BFCM workshop | October 8, 2026Apply to Attend
NEW: Agentic SearchExplore Agentic Search
NEW: Meet Kim, your in-app merchandising agentAsk Kim Anything
Refer & earn up to $1,000 in rewardsRefer Now
Get your free store auditGet My Free Audit

How a Bundle-First Product Page Earns John Beerens 60% of Its Recommendation Revenue

JohnBeerens.com has been delivering beauty since 2004, built by Dutch hairstylist John Beerens into one of the best-known beauty retailers in the Netherlands. The store carries premium hair and beauty brands for two audiences that shop very differently: consumers buying for themselves, and the salon professionals and distributors who order through a separate B2B login.

On its product pages, John Beerens puts ready-made bundle sets first in the recommendations, ahead of the items shoppers usually buy together. Over a 30-day window, that one block earned 60% of the store's recommendation revenue and 65% of recommendation orders. Across all four Kimonix recommendation blocks, 11.5% of clicks turned into an order, and together with Kimonix-sorted collections, 18.8% of store revenue ran through Kimonix.

The operation behind those numbers is sizable. The company moved into a 6,000 square meter facility in 2019, extended beyond the Dutch market into France and Germany, and was voted the most popular website in the Netherlands in the fashion and beauty category. Alongside next-day delivery on orders placed before 23:00, it runs live shopping streams and the JB Club loyalty program.

60%

Of Recommendation Revenue From the Bundle-First Block

11.5%

Store-Wide Recommendation Click-to-Order CVR

14.7%

Personalized Block Click-to-Order CVR

The Discovery Problem at Catalog Scale

A premium multi-brand beauty catalog serving two audiences across three markets creates a product discovery problem that grows with every brand added. A consumer looking for a shampoo and a salon owner restocking the same line want different things from the same category page, and no team can hand-pick what surfaces for each of them in three languages.

The problem shows up in two places: the collection pages where shoppers browse, and the product page where they decide what else belongs in the basket.

How John Beerens Runs Merchandising and Recommendations Together

On collection pages, Kimonix orders products automatically from live sales and stock data rather than by hand, and keeps re-ranking them as demand moves. That is where most of the Kimonix-driven revenue comes from.

On product pages and in the cart, four recommendation blocks each cover a different moment of intent. A personalized block works from what the individual shopper has shown interest in. The frequently-bought-together block handles the bundles, then fills with items shoppers usually buy alongside the one on screen. Two more show recently viewed items and similar alternatives. Every block falls back to a data-driven sort when there is little shopper signal to work from, and skips products that are out of stock or are free gifts, so a click is never spent on something that cannot be bought.

Explore Kimonix product recommendations →

John Beerens - Kimonix case study visual 1

What Each Recommendation Block Contributes

Kimonix reports every block separately, and each one earns its place for a different reason.

Explore Kimonix collection sorting →

  • Frequently bought together, with bundles first, is the workhorse, converting 12.4% of clicks into orders.
  • The personalized block converts best, turning 14.7% of clicks into orders, the highest rate of the four.
  • Similar products earns the most per view of any block, more than four times the average, because the orders it wins are worth nearly four times the average recommendation order.
  • Even recently viewed, the simplest of the four, turns 8.8% of clicks into orders.

Key Takeaways

60%

Of Recommendation Revenue From the Bundle-First Block

11.5%

Recommendation Click-to-Order CVR

Over 30 days, the bundle-first block on John Beerens' product pages earned 60% of recommendation revenue. Across all four blocks, 11.5% of recommendation clicks turned into orders, and the personalized block reached 14.7%.

  • A bundle-first block on the product page can carry the majority of recommendation revenue on its own
  • Collection sorting and product recommendations do different jobs, and running both covers the browse and the basket. Most of the Kimonix-driven revenue came from sorted collections, with recommendations adding to the order on the product page
  • Report every recommendation block separately and judge it on clicks, orders and revenue together. Each of the four ranked differently depending on the metric

Frequently Asked Questions

What results did John Beerens get with Kimonix?+

Over a 30-day window, 18.8% of John Beerens' total store revenue ran through Kimonix, about three quarters of it from automatically sorted collection pages and the rest from product recommendations. Across four recommendation blocks, 11.5% of clicks turned into an order, and the best-performing block reached 14.7%.

What is a good example of product bundling on a Shopify store?+

John Beerens shows bundles inside its product recommendations. On every product page, the frequently-bought-together block reserves its first slots for the store's ready-made bundle sets, then fills with items shoppers usually buy alongside the product on screen. Over 30 days, that one block earned 60% of all recommendation revenue and 65% of recommendation orders.

How does frequently bought together work on Shopify?+

A frequently-bought-together block on the product page suggests items shoppers commonly buy with the product on screen. John Beerens runs it through Kimonix with one addition: the first slots go to the store's ready-made bundle sets, then co-purchased items fill in, and a performance-based sort takes over when there is little purchase data. Out-of-stock products and free gifts are skipped. Over 30 days, 12.4% of clicks on the block turned into orders.

What is the best product recommendation app for Shopify?+

Look for an app that personalizes recommendations per shopper, reports every recommendation block separately, and works alongside your collection merchandising. Kimonix covers all three. At John Beerens, one of the Netherlands' best-known beauty retailers, four Kimonix recommendation blocks turned 11.5% of clicks into orders over 30 days, and the personalized block on the product page and in the cart reached 14.7%. Together with sorted collections, 18.8% of store revenue ran through Kimonix.

Show every shopper the products they are most likely to buy.

Request a Demo