Surveys are viewed as a fair, democratic and legitimate way of assessing a wide range of opinions. Employees like being asked about their thoughts and ideas — it’s human nature. And with web-based tools, a simple opinion survey can be quickly thrown together, administered and yield meaningful results.
However, essential components are missed more often than not and survey results can end up being obtuse or even contain the wrong items. So, a deliberate, systematic approach should be adopted, with attention paid to each step to ensure a successful survey process.
Seven steps should be followed when implementing an employee survey:
Prepare for action: Employee surveys often fail because of a lack of integration into existing organizational initiatives and directives and a lack of endorsement from senior leadership early on in the process.
When the concept of a survey is still forming, key players in the organization need to support and buy into the process. Most senior executives respond well to a return on investment (ROI)-type analysis quantifying the cost and benefits of the survey effort.
Then it’s a matter of addressing the survey’s objectives by asking (and answering):
• What is the survey’s purpose — measuring engagement or change, reinforcing a new vision or strategic direction?
• What are the expected outcomes — improving engagement, managing change, enhancing employees’ understanding of a new vision?
• Who will be involved?
• What will the survey content be?
• When and how will the data be collected — online, paper, interactive voice response (IVR)?
• What will the reports look like — what will be the level of data cuts, what complexity and level of detail are needed?
• How will the data be reported and by whom? What type of training and support is needed? What is the time frame to report back the results?
• What is the process to followup — who will be accountable, how do we tie it back to other initiatives, when do we do another survey?
Design the survey: Focus on the balance between standardized questions and customized content (items) specific to the organization’s needs. If the goal is to measure employee engagement, many standard survey instruments require little or no customization. Off-the-shelf employee engagement instruments offer varying levels of sophistication. If a third-party vendor is chosen, verify its questionnaire has been tested for statistical reliability and validity by asking for a psychometric report. And if the goal is to measure success against a major change initiative, some or all of the items will have to be custom designed.
Keep survey length in mind. Respondents seem to be comfortable with employee surveys that run less than 25 minutes. This translates into not more than 70 closed-ended (radio button) questions and 15 or fewer open-ended (verbatim) questions. As a rule of thumb, it takes one minute to complete seven closed-ended questions and one minute to complete an open-ended question.
The process of designing items should follow five steps:
• gather preliminary information about important organizational issues important to top management
• identify key issues by summarizing and integrating available sources of information (interviews, business plans, CEO goals and objectives)
• discuss and validate the findings with those who provided input
• draft the initial survey document based on agreed-upon key issues
• pilot the survey with your representative group or other organizational members and make changes as required.
Spread the word: The key objective at this stage is to build awareness and understanding of the survey effort. The information provided to staff at this stage should answer questions such as: Why is the survey being conducted? What types of issues are of interest? When will results be available?
For a mid-sized organization, the communication plan should include a letter from the CEO alerting staff of the survey at least three weeks prior to the launch.
At least two weeks ahead, employees should be informed about the what, why, when, who and how through internal communication channels such as newsletters, staff meetings, intranet and email. On the launch date, survey invitations should be sent to the target population with instructions on how to access the survey and general survey information.
And every three to four days after the launch, reminder messages should be sent while managers and supervisors coached by HR should do walk-around reminders.
Launch the survey: Focuson the timing, method of deployment and response rates. Often employers think there is no time for a survey as they’ve just hired a new CEO or there are too many things on the go. Working backwards from the annual budget cycle or from an upcoming senior leadership or management meeting off-site are two common end points to anchor a survey period. Online surveys are usually open for a period of two weeks, with some extending to three weeks if the survey period cuts across the summer or December vacation periods.
The majority of surveys are conducted online as most employees have ready access to the Internet at work or home. And with respect to collecting demographic data, most organizations link individual survey responses to an internal human resources information system (HRIS). This technique reduces the number of questions asked and ensures respondents are placed in the right bucket for analysis.
Building a high response rate adds to the credibility of the data and provides for a more reliable and detailed analysis. Response rates vary wildly across industries, organizations and even internal business units. A response rate above 70 per cent usually indicates an interested and somewhat engaged workforce while low rates often means the opposite. Generally speaking, 75 per cent or higher is a good response rate, 50 per cent to 75 per cent is acceptable and below 50 per cent is low.
Analyze the data: Surveys generate an enormous amount of raw data. Once the file has been cleaned and labelled, it’s about creating and communicating a compelling story regarding the current state of the organization. Interpretation often involves the use of effective statistical techniques to identify themes. These advanced techniques are only helpful to people who know how to use them — with the techniques’ inherent strengths and limitations. Statistical tools in the hand of an uninformed user can cause a lot of damage, not only to the survey process but the organization as a whole.
Three areas of inquiry are often used to analyze survey results. First, statistics are used to identify key drivers or the most important areas that require attention. Second, the use of open-ended comments provides context and colour as well as general themes that may not have been easily identified in the statistical data. Finally, analyzing survey data relative to a norm or benchmark can provide excellent context and interpretation.
Report the results: Focus on determining the rollout process and deciding on the method and sequence to present the results to all those involved (see sample rollout plan below). As a rule of thumb, results are usually cascaded down from the executive committee to management and then to the front line.
Develop an action plan: Even if the first six steps are executed flawlessly, the overall effectiveness of the process will not be optimal if the survey results are not used and acted upon by management. It is important to have managers and employees take ownership of the findings and make the necessary changes. A process that can facilitate this transfer of ownership includes: clarifying results, prioritizing action items and building action plans.
One of the most effective steps after survey results are released is to encourage managers of work groups or business units to further clarify their respective results through discussion and dialogue with staff. Organized focus groups that can be facilitated either internally or by external consultants are one of the most effective ways to fully understand not only the “what” but the “why” of employee responses. A professionally facilitated focus group will provide the tactical information, ideas and understandings that allow you to prioritize all the potential actions that could be taken to help improve the organization.
The final step in the process involves building an accountability framework to ensure the actions are implemented. This may involve tracking and monitoring actions across the organization and tying specific actions to pay or performance. The important aspect of this step is for senior management to commit to keeping the process moving months after the survey.
Sean Fitzpatrick is president and founder of employee survey firm TalentMap in Ottawa. He can be reached at (613) 248-3417 ext. 500 or SFitzpatrick@TalentMap.com. For more information, visit www.talentmap.com.
Sample rollout plan after survey
Survey sponsor (HR)
1 to 2 weeks
Review of initial findings, raw data, trends, concepts, key relationships, hypothesis.
3 to 4 weeks
Formal presentation of key findings and interpretations, outline of next steps.
4 to 5 weeks
Formal presentation to upper management to convey key messages and themes and gain buy-in for changes.
5 to 8 weeks
Overview of corporate results and reports on their specific business units or departments as a catalyst for discussion in work units and to further clarify results.
5 to 12 weeks
General message highlighting survey findings. Senior-level comments and observations are encouraged.