Surveys Can Help IS Keep Users Happy: Gartner
- 04 February, 1999 12:01
IS departments are increasingly under the gun to show results and benefits to the company, but proving value can be a challenge in itself.
However, internal satisfaction surveys conducted by IS can be a good tool in defining the value of IT, according to Donald Feinberg, vice president for GartnerGroup Inc.'s measurement division in Asia-Pacific and Latin America.
"Communication between IS and the end-user has been terrible worldwide," said Feinberg. "IS needs to listen to the needs of users and meet those needs," he noted, adding that surveys open up a channel to improve understanding of end users' expectations and requirements.
While most large organizations would likely call in outside expertise to conduct a survey, it's also a useful tool for small to medium-sized companies -- provided they have the basic know-how and a commitment to resources. "Small companies need to be more reactive to what employees need and want," Feinberg said.
According to Feinberg, there are six steps in creating a survey. First is to design the survey, considering what the objectives are, who will be surveyed -- such as management, end users or both -- and the size of the survey.
In addition, Feinberg suggests including a question that asks what is most important to them so that responses can be appropriately weighted.
Next is to distribute the survey -- then get the end-users to complete it quickly. "Motivating respondents is usually the biggest issue," Feinberg said, noting that incentives are the most effective method for a high return rate.
Another key issue in motivating respondents is to ensure anonymity. "The worst thing you can do is give people the impression that the company knows (how the respondents answered the questions)," he said.
Data is then collected and entered to prepare it for analysis and review. The results are presented and an action plan is developed based on the results. For example, areas that show high satisfaction and that are important to users should be maintained; areas that show weakness should be targets for improvement.
"You've got to deliver on the results (of the survey)," Feinberg said, otherwise people will be reluctant to fill out further questionnaires and IS could lose an important tool.
The key pitfalls that can cause surveys to fall apart are due to inadequate management backing to follow through on the survey, resources drying up or being reallocated, or asking the wrong questions and not getting specific answers.
Feinberg estimates that about half of large companies worldwide conduct regular IS user satisfaction surveys.