Merkur is the leading home & garden retailer in Slovenia, with 23 physical stores and a 125-year tradition of a successful business. They create satisfaction for their customers every day with the best ratio between quality and price and excellent advice.
The company offers a wide range of products for construction, renovation, and maintenance, for entertainment, comfort, and quality living in one place.
Results From Site Search
- Back in 2020, after effortless and fast implementation of SearchNode, Merkur observed how their conversion rates from sessions with search skyrocketed by 9%, compared to their prior solution.
- Since then, the results only kept growing, the company continues enjoying the fruits of advanced site search and continually sees the growing results, which is made possible due to the continuous search improvements process utilized by SearchNode.
Irrelevant Search Results
Their previous eCommerce search engine was only retrieving data from titles and categories. It did not consider such things as descriptions, specifications, bullet points, and other important fields. As Merkur.si catalog consists of many similar goods, it was tough for the company to provide its users with relevant search results that would transform into a pleasant customer experience and high conversions from site search.
Poor Slovenian Language Processing And Understanding
The previous site search solution struggled with understanding the complex nuances of the Slovenian language, which caused poor customer experience, and low conversions as well.
Looking for solution
Merkur was unsatisfied with their site search performance, and they wanted to improve it. The company was growing fast, which made search problems more and more apparent.
Merkur team did extensive market research to evaluate different site search providers. They considered all the options on the table: from enterprise search solutions and open source technologies to SaaS providers and a hybrid between SaaS and the consulting partner – SearchNode.
Ultimately, SearchNode won over the competitors. Merkur chose us because of our unique offering of combining advanced search solution with dedicated search experts and continuous search improvements. Moreover, SearchNode has good references in providing search for home improvement eCommerce stores.
“ Search engine is one of the most important parts of the eCommerce site. Changing the solution can give you a quick win in terms of several sales metrics. We decided to work with SearchNode because of the excellent solutions they offer, saving time handling the search, short response times from the Searchnode team, and it works perfectly with the Slovenian language. “
Ecommerce Manager at Merkur.si
The implementation of SearchNode
We have an established process of SearchNode implementation, so that clients need to dedicate the least possible development time and effort. And Merkur.si was not an exception. The full tailor-made site search solution was launched in seven weeks from the initial kick-off workshop.
The Merkur search team only needed to communicate their desired goals and features for a site search solution. We also discussed a lot about their business specifics, priorities, and objectives to ensure that SearchNode operates in the most profitable way for Merkur.si. We were also in communication during the whole process with the search team to gather appropriate feedback in time. From the technical point of view, they only needed to provide data export API according to our requirements. SearchNode did all the rest – from data analysis and optimization, tracking scripts to the creation of search user interface and custom AI algorithms.
Our search experts keep working together with Merkur as an integral part of their search team, operating as if the company would have internal search engineers.
Solutions created for Merkur
Enhanced Relevance Engine
Data Relevance Algorithms
We supplied Merkur with SearchNodes’ powerful AI search platform, with advanced eCommerce search features and capabilities. However, to achieve the best possible search relevance level, an advanced platform is not enough. Thus we built custom data relevance algorithms for the company.
Our search engineers dived deep into analyzing Merkur product data and user behavior and took into account their business objectives. By doing this, our team was able to come up with data-driven insights and tailor custom algorithms, individually designed only for Merku.si specifics.
We implemented it on top of our core search platform, and this allowed us to create the most relevant search experiences, optimized for maximum revenues.
Even though this instantly generated significant growth in Merkur.si search relevance, this is not where improvements ended. Our solution engineers continue to develop and advance these custom algorithms, which leads to scalable search improvements. Our unique approach of combining AI search platform with company-specific algorithms allow Merkur to have a truly relevant search, optimized for business goals regardless of the circumstances.
Natural Language Processing
To process the Slovenian language correctly, Merkur needed a sophisticated search solution. To solve this issue, we introduced custom stemming and natural language processing (NLP) algorithms based on Slovenian language specifications and grammar rules. With the support of artificial intelligence, Merkurs’ search solution can now process words written in any form and better understand the meaning and context of user queries and return the most appropriate results at any given time. This helped to create a more precise and intuitive search and lift Merkur search relevance significantly, resulting in improved customer satisfaction and experience.
