Have you searched for something today and not found it? To ensure that our customers in the corporate context no longer encounter this issue, we develop customized search solutions tailored to the type and volume of data and search queries. We rely extensively on open source and cloud technologies and guide clients from data preparation to user interface. Additionally, we offer consulting services on how to best handle the topic of search in your organization.
Good Search User Experience
In today’s digital age, we’re drowning in an avalanche of information and data. That’s where search applications come in. When we talk to people in the business world, they often express a desire for something as simple as Google but for their own data – just a search box. It sounds straightforward, but there are many challenges and nuances to consider. In most real-world situations, people need more than just a search box. They need specific ways to interact with it to find what they’re looking for in order to complete their actual goals. Our whitepaper Good Search User Experience is a summary of lessons we’ve learned over the past decade by creating excellent search experiences.
Download now
Why we search
The question may sound simple, but the answers to it are often diverse - even in a corporate context. One significant challenge often lies in the natural language in which documents are written. The same terms can describe different things, and cryptic abbreviations often render content inaccessible. Furthermore, the volume of documents and the multitude of different storage locations often lead to searching for needles in haystacks.
The Redlink team has decades of experience in the field of search and search strategies and the underlying technologies, from classical information retrieval to semantic search. As your partner in search expertise, we offer businesses an ideal blend of strategic consulting and technical solutions.
Technology
In implementing your ideas, we employ a wide range of technologies. When it comes to search and indexing, our solutions primarily use Apache Solr and Elasticsearch. We leverage the features of classical information retrieval combined with vector search.
For natural language processing, we rely on state-of-the-art components such as StanfordNLP and BERT. In ongoing operations, we analyze log data and interactions using common tooling (e.g., ELK). These data serve as the foundation for personalization and contextualization. Additionally, these analyses provide valuable insights into the success of employed UX patterns and the quality of the search algorithm.