USE CASE

, Ecommerce

INDUSTRY

, Retail

REGION

, EMEA

CUSTOMER SINCE

since 2022

FEATURES

, Algolia Search, A/B Testing, Dynamic Re-Ranking, Query Categorization, Query Suggestion, AI Synonyms, Rules, Analytics, Merchandising Studio, Insights API

KEY RESULTS

  • Catalog growth by 3x with no loss in relevance and performance

  • Improved product discoverability across 1 million+ SKU marketplace

  • Multinational search managed by small non-technical team

  • Deeper insights into customer behavior and emerging search trends

As the leading online marketplace in the Balkans — offering everything from electronics and home appliances to beauty, fashion, everyday essentials and far more — Ananas Ecommerce is looking to make waves in digital commerce, and customer trust is at the heart of that.  

Named for ananas, the word for pineapple across the region and much of the world, the company was founded in 2021 to connect buyers and sellers, offer a wide assortment of products, competitive prices and a top-tier shopping experience.  

Operating in Bosnia and Herzegovina, Montenegro, North Macedonia and its home country of Serbia, Ananas is more than a mere marketplace. From its headquarters in Belgrade, the ambitious company manages a complete business ecosystem that includes a courier service, a modern e‑fulfillment center, a logistics network and expanding FinTech and AdTech services.

Aleksandar Marušić, the company’s Ecommerce Merchandising Manager, says the company’s deliciously descriptive name represents the leadership role it hopes to take. Pineapple is, he says, “the royal fruit,” which wears a crown, stands tall and is sweet on the inside,  a reflection of its ambition to deliver satisfying customer experiences.

“Our ambition is to be a leader in customer experience in Balkan countries,” Aleksandar says. “We want to be one of the businesses that drives change in the region.”

That ambition and the company’s growing scope, he suggests, led it to Algolia.

 

The Challenge: Controlling Catalog Chaos and Staying Ahead of Shifting Behavior

  • Multinational and multilingual operations with regional nuances
  • Massive merchant-supplied catalog with inconsistent or minimal product data
  • Search as primary journey for over a million customers
  • Evolving search behavior, such as conversational and AI-generated queries

 

Serving millions of customers with products from thousands of merchants across the region — many of whom provide inconsistent or basic catalog entries — creates a complex, often chaotic marketplace that demands sophisticated search and discovery.

On top of that, the company needed a search engine flexible enough to handle non-English languages and the linguistic patterns and market-specific behavior of multiple Balkan countries. Naturally, Ananas customers are primarily focused on search rather than browsing, making it even more crucial to deliver a seamless, frictionless search journey. Algolia, he says, ticks those boxes and more.

“Being a marketplace comes with the cost of having merchants that don’t always have a rich catalog with good descriptions, titles and images,” Aleksander explains. “Findability is really challenging when you get a catalog of products that has only SKUs. We do a lot of things to make products searchable and to enrich our catalog through different tools.”

He says Algolia makes improving search for such products easier across multiple catalog entries — often little more than a photo and SKU number, and frequently one of countless similar items. “It’s not easy to find a solution that is scalable enough to search across more than a million products at any given moment,” he stresses, noting that having, for instance, 100,000 phone cases compounds and illustrates the difficulty.

Recently, a new challenge has emerged for Ananas.rs: customer search behavior. With its customers increasingly asking questions through AI apps like OpenAI’s ChatGPT or discovering products through TikTok, Instagram and other social media, Ananas needs to meet customer searches where they originate and how they are made. 

That includes satisfying long-tail or conversational-style queries as well as machine-generated ones. “We get a lot of search requests that are not really coming from a person, and we need to adjust the catalog to meet them.”

These challenges made it crucial for Ananas to find a search platform that could not only scale with its impressive growth but also adapt to regional complexities and reduce operational burdens.

The Solution: Scaling with Ease, Reducing Workload and Futureproofing with Algolia

  • Easy to use out-of-the-box solution centrally managed at scale by lean, non-technical team
  • Multi-lingual search across four countries, with local teams handling region-specific needs
  • Machine learning capabilities to reduce workload burden and job tasks
  • A/B Testing to continually refine search configuration based on customer behavior

Aleksandar says the company’s ambition to become a multi-service ecommerce leader in the Balkans is costly but worth it to drive change. “It’s easier and more affordable to just follow the tide,” he says. “But in our case, we want to make the tide; we want to drive the waves and make changes. That’s costly, and that’s why we use Algolia.”

