employee sentiments

Mistake 2: “I already know they are not satisfied with pay. Can we drop that question?”

First off, to be completely thorough, you must complete a holistic survey focusing on no less than the eight categories of engagement factors developed by To exclude certain questions will immediately project to employees that you are evading key issues. More importantly, it will prevent you from developing a fuller understanding of the relative standing of the various factors for your employee population. By keeping an open mind, HR departments quite often are surprised that employees have opinions that defy conventional wisdom, e.g. employees may not view Basic Pay as dissatisfactory but are concerned by Bonuses. Or Career Advancement.

Mistake 3: “Can I save money and just survey a few departments?”

Sure you can but you can also quickly alienate employees by coming across that you don’t care. Sampling is an effective way to get valid results with a smaller employee number if done correctly and random. Our position is that whenever feasible, involve the whole employee population to get the most valid data and to create sense of inclusion and ownership of the survey process

Mistake 4: “Let’s forget the employee briefing- the conference room won’t hold so many people”

One of the biggest obstacles to an effective survey is employee distrust. And one of the best ways to deal with this issue of confidentiality is to communicate directly with the employees. Research suggests that the validity of feedback can improve by 125%  by having a face to face briefing and fielding of questions. This ratio increases even more through the use of an external third party professional. In line with this step, a best practice for pen and paper surveys is to have employees deposit completed surveys in ‘locked boxes’ that are monitored and collected by third party professionals.

Mistake 5: “Can we do it in pen and paper as well as by Internet”

Conceptually, this is possible and might increase the overall response rate. Logistically, we discourage this. The effort and time required to create two survey instruments and to merge the database can itself be quite prohibitive and if not done carefully, can introduce substantial error. In terms of the validity of the results, the online survey has built in mechanisms to ensure that there are no double-entry responses. This cannot be matched in the pen and paper survey which would suggest that the response data is not statistically uniform.

Mistake 6: “Let’s just do it in English”

This is fine if you are an MNC or where your workforce is relatively young. While literacy in English is increasing, at least in Singapore, the level of literacy in older employee segments, age 45 and above, remains low. Having second language translation, at least of the questionnaire proper can pay back quickly in terms of relevant findings. In our experience, a further step to collect very accurate data is to have facilitators walk through surveys in non-English languages and dialects. This increases the effort required by has invariably proved to bring dividends in employee feedback and buy-in to the action plans developed

Mistake 7: “We want lots of data in the report”

What you really want is clear analysis based on lots of data. This means having a clear idea of what action plans that you should put in place based on the analysis of feedback. This is actually a luxury that every company can afford: we call this the bi-factor analysis. If you analyse based on just one factor- i.e. satisfaction of your employee with respect to different factors, you get just that, the sense of how happy they are with different drivers. With a bi-factor methodology, you develop a 2x2 matrix (see below) which graphically quadruples the action options and sensitivity analysis for your action planning. This is the powerful bimatrix approach we advocate at


Mistake 8: “We have never needed to communicate to our employees”

So maybe that’s why your survey is only good for fulfilling someone’s KPI (usually HR). Communication is vital for a survey to flourish the way it can. First off, prior to the survey proper, market the survey actively. Get top management to state clearly the objectives and importance of the survey process. This may be an email from the MD or CEO to every employee, or it can be a formal statement in a staff townhall meeting. The more emphasis placed, the higher the response rate and validity of feedback from staff. During the survey, capitalize on the briefing to reinforce corporate plans and values. Post-survey, use the reporting of results and proposed action plans to project the responsiveness and employee-orientation of the organization.

Mistake 9: “Why is the participation rate so low this year?”

Probably because there was no follow through from the last survey two years ago.

From the analysis of the data, develop action plans with different timelines: immediate low hanging fruits, middle term plans with good return on investment, and long term plans requiring the most effort but having the best payback. A nice balance in reported action plans might have, say, announcement of casual dressing on Friday (long hanging fruit), change of job titling (mid term- printing of new cards), to review of compensation plans (long term – requiring competitive market data).

Mistake 10: “We need comparative data from the industry”

It is very instructive to be able to see and say that,’ relative to our industry competitors, our employees are 5% more satisfied (63% vs 58%). This will be very true if the different organizations use statistically similar surveys- both in format and scope. In normal circumstances, different wordings by different companies nullifies their comparative value, e.g. “Satisfaction with pay” is ranked 6 out of 7 in a Likert scale in Survey A cannot be compared with Agreement ranking of 6 out of 7 that “Pay is good” in Survey B.
Focus on internal issues. For optimal results, compare data with those from a different base year. This will give you the most usable database for your employee diagnostic

About the Author

Woon Chet Choon holds a MBA from the Richard Ivey School of Business, University of Western Ontario. He is an 15-year veteran in research and assessment on peak performance and employee listening. As managing consultant at Satisfied Employees, he has worked with numerous global companies as well as Singapore GLC’s. Prior to developing the online engine, Chet Choon was senior consultant at Hewitt Associates in Connecticut and Singapore. Chet Choon is also a master trainer in Strategy Development at FTC Kaplan and PSB Academy.

[email protected]