Employee Opinion Surveys … Go Ask Your People

By Bill Catlette and Richard Hadden


Even at 6:10 AM, I instantly recognized the top item in my in-basket as a customer survey from Marriott. A believer in and occasional practitioner of the ‘handle paper only once’ doctrine, I gave the form a quick second glance to confirm the observation, crumpled it, and drew ‘nothing but net’ as the paper wad arced across the room and found its way to the bottom of a trash can.

Though disposing of unsolicited junk mail has by necessity become a cold blooded ritual for anyone with a mailbox, this particular piece of paper was at least slightly more welcome in my office than the scads of daily offerings for yet another credit card. Why? Well, for openers, we’re big fans of Marriott, whose hotels we stay in every week because the ‘Marriott experience’ is consistently good, and their people unfailingly polite. The hotel being surveyed was an Atlanta-area Courtyard I’ve stayed at perhaps a dozen times. Marriott’s leadership practices are regularly referenced in our seminars, and in the interest of full disclosure, I also happen to be a Marriott shareholder. So why not take a few seconds and give them some feedback? Moreover, what does this have to do with you?

On the premise that most organizations have at one time or another considered doing an employee opinion survey, and many who have done so have found the process less than fulfilling, this month’s article offers some ‘do’s and don’t’s about surveys, and in particular, for taking the pulse of your workforce.

Following are what might be termed the ‘Critical Success Factors’ for using such a tool, a few of which Mr. Marriott (and others of us) would do well to pay attention to.

1. Commitment to the Process – Organizations deploying an employee survey absent the genuine support and commitment of top management are wasting time, money, and valuable management credibility. In fact, they are running the risk of making things worse.

Conversely, given the demonstrated resolve of senior management to look under ‘every corner of the rug’, including their own, and to make genuine (often time consuming) efforts to understand the data, and to act on it, the effort is likely to be rewarding.

In most respects, this comes down to not one, but two ‘C’ words – Courage and Commitment. Do you have the courage to risk being told that “your baby is ugly”, and the commitment to do something about it? If so, read on.

2. Understand What You’re Getting Into – On the surface, conducting an employee survey (or any survey for that matter) is a relatively straightforward process…You compile some questions, go ask people what they think, and tabulate the results. Often overlooked is the realization that the very act of doing so creates expectations on the part of those being surveyed – particularly when that population is your workforce. Expectations that:

A. The results will somehow be shared with them
B. Something will actually happen as the result of the survey, and
C. No harm will come to them for telling you what they really think

In the case of the Atlanta Courtyard, I had completed a survey on this same property a few months prior, and though I awarded relatively high marks, a couple of specific problem areas were identified. There was certainly no expectation of receiving a personalized response, yet I had hoped that the problem areas would be corrected. They were not.

Though we have no scientific data to back this up, three decades of dealing with employee surveys have produced a few hard conclusions. One of them is that it is imperative to do a good job of managing these expectations. One possible consequence of failing to do so is that you quickly reach a point where the only people who bother to respond to your survey are more interested in venting their spleen than offering helpful feedback.

Employee and customer surveys are, by nature, perceived by line managers as a threat, in that they inevitably ask questions about the quality of supervision (both direct and indirect). This makes it imperative that you establish and communicate clearly, right up front, why you’re doing the survey, what will happen to the results, and what you hope to gain from the whole process.

If the results are to become part of your business metrics (they should), improvement over time should be a significant factor.

3. The Survey Itself – Too often, organizations trip themselves up from the start by asking too many questions, questions that aren’t relevant, or questions people don’t understand or could interpret differently. When it comes to the actual number of questions being asked, more is definitely not better. Twenty to thirty well-phrased questions will, in all likelihood, produce as much data as you and your management team can effectively manage. Mr. Marriott’s marketing/customer service folks flunked this one. The survey was a couple pages in length (in type that sent this 50 something reader scurrying for his glasses). And just what does my socio-economic status have to do with how well the Peachtree Corners Courtyard is performing?

Should we develop our own survey or use one that is commercially available? Clearly, there are advantages and disadvantages either way. The major advantage of designing your own survey is that it affords you the ability to ask exactly the questions you want to ask. The disadvantages are that you may shy away from asking questions you should be asking, and by definition, preclude any opportunity to benchmark your results to others.

Surprisingly, cost is not much of a differentiating factor since whatever you might save by going the self-serve route will likely be spent in internal haggling over the questions, crunching your own numbers, or managing the database. Often, in the beginning stages (i.e. the first few years) of using a survey, it may be advantageous to use one of the commercially prepared (externally processed) variety, as it tends to ease the minds of those who worry about where the responses are going, who will see them, etc. Just make sure that if you take this course, you’re working with a survey instrument and service provider you’re prepared to stick with for a while, as it’s important to be able to establish some consistency in the data that is generated. Otherwise, you’ll never be able to measure the results of your efforts over time.

4. Survey Administration – Some important things to consider here have to do with how the survey is actually administered (i.e., face-to-face, by mail, electronically, etc.), and the instructions people are given about completing the survey. Generally, face-to-face administration yields a higher participation level and the ability to answer any questions people may have about the survey effort. Administration via the Internet is faster and cheaper. Regardless of mode, it is vital to ensure that everyone gets a consistent message about why the survey is being conducted, what will happen to the questionnaires/results, etc. And, calling to mind the televised images of hanging chad inspections so popular in Richard’s home state, you might want to think about data integrity BEFORE doing the survey.

5. Data Presentation – In this area, there are two important considerations:

A. The survey effort will be successful only to the extent that people below the rank of vice president actually take ownership of the data. Take pains to ensure that the data presented to them is indeed relevant to them. Generally, each workgroup or team should get its own discreet report. (Hint – dumping a 50-page report on a manager’s desk is not seen as helpful.) The data should be presented in a fashion that enables them to make sense out of the report and begin putting it to use without having to whip out their Statistics 101 text to look up what a standard deviation is, or become a ‘survey expert’. In short, you should insist that they be provided usefully formatted data, accompanied by user-friendly tools.

B. Stale information is of little or no value, be it financial data or employee survey results. The time between the actual administration of the survey and the return of the results should be kept to a minimum…certainly no more than a month.

6. Survey Frequency – Given that one of the major benefits of a survey process is the opportunity to measure results over time, organizations should commit themselves to periodically resurveying their workforce. Much like an operational or financial audit, an annual cycle is generally an acceptable interval; except in cases where either the results themselves or the existence of some significant internal events might call for an earlier resurvey.

Here again, more is definitely not better. We all have enough noise in our lives (can you say spam?), so be considerate and make your survey efforts count for something.

Oh, and another thing… Mr. Marriott, when you get a minute, please get that air conditioner control in room 122 fixed.

Why not have a look at PeoplePulse today?

If you are interested in a demonstration of full featured online survey software offered by a full service provider, please request a demo.

Alternatively contact PeoplePulse on ph +61 2 9232 0172 to discuss how we may be able to help you

Next Article:

Exit Interview Surveys