Implementing dynamic filters and facets to Merkur.si search was another solution, which significantly improved customer experience. This way, they don’t need to spend too much time browsing through different categories or results pages. Even more, our solution, enriched with machine learning, automatically analyzes the context behind each query. Based on that, it displays only relevant options. Now Merkur users are empowered to refine their search, as they like swiftly and effortlessly.
Proactive Search Assistance
Another great source of revenue for Merkur was well constructed autocomplete. SearchNodes’ smart query suggestions render keywords from Merkur.si product data to make conversion-optimized product suggestions. Similarly, it shows queries inside product categories and makes intelligent suggestions for all content types, simplifying the discovery process. With the help of an intelligent autocomplete solution, Merkur.si can now enjoy a highly-converting search experience.
Dynamic autocomplete suggestions
Based on Merkur’s request, we created and implemented dynamic autocomplete suggestions. On the right side of the autocomplete, customers can see smart query suggestions. On the left side, product suggestions are displayed, which is changing when the user is hovering the mouse over different suggestions on the right side. This way, Merkur.si customers can easily browse and access the desired product right from the autocomplete.
Merchandising Beyond Search
We implemented sophisticated search results merchandising solutions for the company. During the kick-off workshop, SearchNode and Merkur teams agreed on rules and strategies for a custom merchandising solution to help the company achieve its goals. The developed solution combines Merkur merchandising logic and sales data from Google Analytics with SearchNode search relevance algorithms. This allowed our engineers to tailor intelligent merchandising algorithms that automatically adjust results for optimal conversion rates.
“ Due to the fact that www.merkur.si is operating with more than 64.000 products in more than 1.400 online categories, SearchNode helps users to find products with ease. It provides a better experience, and due to this fact compared to the prior solution, more visitors use the search engine, saving time scrolling through the categories. “
Ecommerce Manager at Merkur.si
Continuous Search Improvements
As we mentioned in the previous part, our work doesn’t end when a solution is live on clients’ websites. The team of search experts from SearchNode keeps proactively working as part of the eCommerce search team in the company, striving to make the site search better each day. With Merkur, we have one sprint of continuous search improvements each quarter, divided into two parts.
1st Part: Ongoing Search Maintenance
We commit a fraction of sprints’ time to maintain the search intact. Our search engineers ensure that all the data is updated promptly during this part and that all the bugs are patched immediately. Similarly, they spend some time introducing user interface modifications and carrying out unforeseen or urgent tasks.
TYPES OF FIX
- Quick fixes
- Minor UI changes
- Promotions, landing pages, redirections
- Product priority adjustments
- Both: Merkur team or SearchNode team
- To quickly fix individual cases
IMPACT TO SEARCH METRICS
- No significant improvement
2nd Part: Scalable Search Improvements
The second part of the sprint is devoted to implementing new search features or creating new functionality. On a monthly basis, search teams from SearchNode and Merkur develop a list of possible search improvements and hypotheses for testing. Then, Merkur’s product owner analyzes and prioritizes the list items to select what should be done in that sprint. A new improvement is released in the form of a test so that SearchNode engineers could observe and analyze which of the variations is performing better and act accordingly.
TYPES OF FIX
- Relevance algorithm improvements (fixing a pattern of issues)
- Ranking strategy improvements
- Merchandising automation
- New search features
- User behavior analysis
- A/B testing
- Merkur manages priorities and sets goals
- SearchNode helps to identify nonperforming areas and raise hypotheses
- SearchNode carries the implementation
- Making impactful search improvements
- Reducing the need for manual work
IMPACT TO SEARCH METRICS
- Search metrics rise over time
Continuous Search Improvements Preserves Growth
As we continue working together with Merkur as a single search team, the site search is continuously improved in the most scalable and efficient way. This process ensures that the search is always optimized for both – Merkur and its customers. It allows the company to achieve the best possible business results while providing an amazing search experience for the clients.
SearchNode engineers develop custom algorithms specific only to the Merkur case in each new sprint. Thus, we enrich our AI-based search platform with custom algorithms, maximizing Merkur site search potential and company revenues.