Last post I outlined 10 reasons not to copy Amazon, with a promise to follow up with reasons why you SHOULD copy Amazon. While I could easily have compiled a list of 50-100 tactical items that could improve your conversion like unusual shaped cart buttons, delivery cut-off dates, customer reviews, mobile apps, triggered email or point-of-action assurances, I chose today to focus on 5 strategic things Amazon does really well, that could make you or save you millions of dollars.
Amazon Never Redesigns Its Site (Anymore)
Hands up if you remember when Amazon looked like this:
It was Amazon that popularized "tabbed navigation" for online shops. Who could forget Amazon's tombstone menu?
Which evolved into this:
In Amazon's early days, it launched redesigns much like any other business would. But now that it's older and wiser, it makes piece-by-piece changes with slow roll-outs. Amazon doesn't pull the rug out from beneath its customers with a complete facelift. Rather, changes are made gradually so it's easy for customers to learn the new piece within a familiar UI. In fact, many customers will hardly notice a change has even been made.
The dangers with major design overhauls are many:
- You risk confusing the heck out of customers who have taken time to learn your old site's processes
- You risk "breaking" something that's working, and you can't just roll back to the old design
- It takes a lot more resources to design a new UI from scratch than to make smaller, incremental changes
- The resources you place on the redesign project are often taken away from your existing system, so the existing system may go without crucial updates for a year or longer
- If you outsource your design to a 3rd party, decisions may be made based on aesthetics/design trends rather than your business interests
- If you change too much at once, you can't measure which change is responsible for an increase or decrease in revenue (or other KPI)
Usability expert Jared Spool once shared a story of a big-box retailer that spent $100 Million on a redesign. Sales dropped 20% after its roll-out, and it took 3.5 years to recover from it.
Radical site-wide redesigns are very risky, especially when design decisions are made without input from customers.
Amazon Understands Its Customers
When Amazon decided to radically redesign its navigation, it had already conducted extensive usability tests and customer surveys. From this research, Amazon learned that the 4 activities most important to customers are shopping, searching, saving and buying. So, Amazon made sure each of these options were prominently featured at the top of every page on the site:
Amazon Reduces Risk By Dripping Out Radical Changes
Despite the genius of Amazon's design team, Amazon didn't cockily roll out the new navigation scheme to everyone all at once. Rather, it took a very conservative approach to roll out. It began by showing the new design to a small sampling of "new" visitors (those without a cookie), as they were less likely to be familiar with the current design, and therefore less likely to freak out at the change. Slowly, more and more non-cookied users were exposed to the design. Once confidence in the new feature was high enough, it was introduced to loyal cookied customers. Eventually, after weeks of dripping out the new version, it became the default design. (Hat tip to Jared Spool for sharing this story)
Amazon is Inconsistent
Usability consultants may review your site and criticize it for inconsistency. For example, your Widgets category page might look different than your Sprockets category (different sales tools, merchandising zones, filtered navigation options, etc.), recommending that you keep your design consistent to avoid customer confusion.
In reality, customers aren't as obsessed with consistency as designers are. There are no “rules” that say your site has to look and feel the same apart from the main theming (colors, logos and main navigation), i.e. as long as the visitor is aware they have not left your site, you're okay. You have freedom to organize content in each category in the way that best suits the product type or customer segment.
Amazon understands that the purchase decision for a camera is different than for consumer packaged goods, clothing, books, software and furniture. For example, a typical category page might look like this:
While the Digital Camera category includes a product finder:
The Grocery & Gourmet Foods is merchandised much differently. A customer may purchase many items in one go, or purchase very frequently. Customers may also be interested in discovering new products, so showing New Arrivals, "Top Sellers in Beverages" and "More Items to Consider" based on previous browsing history makes more sense than with consumer electronics that are purchased less frequently:
Amazon has even has built different advanced search tools for different product categories:
You don't need to fear inconsistency on your site. Building different features and tools for different product categories can make your site more usable and useful than sticking to a rigid framework across all.
Amazon Never Stops Testing
Amazon continually tests its designs and features, and you should too. You're never finished learning about what works and what doesn't.
Looking for tips on A/B and multivariate testing? Join me on Tuesday, July 20 for our free webinar Taking Your Site Performance to The Next Level With Optimization Testing. Sign up today!