Rob Kaminoff

Before the web came along, catalog companies were the first to experiment with merchandising techniques designed to overcome the lack of a fully immersive in-store shopping experience. Punchy and persuasive product descriptions, detailed specifications, and imagery were needed to compensate for the lack of interaction with sales agents. The better catalogs have included lifestyle shots of how customers use their products, rather than merely listing products or showing photographs of products. Instead of listing products in alphabetical order, they tried groupings based on themes, styles, designers, complete outfits, etc.

BASIC INTERACTION AND NAVIGATION

The first generation of e-commerce websites did not take advantage of the medium’s interactivity. Instead they typically took content from a catalog and simply converted it to HTML for the web. The first improvements leveraged the fact that individual customers can navigate a website in any manner they choose. No longer a linear catalog with pages 1-30, merchandisers enabled users to shop by brand, by price, by style, by designer, by occasion, by room in your house, etc.

CROSS-SELLS AND PERSONALIZATION

The next innovation promoted cross-sells by showing additional products that complement products already browsed or inserted into shopping carts. Many of these recommendations were generated statistically a la Amazon.com, whereby “People who bought product x also bought product y.” Others are based on sheer logic as in “You bought the razor, now buy the blades.” Advances in personalization continue to drive higher conversation rates by presenting products that are most likely to be purchased by a given consumer. Of course good market research is helpful to this effort as well as more sophisticated analysis of individual click-stream behavior.

CONTENT AS KING

By providing interesting, relevant and fresh content on your site and emails, you provide non-promotional reasons for people to return to your site on a frequent basis. Blogging is a terrific way to provide this content. By linking your content directly to relevant products pages, blog posts provide a rich context that encourages higher conversion rates.  Such content should also be re-purposed for use in white papers,  e-books, emails, social media, etc.

RICH MEDIA

Additional merchandising strategies emerged as more consumers acquired broadband internet access and e-commerce brands added power and speed to their web servers. This gave consumers greater opportunities to manipulate product images (e.g., rotate, pan, zoom, change colors and backgrounds, see how they look on a model or a photo of oneself). Audio clips can add character to avatars used to help sell products on site. Flash and Videos show how products can be used across time and space and are especially effective for conveying the lifestyles associated with products.

SOCIAL SHOPPING

Of course, people (especially women) often like to shop with their friends because it adds to the “fun factor” and enables them to get opinions from people who they trust and respect. This broadens the selection (”how about this one?”) and helps close the deal (”that looks great on you!”). By offering user reviews, testimonials and discussion forums, e-commerce sites are able to provide some of the benefits of social shopping. By encouraging customers to share what they like within their own social networks (e.g., Facebook, Twitter, YouTube) they are to obtain feedback from their friends and followers. By “voting” on them on such sites as Digg, Delicious and StumbleUpon, they get additional opportunities to share their preferences with, and receive feedback, from others.  Other innovations, including publishing FAQs and providing web chat or click-to-phone capabilities,  provide social opportunities that compensate for in-store salespeople.

THE FUTURE OF ONLINE MERCHANDISING

Soon, such innovations as screen-sharing will become easier to use so that friends can indeed shop online together while geographically separated. People will be able to comment to one another via texting to their friend’s screens while they shop. Applications for iPhones and Tablet PCs will further provide a fuller, more immersive shopping experience and allow for better geo-targeting (e.g., present different products to people on the East Coast, West Coast, in different climates or within cities, suburban or rural areas). And although 3D web environments such as Second Life have not yet caught on in any major way, their ability to simulate real-world shopping experiences are likely to further enrich merchandisers’ future ability to convert browsers to buyers, increase average order size, frequency of purchase and the lifetime value of customers. After all, isn’t this what its all about?

What types of merchandising do you use or envision to facilitate e-commerce?

