5 Steps to Capturing Fickle Consumers

Behavioral targeting isn’t limited to online behaviors. Don’t forget about your existing customer data, such as current products owned or a propensity score, which can be used to identify the best offer for each customer. This data can then be married to the customer’s current online behavior.

While targeting can be effective, more automated and individual targeting using behavioral information can quickly require a complex set of rules that’s impossible to maintain or even test. A more sophisticated method of behavioral targeting uses mathematical models to predict the most compelling content and offers based upon all that’s known about a visitor. This type of model learns and adjusts over time to dynamically optimize visitor experiences with content that yields the highest conversion rate. This approach enables a broader range of content, such as destinations or travel packages to be presented to the right visitor based upon predictive attributes for that visitor.

Step 3: Design and Test Product Pages
It’s safe to say that nearly every aspect of your site — aesthetic, navigation, products, product descriptions and online features — has an impact on consumers’ shopping experiences. Their online experience ultimately influences their decision to keep buying from your brand. Once you’ve captured visitors’ attention with personalized content on the homepage or a landing page, they’ll begin to navigate through your product pages. Therefore, it’s important to determine the best online experience for them.

A proven way to do this is through A/B and multivariate testing. Virtually all site content can be tested, from the homepage and category pages to site navigation to product page content and layout all the way through purchasing funnels and shopping carts.

To get started, identify your business and site goals. Figure out what you want to optimize for your visitors, such as calls to action, layout images and user reviews. Then determine your optimization goals, such as clicks to add to cart, revenue per order, cart size, response rates on offers, etc. — any metrics that you feel provide insight into the consumer experience and impact on your brand.

Testing a variety of experiences across thousands of visitors will help you not only improve your site, but will validate the impact of these content changes on your business, thus enabling you to determine their impact on live visitor conversions before making any permanent changes to your site.

Step 4: Streamline and Optimize Your Registration, Log-In and Checkout Processes
Abandonment rates continue to plague online retailers, and ease of use with the checkout process can be a decisive factor in drop off as well as loyalty. Forrester Consulting found that 88 percent of online shoppers are likely to return to a site if it’s easy to use, particularly during registration, log-in and checkout processes.

Here’s where multivariate testing can be used to validate that your user log-in, registration and checkout processes are optimized for site visitors. Even the slightest changes to a checkout process can impact your overall conversion rates and customer loyalty. By testing metrics such as clicks to the next step, errors in entering form data, completion of the process and revenue per order, you’ll be able to evaluate different visitor experiences and their impact on results.

The checkout process is a sensitive area for presenting offers to visitors. For starters, you shouldn’t offer competing products at this stage. Price sensitivity relative to their current cart size is important with offers that you present. The offer content and placement also need to be tested. For example, retailers may offer a premier service which gives customers free delivery on all orders for the next 12 months if they sign up and pay a small annual fee. Different options for copy and placement of the offer during the checkout process can then be tested to find the option that increases sign-ups but doesn’t distract visitors from completing the checkout process.