For Better Email Results, Survey Your Audience

One of my recent consulting projects involved developing and analyzing a survey for a professional organization. “A survey?” you may ask, “What does that have to do with email marketing optimization?” Actually, they have a lot to do with one another.

Although they are underutilized by many doing email marketing and publishing email newsletters, surveys can help you learn about your audience and better target your email initiatives to their wants and needs — which is the key to long-term success with email. Best of all, surveys are one of the most cost-effective forms of market research, especially today when you can use the Internet to collect and analyze the data. Below I’ll share some tips about when to use a survey and provide some keys to successful surveys.

When to Use a Survey

The short answer is you should use a survey any time you want to learn more about your audience, but the following are some specific examples of when surveys can be particularly useful:

• If you have a Web site and are thinking about adding an email newsletter: Use a survey to find out who is visiting your Web site and learn what type of information they’d like to receive via email.
• If you have a house list you use for promotional email mailings: Ask the recipients about things like:
  • Product mix — How often are the products promoted relevant to the recipient? Are there specific product lines that are of interest while others are not?
  • Frequency — Are you hitting them too often? Not often enough?
  • Content — Do they prefer emails with a single product or emails featuring multiple products? Would they also like to get reviews? Industry articles? What could you provide that would add value?

   

• If you have a house list but don’t know much about list members: Gather additional demographics on your list. Most people are willing to tell you more about themselves on a voluntary basis if you’re going to use the information in the aggregate rather than tie the information directly to their data record.

Keys to Successful Surveys

Creating a good survey and getting statistically significant results requires some thought and planning. Below are some things to keep in mind.

Offer an Incentive

You’re asking people to take time to help you; offer them something in return. Entering all survey respondents in a drawing for a book or gift certificate is a cost-effective incentive to offer (for the client mentioned earlier, we gave away a windbreaker emblazoned with the company’s logo). You can also offer all respondents a premium, in the form of a PDF white paper, access to a password-protected area of your site, or other benefit.

Keep It Short

Be respectful of people’s time: Keep the survey short and sweet. You can always go back and do another survey in a little while (not too soon!). Focus on the things you most need to know now.

Make It Focused and Well Organized

Group similar questions together, and put the more personal questions toward the end. The survey questions should flow in a natural progression. You should leave the more personal questions to the end so respondents aren’t put off by them.

Don’t Require “Identity” Information

If you’re offering a physical incentive you’ll need to send through snail mail, don’t ask for everyone’s snail mail address just so you will have the address of the one winner. Just ask for an email address — then you can contact the winner and get a mailing address. You might also consider making the email address voluntary for those who want to enter the drawing; don’t require people to provide it.

Take Advantage of Technology

There are powerful survey programs out there that are much less expensive (and less hassle) than having your IT group build you a survey and back end to analyze the data. SurveyMonkey.com is one that I use; Zoomerang is another, and I’m sure there are many more. Don’t reinvent the wheel.

Ask Colleagues and Friends to Vet Your Survey

There’s nothing worse than an ambiguous survey question or a drop-down menu where none of the options represent your answer. Ask a small group of colleagues and friends to take your survey and provide feedback. You want to be sure all the questions are clear. Ask people you work with, but also include folks less familiar with your organization and your industry. Spouses, other friends, and family members can be a great test group.

Use Multiple-Choice Questions as Much as Possible

Multiple-choice questions are quicker for the respondent to answer and much easier to analyze. If you’re looking for a numerical answer, use ranges. If it’s a job title, state of residence, or other qualitative answer you’re looking for, put the possible responses in a logical order (e.g., alphabetical for state, rank for job titles). Don’t get caught trying to manually group and analyze free-form answers, such as “director of marketing,” “marketing director,” and “dir. mktg.”

Always Include “Other” in Multiple-Choice Lists

Also, give respondents who choose “other” a field to self-define their answers. This gives them an option if nothing in the lists fits. It can also help your results. In the survey I did recently, we included 11 broad professions and “other” as options for respondents. Almost a third of the respondents chose “other”; this shows the audience is very different than we thought. Because we included a field for the “others” to give us a free-form answer, we were able to group them into categories and include them in our detailed analysis of that question.

Don’t Take It Personally

You may get all glowing feedback and praise if you ask people’s opinion on something. Or you may not. If not, don’t get defensive. You need to make constructive use of the comments. If everyone wants to know why your print newsletter isn’t available via email when it actually is, that just means you need to do a better job of letting people know. Don’t beat yourself up about it; and don’t blame your audience for not knowing.

Use Statistically Significant Sample Sizes

You’ll want to be sure you have enough survey results to accurately represent your entire audience; this is where sample size comes in. Rather than go through the mathematical formula, here’s a link to the NCS Pearson Sample Size Calculator. You can input your population size, and it’ll give you a sample size. The instructions on the page are very good, so I’ll forego the explanations. Surveys often have an element of self-selection, so be forewarned you may not get a perfectly random sample. As long as the sample size is significant, you should still get valuable (not perfect, but valuable) data.

Wrap Up

Surveys are a great tool. They can confirm what you already “knew,” or they can show you things you didn’t know about your audience. The big surprise that came out of the survey I mentioned is only a quarter of the visitors to the professional organization’s Web site were members. In addition to opening up an opportunity for membership growth, this dramatically changed the email plan. Rather than just looking at email as a way to communicate with membership, we’re now looking at ways to use it with prospective members — and even interested parties who aren’t eligible for membership but who obviously have an interest in the organization and its activities.

Give surveys a try, and let me know what you learn about your audience!

---

“E-Mail Newsletter Publishing Fundamentals: A ClickZ Guide to E-Mail Marketing” — an in-depth walk-through on how to start your own email newsletter for profit

$129 PDF

Author and e-business expert Alexis Gutzman undertook the complex process of starting and publishing an email newsletter and details her experience in this briefing. “Publishing Your Own Newsletter” originated as a multipart series on internet.com. This briefing is a compilation of Gutzman’s essential writings about the email newsletter publishing process. Along with tips, tricks, and advice on what works best and what pitfalls to watch for, this ClickZ Guide includes product evaluations, code for capturing user information, and sound advice on user privacy concerns before implementing some of the tools discussed.