When the World Wide Web was first utilized by companies before the end of the 20th Century, the objective was simply to have a website.  Just as in the movie, Field of Dreams,  there was this idealist belief that “If you build it, they will come.” Since then, you have been able to leverage search engine optimization (SEO), search engine marketing (SEM), banner ads,  contextual ads, comparative shopping engines, affiliate networks, emails, blogs and your presence on social networks to drive traffic to your site.  As a result of these efforts you hopefully have lots of people coming to your site.   But what have you been able to do with these visitors once they arrive on your site?  Hopefully, you’ve either converted them to a qualified lead or sold them something.  Right?

If They Come, Will They Convert?

Unfortunately conversion rates continue to be low for most websites. So you get your graphic artists and copywriters to create landing pages that contain a compelling offer,  supporting content, and the best graphics to close the deal.  But why are only 1% of the people coming to your site actually filling out your lead form or purchasing something on your site?  You’ve done your homework of segmenting your customers and understanding the types of messaging that will resonate with each buyer persona.  And you have terrific creative people to translate these messaging strategies to the perfect copy and graphical elements.   So what went wrong?

No matter how good your research and creative talent are, the results are simply hypotheses regarding the elements of the landing page that you think will most appeal to your prospects.  But I’m learning that the hunches of even the best online marketers are simply wrong!   So where do you go from here?

If you’ve been following this blog at all (and I apologize for my lapse in posting), you know that the answer lies in testing the different creative treatments to learn which one’s actually convert the most customers on your site.  Consider your first stab at a landing page to be the control group and test various combinations of creative elements to determine the combination that leads to higher conversion rates.  Use your creative talent to come up with the various alternatives, but do not get ego invested in only one solution.

Let your customers vote with their wallets

Your customers are the actual experts as to the elements of your marketing campaigns and landing pages that work best.  You need to dynamically present the various combinations of creative elements (e.g., offer, body content, graphics and layout) to your customers.  Once the customers have voted with their wallets, statistics identify the combination that works best to convert visitors to customers.  You should then use that landing page as the new “control”, and develop additional versions to see if you can achieve even higher conversion rates.  This results in the continuous improvement, or optimization, of your conversion rates.  But how can you dynamically generate the different combinations and measure their results?

A/B and Multivariate Testing To the Rescue

Most of us can easily comprehend the idea of A/B testing, also know as  split testing.  By exposing half of your test audience to one creative treatment and the other half to another,  you can easily understand which treatment works better.  Take email subject lines for example.  Send emails with one subject line to one group of prospects  and send an email with another subject line to a different, but comparable, group of customers.  Whichever subject line results in the higher open rates should be used going forward, while the poorer performing subject line should be dropped.  The winning subject line now becomes the  new “control” against which you can test additional subject lines.  Present, test, analyze,  and repeat this process to see continuous improvement, or optimization, of your email open rates.

OK, so we tripled the open rates of our emails over the course of three weeks.  So we should declare victory, break out the beer and pizza, and call it a day, right?

Not exactly.  If people open their emails but don’t click through to your landing page, what have you accomplished?  Nada!  You now need to optimize the body of the email itself.  This requires a more complex form of testing named multivariate analysis.  Multivariate analysis enables you to test many variables at once to determine the combination of  elements (e.g., offer, content, graphics, layout) that lead to the highest percentage of clicks.

OK, now you’ve tripled your email open rates and doubled your click-through rates to drive six times the amount of targeted traffic to your site!  Now its definitely time to break out the caviar and champagne and give the crew a day or two off.  Right?

Unfortunately,  if  all of that traffic comes to your site, but doesn’t sign up for your newsletter or buy anything from you, it sounds like you’ve been spinning your wheels, and burning through your budget, without any return on investment (ROI).  Now you need to shift your focus to the landing page your prospects are sent to once they click-through on your email (or SEM ad,  web banner, blog, affiliate’s site, Facebook post, etc.)

Conclusion

The objective is to spend as little money as possible by testing the various permutations of the creative elements (e.g., offer, body content, graphics and layout) with just enough prospects to obtain a statistically significant result.  Declare the landing page that converted the highest percentage of customers the winner, and let it serve as the “control” for additional treatments to be tested.  The result?  Continuous improvement, or optimization, over time. You not only increase your customer acquisition costs, but also reduce your cost-per-acquisition (CPA).  Now that’s something you can take to the bank!