Almost anyone can pull together a web-based survey using free or almost free survey tools available on the Internet. In fact, these do-it-yourself tools can help you save money and time by making use of resources already developed by professionals.
However, as some of you might know from getting involved with some other do-it-yourself activities, you still need to make sure you know what you are doing, follow instructions and allocate enough time and resources to getting the project done. Otherwise, do-it-yourself projects can sometimes become more costly and time consuming than working with a trained professional in the first place.
In browsing some of the available web survey tools, we found little information on how to ensure your survey efforts are successful. So, we´ve put together the following guide for those of you considering taking a do-it-yourself approach to gathering feedback from employees or customers. We´ve broken down our guide into three sections - getting usable data, managing expectations and acting on findings.
Getting Usable Data
Getting usable data depends largely upon the questions you ask. If you ask the wrong questions or ask the right questions in the wrong way, you´ll end up with survey results that may be difficult to use or interpret. Here are some helpful tips to get you usable data:
Less is more, especially the first time you are conducting a survey. The more questions you ask, the more answers you´ll get and the more time you will spend trying to analyze, interpret and act on the findings.
If you are using a template provided by the web survey company, make sure you can edit those questions to make sure they meet your goals. Using the off-the-shelf survey without reviewing it can lead to survey results that end up sitting on the shelf. So, think about why you need feedback and use these reasons to guide the questions you ask. Some of the templates that your web survey service provides may help you meet these needs but often times you may have unique needs. If you are going to write your own questions, here are a few item writing guidelines to get you started.
- Avoid double barreled questions
- Make sure your questions and scales agree with eachother
- Avoid jargon and make sure your survey is written for your audience (e.g., reading level, language proficiency and educational background)
- Avoid leading/biased questions
Make sure the websurvey service you use allows you to look at the results in mulitiple ways and not just by overall results. Important features include being able to conduct cross- tabulations (comparing different groups of respondents' results on the same items), and filtering (looking at results of only one group or a group based on certain criteria). Also, if you can´t conduct the analysis that you need using the service´s reporting tool, make sure you can download the results into a spreadsheet or other software program so you can conduct your own analysis.
Managing Expectations
Only ask questions that you are willing to act upon. For example, if you can´t move the location of the office, don´t ask employees to rate their satisfaction with where the office is located. Similarly, if budgetary constraints hamper your ability to enhance employee benefits, try to limit the number of questions you ask about benefits.
It is also important to manage expectations about confidentiality and anonymity of the survey results. A confidential survey means that somewhere someone will be able to identify individual survey results (e.g., by linking responses to an email address or unique log-in) but individual results are not reported, they are reported at the group level to maintain confidentiality. An anonymous survey means that there is no way to identify who said what, there is no way to identify an individual´s survey responses. As a rule of thumb, if you are gathering feedback from employees, you will want the results to be both anonymous and confidential, especially if you are conducting the survey yourself. If you are gathering feedback from customers, you may want to be able to identify your responses, especially if a customer is dissatisfied, so you can more easily act on their input.
Another way to manage expectations is to clearly communicate the goals of the survey, how the data will be collected and what will be done with the data. For employee surveys, it is especially critical to inform employees everytime you take action based on the survey results.
Acting on Findings
The third area to focus on when conducting your own survey is related to acting on the the findings. No survey effort is complete unless you do something with the results. Whether you are surveying employees or customers, your efforts are not fully realized unless you act upon the findings. Here are some helpful tips to get you started. Some of these may look familiar as they appeared in the Critical Metrics Checklist for Successful Surveys (which can be found by following the first link below or by clicking on the second link to have the list emailed to you).
The first step to successful action planning is communicating the findings and plans for action to the appropriate audience. Employees who were surveyed should be informed of the high level results and next steps. For customer surveys, you may want to share some key findings with customers but also with your employees, especially those who interact with customers the most.
We also recommend palnning for action from the beginning. One of the most common survey pitfalls is too much focus on survey administration, reporting and analysis and not enough on action planning. This can be due to the event-based nature of survey administration (there´s a start, middle and end) and the deliverable-focused nature of reporting. Action planning is more abstract and variable in shape and form. So it is important to set a plan from the beginning for action planning. Make sure to allocate resources from the start.
Hold people accountable for action planning. When someone's performance goals include acting on survey findings, you are more likely to get action out of the survey. Incorporating action planning as part of each manager's goals can go a long way to getting more action out of the survey results. Be careful to not overwhelm managers with too much action planning. If each manager acted on just one finding from the survey, the collective impact could go a long way to improve your organization.