Algolia was selected, he notes, because it was “the most promising solution” to overcome Ananas’ several hurdles and support its ambitious plans. It helps the company cost-effectively build the trusted service modern customers expect at the scale the business demands, creating the consistency, trust and longevity its ambitions require.

So deeply embedded is Algolia in Ananas’ tech stack that Aleksandar says it’s fundamental to its platform: “It’s almost like, I would say, that 80 percent of our solution is handled by Algolia. We are tightly bound together.”

The company implemented Algolia Search and a broad suite of features, including A/B testing, Dynamic Re-ranking, AI Synonyms, Query Categorization, Query Suggestions, Rules, Analytics and Merchandising Studio. These help the ecommerce company strengthen relevance across its massive marketplace catalog without growing headcount.

Thanks to Algolia, Aleksandar says a lean team of three can manage search and discovery across four countries from its Serbian headquarters. “Only a few people can actually manage the whole marketplace platform that covers four countries at the moment, and millions of customers.”

He adds that the ease with which business users can manage Algolia is a major advantage, noting that the company can onboard new graduates — with no coding or engineering experience — to full operational proficiency in just a couple of months. “With other similar search tools, you definitely need to rely on engineers.”

In addition to Algolia itself being straightforward to use, it helps reduce the complexity of handling queries across the multiple languages and countries Ananas serves. Configurations can be reused across regions, with search managed centrally and language‑specific adjustments handled locally. With Algolia, “it’s not really hard,” he says.

Aleksandar says that while Ananas is still only considering AI‑driven capabilities like Agent Studio and NeuralSearch, the company already benefits from the machine‑learning features built into Algolia. 

“Machine learning is doing a lot of work when it comes to labor and job tasks,” he says. He also praises Algolia’s new MCP Server — a bridge between AI models and Algolia’s data — as a “cool tool” that lets him, as a non-technical user, analyze data, understand patterns and drive meaningful changes simply by using natural language.

The Results: Growth Without Compromise and Deeper Insights

  • 3x catalog growth with no loss in relevance or performance
  • Improved relevance and product discovery across a 1 million+ SKU marketplace
  • Stronger relevance and catalog stability through Query Categorization
  • Deeper understanding of customer behavior and emerging search trends

One of the clearest demonstrations of Algolia’s fundamental impact on Ananas is the company’s own growth and how well the solution has supported it. Aleksandar notes that after a brief soft-launch period using another tool, Ananas moved to Algolia very early in its development, and the solution has grown in lockstep with the company.

“Even though we grew by about three times in scope, we haven’t suffered any relevance or performance drops,” he notes. In fact, he adds, “We’ve increased all the KPIs in every way.”

Algolia’s features have become vital to Ananas’ performance, especially Query Categorization, which helps the company address the catalog challenges caused by insufficient or inconsistent information from merchant partners. “It helps us really get the most relevant results when, by the catalog alone, we would not give them.”

So essential is this functionality that, when it’s turned off for technical or strategic reasons, Aleksandar says “everything breaks down.”

“Without Query Categorization, it would be much harder to sleep at night,” he reflects.

Importantly, the feature is helping the Ananas team glean customer behavior patterns, understand intent more clearly, and identify gaps in merchant-supplied catalog data. It’s not just fixing problems but showing the team where those gaps exist.

Insights from Algolia are helping Ananas optimize both performance and operations. Aleksander explains how A/B Testing is helping the company refine elements like typo tolerance, determine optimal configurations to maximize customer outcomes, and find the right balance between flexibility and control.

Meanwhile, Analytics shows the near-continuous shifts in customer and search behavior, informing the team as query types evolve and more requests originate from AI or automated systems. These insights help Ananas adjust catalog structure and search attributes accordingly. “We get a lot of search requests that aren’t coming from a person, and we need to understand where they came from and adjust the catalog and design new attributes that comply with what customers are searching for,” he says.

As Ananas continues to evolve in its ambitious journey to become even more of a regional powerhouse, Aleksandar says the company is constantly evaluating how new and emerging Algolia features can help it meet its goals, particularly Agent Studio and Personalization.

“My ambition is to use more of the features Algolia has, like agents. This is something I want to explore, and as a business we want to make better use of them,” he says. “We’re exploring it now, but we're not using it with customers. So that’s something that would be great.”

Get a customized demo from our search experts

Get a demo