Testing and optimization are a necessity in any marketing endeavor. I’ve gone deeper into the subject in several columns, such as “Conversion Folly Funnel” and “We Tried That Already.” Today, I want to focus on one aspect of optimization: customer insight.
Success in testing doesn’t necessarily indicate success in customer insight. For example, you can test landing pages, determine the best landing page, and enjoy an increase in conversion. But do you know why it converts better? Oftentimes marketers gain knowledge of customer behavior, which is inferior to customer insight (defined as learning why customers are behaving the way they are).
While it’s possible to optimize and see increases without customer insight, you’re chasing diminishing returns. Exclusively chasing better numbers gives the marketer a weaker 2-D approach in a rich 3-D world. Gaining customer insight is more efficient and typically more powerful in maintaining an upward trend toward your goals.
How do you gain this customer insight? Customer surveys are one means.
Web analytics expert Avinash Kaushik collaborated with iPerceptions to give marketers the 4Q survey platform. 4Q is a free, permission-based on-exit customer survey. It’s delivered post-conversion and asks customers four powerful questions:
- What is the purpose of your visit to our Web site today?
- Were you able to complete your task today?
- If you were not able to complete your task today, why not?
- If you did complete your task, what did you enjoy most about the site?
At the recent eMetrics Summit in San Francisco, iPerceptions shared some early results of using 4Q. For retailers, it learned that 39 percent of visitors went to learn about products, while 27 percent went to buy. Of the 27 percent who went to buy, roughly only two-thirds actually completed that task. Visitors also told why they did not convert: 31 percent wanted better product selection, 24 percent desired better shipping options, 17 percent cited problems with the online shopping cart, and 14 percent said prices were too high.
Analytics will only tell you what people are doing, but knowing why they are doing it is a powerful optimization tool.
In this case, the retailer can make much better optimization decisions. While a retailer may already be working on an initiative to offer more shipping options, it now has data to support accelerating the project. Knowing that 17 percent said they had shopping cart problems, the retailer can dig into the analytics and gain better insight into what is happening.
You can also use this data to create personas to help your marketing initiatives.
Another simple means of customer insight are customer product reviews. Here’s how you can optimize using them:
For example, here are two bullet points from the product description for a lady’s watch before optimization:
- Contemporary style adds bold look to any wardrobe.
- Water resistant to 30 meters.
Now, here are two snippets from “emotional” customer reviews for a lady’s watch:
- It’s like wearing two silver chain bracelets with a beautiful watch centerpiece.
- I’m a constant hand-washer, and I don’t have to worry about “time stopping” just because I have to have clean hands.
Now here are the optimized bullets:
- This unusual double chain bracelet band and watch is an instant attention getter.
- No worries while washing hands, because this watch is water resistant to 30 meters.
Which description do you think converts better?
With customer insight you can more easily duplicate your successes, create more effective campaigns, and apply that insight to other site areas. And with our current economic situation, you can better budget and prioritize your optimization efforts.
Now go and learn what your customers are saying about you and your Web site.
*This article is cross-posted on ClickZ.
. .
About the Author: Bryan Eisenberg is co-founder and Chief Persuasion Officer at FutureNow. Join Bryan on June 3rd in Manhattan at the Call to Action seminar, the popular one-day course based on his New York Times and Wall Street Journal bestseller Call to Action: Secret Formulas for Improving Online Results. Not only will you learn the most effective online persuasion and website optimization techniques, you’ll get a chance to chat with Bryan over hors d’oeurves and cocktails at our “Happy Hour with the Experts” reception.
May 23rd, 2008
11:54 am
Love the idea of retooling product descriptions based on review content! To take it one step further… take review descriptions and use them in campaign copy.
May 23rd, 2008
1:20 pm
Bravo on your statement of “Oftentimes marketers gain knowledge of customer behavior, which is inferior to customer insight.”
And why does this oftentimes happen?
Because marketers, always pressured by time/resources/budget, end up relying too heavily on technology and not enough on gaining customer insight. Technology is great for a lot of marketing tasks, but it will never answer the all-important customer insight questions, like “Why didn’t visitor A convert?”
May 23rd, 2008
2:10 pm
Great post Bryan. I have been using 4Q since the day it launched and have had brilliant insights from it. While numbers for analytics are great - insights from customers are unbeatable. I was surprised at how many customers took the time to give us their invaluable feedback!
May 24th, 2008
2:16 am
Some really great insights into the psychie of the consumer, learning more about your customers is something every business needs to do and in my opinion the ariline industry needs to really pay attention to its customer smore.
May 24th, 2008
2:33 pm
You can also get customer “insight” by analyzing site visitor behavior. see for example http://crazyegg.com/ heatmaps or http://www.pagealizer.com/ page optimization suggestions based on user stats
May 25th, 2008
5:15 pm
I took a quick look at that survey site and the first thing I thought was that having a survey pop up over my site would irritate more people than would use it, thereby reducing conversions and causing more visitors to leave.
I’d be interested in hearing how and where people are using 4Q
May 27th, 2008
10:33 am
Thank you so much for these specific examples, Bryan. There is nothing better than actually asking a customer why they behaved a certain way! Too many organizations are basing strategic decisions on deduction without asking customers what they think.
To Mark, the best way to see whether it would impact customer behavior would be to test it for a sub-group of customers coming to your site. You may be surprised by what you see.
May 27th, 2008
5:14 pm
Thank You. We’ve taken our customer reviews and applied them to the prod description as you’ve mentioned. Will try the iperceptions questions. thanks again for your great insights.
May 28th, 2008
2:36 am
This commment is what seperates a talented SEO from someone who simply knows the steps involved. “Analytics will only tell you what people are doing, but knowing why they are doing it is a powerful optimization tool.” It’s why we must constantly work to see not only how we are ranking, but how the prospects that are visiting the site respond to it.
May 28th, 2008
10:44 am
Just something about 4Q since so many articles at Grokdotcom mention it. We use 4Q for two of our websites. It works really well for one of the websites (On the other website, no one seems to be taking the survey for some reason) and lots of people are giving us their opinions.
However, there is one small glitch- for most visitors, the survey window pops up too early. And though the instructions clearly say that the visitor can take the survey at the end, visitors don’t seem to be reading them. They assume they are supposed to take the survey right away and we get comments such as “How do I say anything about your website without surfing it?”:)Anyone having the same problem?
May 28th, 2008
4:07 pm
I haven’t used it, but what you are saying has confirmed what I thought when I saw the demo of the service.
Surely, it would be better as an exit pop-up, pop-over, pop by for a cup of tea.
May 29th, 2008
3:23 pm
Bryan, this was an extremely insightful post and I appreciate that sometimes there are insights that just the numbers themselves cannot surface.
I work as a Marketing Analyst at Bazaarvoice, and I can tell you from experience that product reviews will surface more than just product description insights. Some consumers are very active when given a free text field such as a product review, and we have found that an average 3 percent of our reviews contain questions. Some questions are rhetorical, but many are real customer questions for which they are hoping to find answers.
More importantly, reviews containing questions have an average 17% lower product rating and 18% lower net promoter score, so retailers are losing and frustrating customers by not addressing these queries.