We recently announced the launch of Recommendations Web, and we couldn’t be more excited! This extension of our recommendations offering for email allows you to easily and quickly add dynamic, personalized product recommendations anywhere on your website and provide a consistent brand experience throughout the buying journey. With the ongoing projects and maintenance required for your website, we know your IT request queue can become quite unruly. So we set out to develop a solution that would support marketer-driven design and deployment and limit the need for assistance from IT. And that’s exactly what we’ve done. Our solution gives you, the marketer, the tools you need to maintain and enhance your web recommendations experience. Let’s take a look at what it takes to get your site up and running.

A Simplified Setup

I won’t lie. Adding this new content to your website still requires adding some code. But we have limited the initial setup to the placement of the following: Two JavaScript snippets in your website’s footer or tag manager. The first loads the necessary scripts to collect needed data on the page, such as contact and product info. And the second initiates the actual request for the recommendations and displays the products on the page. Div tags on each page where you want to place a recommendation. These div tags act as placeholders for your recommended products. In most cases, this can be done in just a few template files that lay out the general HTML for similar pages, such as your product and cart pages. Define a page identifier (if applicable). In most cases, you’ll want to use the same recommendations across similar pages on your website, such as product pages. If you have a well-structured URL that can identify these pages, the page identifier isn’t needed. But if you don’t, you can set a variable on the page that defines the page type. Define reference products and request time criteria (if applicable). Depending on the features you leverage, a few additional function calls may be required to supply additional information to the recommendation, such as the reference products to use when determining the results of your recommendation. For example, you would need to supply the product ids for the items in a cart for a Frequently Bought Together recommendation based on carted products. That’s it. Unless you fundamentally redesign your website or re-platform, this is a one-time setup by IT. “The team at Oracle Bronto has created an extremely marketer-centric tool for creating flexible on-site product recommendations,” said Justin Bergson, director of digital marketing, ecommerce and creative, for Luna Gourmet Coffee & Tea Company. “Implementation was a breeze, and we were up and running in no time. Kudos to the team.”

Marketer-Driven Design and Deployment

With the website configured, it’s the marketer’s turn. The actual decision-making on what recommendations to display on each page and what they should look like is 100% marketer-driven, all from within the Recommendations app in the Bronto Marketing Platform. The steps are easy:
1. Build your recommendations.
If you’re already using Recommendations Standard or Premium, you may have everything you need. If not, you’ll use the same solution used for email to build your recommendations, defining the business rules and criteria to use and choosing from various predictive models (if you’re using Recommendations Premium). Recommendations Web
2. Design your layout.
Use our drag-and-drop layout editor to design how your recommended products should look on the site. This editor is similar to our email message editor, so if you’re a pro at using that, designing layouts will be a snap. You can even design different layouts for different recommendations and/or pages. Layout Example 1 Layout Example 2
3. Define and configure your identifiers.
Configure one identifier for each set of similar pages where you want the same recommendations and designs to be used. The identifier will be matched against the URL or page identifier you implemented on the requesting page. In most cases, you’ll start with three identifiers: one for your home page, one for your product pages and one for your cart page. As you become more experienced, you may choose to get more granular with your site or add recommendations to other types of pages, such as your wish list or category pages. Then decide what recommendations to use for each identifier and what layout to use for each recommendation. Recommendations Web When you want to change which recommendations to display or what layout to use, simply make those edits right in the app. No need to contact IT. “After our IT resource deployed the Recommendations Web code, our marketing team was up and running with live recommendations on our website in hours,” said Tim Taylor, ecommerce manager at RST Brands, a manufacturer of premium outdoor furniture. “And we’re able to modify the appearance and behavior of our recommendations with no additional IT involvement,” he added. And as I mentioned earlier, Recommendations Web leverages the same recommendations that are used in your emails. We use the same predictive analysis and apply the same business rules to filter your results for a specific consumer both on the web and in their email. As they move from one to the other, they’re seeing consistent product recommendations to help them make good buying decisions, which increases their loyalty to your brand. After having great success with email, Taylor’s very optimistic about the brand’s website results. “We’ve seen our email recommendations offer a 64% increase in conversion rate, so we are very excited to see a positive impact with recommendations on our site.”