trbo Insights – Giovanna Tramontano’s Favorite Feature

In our blog series, we asked our colleagues what their favorite trbo tool feature is and why. After Minea recently gave us insights into personalized product recommendations, Client Success Manager Giovanna Tramontano tells us which of our features she likes to work with the most.

1. Which trbo feature do you enjoy implementing the most?

In my daily work, I am particularly happy when I can set up a multi-variant test and determine the optimal recommendation logic for a recommendation.

2. Can you describe a multivariant test’s function in a little more detail? Where can it be included on a website and what opportunities does it create?

With trbo’s multi-variant testing options, it’s child’s play for customers to compare variants and understand what is most likely to appeal to users. In a multi-variant test, different versions of a measure can be tested against each other — in this case, different logics of product recommendations. For example, product recommendations from the store’s ‘top sellers’ category are tested against ‘recommended products’ based on user behavior. Using filters, we can then make the selection of products displayed in the recommendation context even more granular in relation to price or category. 

The exciting thing about testing is determining where the recommendation should be integrated on the website. Two pages are particularly suitable:

• Product detail page
• Shopping cart or checkout page

A recommendation on the product detail page, based on the product category and showing similar products in case of an out-of-stock item, is a must-have for me! Also, the shopping cart page offers the chance to show similar products with the logic ‘match’ or ‘is often bought together’, which can increase the conversion value.

3. Why do you like this measure so much?

Recommendations help store operators to suggest relevant products to their users and encourage them to buy more items. With the multi-variant test, you can find out which recommendation logic appeals to users the most. Do users like the products from the ‘recommended products’ or ‘top trending’ content option? In this way, you can decide which logic to use for recommended products with reliable figures instead of relying on your gut feeling. 

But we can go one step further: compared to some of our competitors, our multivariant test results have proven that trbo product logics achieve superior results and have a positive impact on conversion.

4. Let’s say an online merchant wants to implement a multi-variant test on their site. What would you advise them to do before implementing it?

Before a test is set up, the framework conditions must be defined. Otherwise, you risk not getting any reliable results in the end. To do this, set up a hypothesis of what the test is intended to achieve and define the KPIs to be tested. In order to be able to assess whether the results are transferable to all users after the test has been completed, the significance level of the results must always be taken into account.

But there are also a few points to consider in advance when setting up the recommendations: The product feed should always be up-to-date and maintained. The more information there is about a product, the better the recommendations will be. It is also important to include certain attributes, e.g. whether a product is already out of stock or how an item has been rated.

5. Is the option a classic favorite or do you think its impact is underrated?

I think that testing different recommendations is an all-time classic. Nowadays, users expect more and more personalized information adapted to their needs when shopping, as part of a good user experience. This requires store operators to know and understand their users’ preferences. That’s why testing is essential.

Conclusion: Product recommendations on the product detail or shopping cart page are an effective tool for increasing order value and conversion rates. Users don’t have to scroll through long category pages and can quickly find what they’re looking for. A multi-variant test allows you to determine the preferences of your (potential) customers. If the results are valid, you can make adjustments for all store users on this basis.