For Italian cosmetics brand KIKO Milano, beauty isn’t just skin deep. And helping customers find the confidence — and the perfect products — to express their inner selves takes a lot of behind-the-scenes work.

Headquartered in Bergamo, KIKO Milano is just a short drive from Milan, the fashion and design capital of Italy. It was there that founder Stefano Percassi opened a small makeup store in 1997, the origin of a brand rooted in traditions of fashion, creativity and artistry. Fast forward nearly 30 years, and KIKO has seen a remarkable glow-up: it now boasts over 1,000 stores in more than 60 countries and a portfolio of 1,200-plus products in virtually every shade, hue and texture.

Despite offering a highly sensory retail experience, KIKO has always had a strong web presence. However, to improve conversions, it needed to better emulate the in-store journey and sense of discovery customers associate with shopping for cosmetics. After all, buying cosmetics is a tactile and visual process, built on a foundation of trust.

“It’s really hard to buy something remotely unless you try it,” says Nadia El Ghaouat, Web Catalogue Coordinator at KIKO Milano. “The online experience is very different from the store, where a beauty assistant sees you and advises you on what suits you best.”

It was with that goal of helping its customers better find products that bring out their inner and outer beauty that the team at KIKO sought to upgrade and enhance their search and discovery, and which led them to Algolia.

The Challenge

Giving Search a Performance Makeover

As KIKO made search and discovery a central part of its ecommerce strategy,, the company’s website faced several frictions: its native search capabilities were static and underperforming, it needed to support multiple languages and regional nuances, and, importantly, it had to handle the company’s impressive 2,000 SKU inventory.

At the same time, the growing shift toward conversational and long-tail searches made a more advanced search platform essential.

“Our native search was very static, was not helping the customer to find the right product, and produced a really low conversion rate,” Nadia says, adding that to bridge the experiential gap between in-store experiences and the online customer journey required a more intelligent search engine.

For KIKO customers, she suggests the limitations of the company’s existing search platform led to frustration around finding the right product, in sharp contrast to the more personalized guidance they’d receive from an in-store beauty assistant. Beauty needs vary widely — from subtle skin tone matching to creating dramatic seasonal looks — so customers often rely on guidance to understand what truly suits them. 

Static search simply wasn’t equipped to support that level of nuance, so KIKO investigated solutions that would deliver more relevant, intuitive and personalized product discovery. It found this with Algolia.

The Solution

Blending AI and Personalization into the Beauty Journey

To bring customers a more seamless online search and discovery experience and drive conversions, KIKO Milano adopted Algolia in 2021. Nadia says the solution’s ease of use by business teams and its strong multi-language support and flexibility — allowing them to design and control the user experience — made Algolia stand out among competitors.

“Algolia provides the engine, and then it’s up to the brand to supply the proper data and design it as you wish. The user interface is basically designed by our IT and designers,” she says, noting that its previous search vendor required the company to embed fixed components into the site.

“Of course, that doesn’t smooth graphically with your website. With Algolia, we have something that lets us design search in a way that looks like it’s part of the website right out of the box.”

KIKO built up its search and discovery experience in layers, all atop a foundation of Algolia Search. 

Semantic search was implemented, with country-specific indexing to aid customers searching its primarily English websites in their own language, better handle long-tail queries and accommodate misspellings. They then added Algolia Recommend to surface market-specific and frequently-bought-together, related or trending items without the heavy merchandising burden previously required.

“It was very helpful, and not just for the customers,” she explains. “Fetching products for every product across every country with that level of accuracy is something no editor could achieve. And not just because of the numbers, but because it changes every day. With Recommend, it’s super-autonomous: what’s trending today appears today, and updates as new trends emerge.”

KIKO recently deployed Algolia across its Product Listing Pages, where products were previously being listed statically, and polished relevance with the addition of AI-powered Dynamic Re-ranking (DRR) to combine sales data, product attributes and real‑world performance to create a more dynamic, customer‑driven ranking.

“This also gave us the ability to reorder by newness, which we weren’t able to do without static ranking,” Nadia says. “It was a milestone, changing and impacting the way customers navigate products.”

Prior to the 2025 holiday shopping season, KIKO took its search-refinement journey further by introducing token-based Personalization for logged-in users, aligning product discovery with customer behaviour and individual tastes. 

“If I want glossy lipstick, I will have glossy products appear in the first position; if I like red lipstick and only red lipstick, it will not show me pink ones,” Nadia explains. She adds that this feature honed relevance even further and was a natural next step after DRR.

Next up, the company plans to federate search this year across its wealth of content — moving beyond the product experience to aid customers by giving them faster, frictionless access to tutorials, seasonal collections, help center resources and, importantly, inspiration.

The Results

Higher Conversions, Smarter Merchandising and Less Manual Effort

Like the product experience it sells, improvements in KIKO’s online presence go well beyond the surface. Algolia is now an important part of a continuously improving tech stack as the company keeps its website fresh, engaging and highly relevant.

The combination of Algolia Search, Recommend, Dynamic Re-ranking and Personalization as well as other features has seen significant results. In fact, the company saw a 19 percent uplift in conversions after going live with Search, Recommendation, Product Listing Pages and Personalization compared to previous periods.

“I don’t think it’s a coincidence that conversions started improving last year, and step by step as we added search, listings and recommendation,” Nadia says. She’s quick to note that Algolia is part of a broader effort: “It’s an ongoing process of improvement, and Algolia contributes as one of several factors helping us optimize the website.”

Although the company has seen meaningful improvement in conversion rate, relevance and the ease with which customers find products, Nadia says, for her, the biggest values are internal: customer insights and reduced manual merchandising.

“The top benefit for me is that it gives me the ability to understand the customer better,” she says. “There are very deep and specific analytics in Algolia where I can see what customers are searching for, what they can’t find, what they are viewing, what converts better and what categories perform better.”

She reiterates how overwhelmingly complex it would be to manually set up fresh, trend-centric carousels by country across KIKO’s 26 online markets. “Time is the real asset,” she emphasizes. “Time is money, as they say. So having all that time spared and saved, it's already a big achievement.”

As KIKO continues to refine its online experience, the company is carefully considering how AI can make search further more conversational, reflecting the longer and more specific queries it’s increasingly seeing from customers. Its goal is to further bridge the gap between the trusted guidance of an in-store beauty assistant and the online journey, and to align with the ongoing changes in how customers search, browse, discover and buy.

“It’s hard online to shorten the distance between the product and the customer,” Nadia suggests. “This is where AI comes in, because regardless of how much content you have, there’s always going to be something missing: a seamless human interaction.”

She adds that AI won’t replace the human experience but will augment it, creating a feedback loop in which search behavior from Algolia informs the model, and the model, in turn, gives customers a more conversational journey that reflects how they now search. 

 “Humans and the AI can work together. I see AI as an ally, if used properly, to provide the customer a very close human experience.”

It is a retail inevitability, she says, and one KIKO is preparing to meet. With that in mind, KIKO is beginning to explore Algolia’s Agentic Studio and NeuralSearch capabilities, although the company is currently in the early stages of evaluation. 

As KIKO continues to refine its digital experience, Nadia sees Algolia’s retrieval and agentic capabilities as a likely next step in creating even more personalized, conversational and customer-centric beauty journeys.