Make no mistake: The retail e-commerce industry is in excellent shape. With growth rates still approaching 15% year over year, the business of making money in online retail is certainly not difficult.
But then there's Amazon. Between its $75 billion in revenue last year and its 25% annual growth rate, chances are that Amazon.com outperformed you in 2013.
What can your business learn from Amazon, and what can you do to close the gap?
Here are 10 ideas that your company can use to achieve growth rates equal to―or greater than―Amazon's.
Five Tactics Amazon Uses That You Should Be Using
1. Deliver recommendations after, not during, checkout
Although intended to increase impulse purchases in the form of upsells and cross-sells, recommendations delivered during the checkout process invite customers to abandon the checkout and continue shopping. Of those who do, many will never return to checkout.
Recognize that customers in checkout are in a "buy now" state of mind. Amazon doesn't distract them with product recommendations, and neither should you.
2. Create urgency through 'out of stock soon' messages
Is your company proud of always having enough inventory to satisfy customer demand? Amazon's not, and it uses that to its advantage.
Amazon creates urgency to buy products by integrating real-time inventory levels into its website. Messages such as "Only 2 Left in Stock" suggest that the customer is looking at a popular product. And there could be no better answer to the question, "Why buy now?"
3. Use sessions and authentication to great advantage
Few realize that Amazon.com uses two authentication layers. The first layer keeps the customer perpetually signed in—which enables the company to deliver personalized recommendations whenever the customer visits any part of the website.
The second layer, however, requires the customer to re-input his/her password to initiate the checkout process.
Amazon's use of split authentication achieves the perfect balance between convenience and security. Customers feel safe, but also respected in the form of an intelligent shopping experience. In addition, Amazon keeps nonpurchased items in the shopping cart for as long as the customer remains signed in.
4. Use a data-driven approach to what works
Some marketers get caught up in the "Algorithmic Arms Race," whereby they select a product recommendation based on whoever has the most, or coolest sounding, algorithms.
Don't be one of them.
The great minds at Amazon learned a very important fact: The algorithm that works best is "Viewed / Also Viewed." Other algorithms may help sell technologies, but they're not quite as effective for selling products.
5. Take baby steps
Along with eBay, Amazon is one of only two companies on the IR 500 never to re-platform.
Why?
Amazon understands what many retailers still do not: In most instances, re-platforming initiatives and full website redesigns more often than not result in a decline in conversions.
Five Tactics You Can Implement That Amazon Doesn't
1. Make your website search visual
Customers cannot purchase what they can't find, but traditional text-based search results are not up to the task of helping customers find what they're looking for.
Consider the following:
- On some retail sites, up to 50% of visitors never reach a product detail page.
- Search results often rank among a website's top exit pages.
The product catalog is the nerve center of every retailer, and integrating that catalog―including product images, names, prices, and other information―into search results is a key advantage.
2. Think beyond the product
Amazon treats a customer's most recent behaviors as the most reliable indicator of what the customer wants in the future.
Why is that a problem?
Because throughout most of the year customers shop for themselves, whereas in the two months before Christmas they're more likely to be shopping for others.
Your recommendations can be more relevant if your algorithms simply think beyond the product by including additional data in the decision-making process, such as weather, times of the year, and more.
3. Emphasize what your customers care about most
Recommending the right products―not just any products―is not easy. As a mass merchant retailer, it's difficult for Amazon to know what customers want the most at any point in time.
If you're not a mass merchant retailer, you can take advantage of more limited interests that customers have on your website. Pay attention to things like...
- Items previously viewed or purchased, including colors and sizes (if relevant)
- Typical purchase frequencies or date ranges
A gift company may discover that customers who view chocolates also view flowers―but only during the Valentine's Day and Mother's Day holiday seasons.
The point is that companies serving niche markets can better afford the time and resources to mine data in an effort to deliver the best product recommendations possible.
4. Turn valuable experiences into relevant ones
There's a big difference between a valuable experience and a relevant one. Valuable experiences are those that express a quantified benefit to the customer, but the most forward-thinking marketers are moving beyond price-based merchandising strategies in favor of authentic ones derived from a deep knowledge of a customer's preferences, affinities, behaviors, and interests.
Compare Amazon's "5% Off" badge to the hypothetical retailer that badges home theatre systems with "Just 4 You" based on the knowledge that the customer recently purchased a television.
Which would get your attention?
5. Send relevant email
It's no longer enough to trigger emails to customers who have abandoned their shopping cart. The best email messages are personalized at the time customers open them, rather than email with predetermined content at the time the message is sent.
Open-time personalization enables marketers to deliver the most relevant experience possible, and if you trigger a cart abandonment email ("You Still Have Items In Your Cart"), but the customer subsequently purchases those items before opening your email—you can modify the message's content dynamically with new recommendations.
