How good is your Web site? How do you know?
Even if you're an Internet novice, gathering content for your Web site can be a fairly straightforward issue. But when you put it all together, does it make sense? Can your visitors find what they need? Chances are you're too close to your own site to look at it with an impartial eye. Yet evaluating the clarity and usability of your Web site is a critical step toward optimizing your nonprofit's on-line presence.
Most nonprofits haven't solicited feedbackGuideStar recently helped conduct a survey on the interaction between nonprofits and Internet users. Sixty-five percent of the more than 1,000 nonprofits questioned indicated that they had not solicited any formal feedback on their Internet presence.
- 20 percent had surveyed their Web site visitors and/or e-mail list subscribers
- 15 percent had conducted usability testing
- 11 percent had surveyed their volunteers
- 11 percent had surveyed their donors
- 10 percent had conducted interviews or focus groups with Web site visitors and/or e-mail list subscribers.
Conducting a surveyThere are some processes that the Internet has made easier and there are others that the Web has made a lot easier. Conducting user surveys is quicker, cheaper, and easier than it was in the past and can be done entirely on-line. Numerous companies provide on-line survey tools as part of their services. If your needs are unsophisticated, there's no reason to purchase any software, and you don't have to be particularly tech savvy.
Some companies offer very basic survey tools at no cost. If these meet your needs, you won't find a more cost-effective method of gathering user feedback. Most free tools are fairly stripped down and saddled with usage limitations, but they often have fee-based options if you feel the need to upgrade. Some free sites to look into are:
Depending on how much you are willing to spend, you can easily conduct complex surveys and receive detailed breakdowns of the results. Check out such sites as Tech Soup and Nonprofit Matrix to learn more about individual ASPs and what they offer.
What to askSo you're ready to create a survey. What do you ask? Be sure to get important information up front. Seemingly mundane details, such as the type of computer participants are using (Mac or PC?), the version of the browser they're accessing your site through, and whether they have a dial-up connection or DSL, may prove to be extremely useful if users report slow page loads or odd behavior that you're not observing.
If you think you might want to ask some follow-up questions, obtain permission to do so and be sure to get the appropriate contact information.
Before you begin to craft the questions for your survey, make sure you have a clear objective. It may sound obvious, but what you ask depends on what you want to know. Are you looking for general reactions to your Web site or do you have specific concerns? Identify the issues related to your site that are most important to you and concentrate on getting feedback focused in those directions.
Make sure your questions are short and clear, and that there aren't too many of them. Convoluted questions will lead to meaningless results. Don't use jargon or ambiguous language that is open to interpretation.
The bane of all surveyors is the half-taken survey; if users are confused or overwhelmed, they'll bail out without finishing and leave you without usable data. Remember, you are asking your site users for a favor. Assume that they don't have a lot of time. Keep it short and sweet.
Statistical vs. anecdotalThere are different formats you can use for your questions, depending on the flexibility of the survey tool you choose. Use yes/no, multiple choice, or scaled questions if you're looking to qualify user feedback into easily understood statistics. Open-ended questions can produce the most interesting responses, but contribute little toward statistical analysis.
If you suspect your pool of potential survey takers is limited, open-ended questions are probably the way to go. Even without enough responses for statistical validity, you can still learn a lot from users' comments, views, and opinions.
If you're interested in conducting a more scientific survey, questions based on a sliding scale will give you the best results. Keep the scale consistent, the wording clear, and avoid neutral middle ground answers that can cloud your results, e.g., very good, good, okay, bad, very bad. "Okay" gives hesitant survey takers an easy way out without offering a real opinion.
Keep in mind, if you want to conduct a truly scientific survey that will produce hard data that you can base important business decisions upon, your best bet is to invest some money in the services of a professional.
If you build it, will they come?
So now you have a survey and you (you hope) have some Internet users accessing your Web site. How do you bring the two together? Simply posting the survey on your site should be enough.
If participation rates are low, however, you may have to get proactive. If your organization has a database of engaged supporters, an invitation to take the survey can be sent directly to them. Don't be afraid to let colleagues and business contacts know about your survey.
But even if you're sending out invitations, it's best to leave the survey on your Web site where it can be accessed by the general public. This way you'll reach a wider range of your site's audience and possibly connect with supporters (and potential supporters) you didn't even know you had.
The preceding is a guest post by Patrick Ferraro is the Editor of the GuideStar Newsletter.