CHALLENGE

Michael Jasma, CEO and Founder, Gem Affair, had been selling jewelry online and in brick and mortar stores for years. Gem Affair’s ecommerce site was successful, but Jasma worried that he wasn’t getting enough customer feedback from the website, and might be missing opportunities.

“An email blast is a way to show them what we’re thinking, what’s on our minds. But unless they send an email that says ‘Hey, I think you should do X,’ we can’t find out what’s on their minds,” says Jasma. “So you’re kind of looking at a one-sided conversation.”

Jasma and his team wanted to a way to complete that conversation, and learn more of their customers’ thoughts and opinions about the ecommerce site. That way, they could improve the online shopping experience.

CAMPAIGN

Jasma and his team used an old standby, the survey, to open the second-half of the dialogue with customers. They focused on questions that would identify problems with the company’s website, and uncover additional services that customers wanted.

Here are the four steps they took to create a survey program and use the information they gathered:

Step #1. Build the survey

The survey asked five questions, using a combination multiple choice and open-ended formats.

Multiple choice questions:

1. Based on today’s visit, how would you rate your site experience overall?
o Scale of 1 to 10

2. Which of the following best describes the primary purpose of your visit?
o Buy
o Check prices
o Check order status
o Add a review
o Enter contest
o Other, please specify

3. Were you able to complete the purpose of your visit today?
o Yes
o No

4. Which of the following best describes how often you visit the website?
o This is my first visit ever
o First visit in 3 months
o 2-5 visits in the last 3 months
o 6+ visits in the last 3 months

Open-ended question:
5. What do you most value at the www.gemaffair.com website?

Step #2. Launch the survey

The team started running the opt-in exit survey in November 2008. When visitors first arrived at the Gem Affair website, an overlay screen asked if they would be willing to take a survey (see samples below for screenshots).

If visitors selected ‘Yes,’ the survey loaded in a separate browser window. Visitors could then shop the site and complete the survey afterward. If visitors selected ‘No,’ the survey was not loaded and the overlay screen disappeared.

Two additional surveying tactics:

- Query half the visitors.

Jasma’s team set up the survey to be shown to half of the visitors to the site to avoid losing too many potential customers. Also, to avoid annoying returning visitors, the survey did not ask people twice to partake.

“With a survey, you’re always concerned about bouncing people, or turning them away because they don’t want to fill it out,” Jasma says.

Jasma’s team ran the survey continuously -- and continues to do so. At times, the team only shows the survey to 10% of visitors, when they are not seeking to answer specific questions, Jasma says.

- Let it run.

The survey needs to run long enough to provide meaningful results. Jasma suggests analyzing at least 30 to 45 days worth of data to find trends. Looking at data from a shorter period can skew your results and provide false insights.

Step #3. Analyze data

Jasma’s team exported the survey data into an Excel spreadsheet to look for trends.
They used a simple analysis process that counted the number of times a specific recommendation or complaint was mentioned. They then looked for items with the highest number of mentions.

Here are the top four opportunities they found for adding services and improving their website:

->Insight # 1. Sizing

Gem Affair did not offer ring sizing, and customers stipulated that this was an option they wanted when ring shopping online.

->Insight #2. Engraving

The ability to personalize jewelry with messages or initials was another service customers requested.

->Insight #3. Site load time

Customers complained that certain portions of the website were slow to load.

->Insight #4. Navigation

Customers liked Gem Affair’s navigation and found it to be easy to use -- which was a surprise to Jasma and his team. They had been considering a new navigation structure prior to launching the survey project.

Step #4. Prioritize and act on insights

Jasma and his team knew better than to let a single survey dictate everything from their marketing to their website design. “We’re certainly not going to let this drive our entire road map, but it certainly does influence it towards trying things and putting things on hold,” he says.

After weighing everything from resources to the costs and benefits of each insight, here are the actions they took:

->Action #1. Add sizing

The team hired an expert jeweler to provide sizing on any ring to any size for one fixed price.

“We had lots of projects going on that we thought were going to drive conversions -- none of which was ring sizing,” says Jasma. “It’s not that we hadn’t thought about it. We just hadn’t prioritized it over some other things.”

->Action #2. Prepare to offer engraving

The team plans to offer engraving in the near future.

->Action #3. Fix the website

The team discovered that the website had a JavaScript problem that was slowing load times. This problem was quickly resolved in-house.

->Action #4. Leave the navigation alone

Before the survey, the team was designing a cascading navigation structure to add to its site. That idea has been put on hold, Jasma says.

“We just had our own internal impression that we were kind of falling behind the times in technology, that everyone was doing cascading roll out navigation, so we probably needed to do that,” Jasma says. “I don’t want to say that we scrapped the idea, but if we do it, we’re going to test it thoroughly.”



RESULTS

The two main steps that Jasma’s team took -- offering ring sizing and fixing the website -- are showing promising results, and overall satisfaction with the site has increased:

- Overall conversion rate increased 0.25% percent after fixing the website’s JavaScript problem.

“That is pretty huge,” Jasma says.

- The team has seen a 4.5% lift in the sale of rings with the sizing option, compared to rings without sizing, year over year. (Jasma noted that he cannot be absolutely sure that these sales would not have occurred without the changes.)

- Customer satisfaction ratio has jumped from 77% to about 84% since starting the survey and making the changes.

Useful Links related to this article:

Creative Samples from Gem Affair’s survey program:

http://www.marketingsherpa.com/cs/GemAffairs/study.html

How to Get Double Digit Response from Customer Surveys
https://www.marketingsherpa.com/barrier.html?ident=3053


4Q: Ran the team’s survey
http://4q.iperceptions.com/

Gem Affair
http://www.gemaffair.com/