Search Engine Query Reporting
Client
LINK4
Industry
Insurance
Technologies
Drupal
ElasticsearchScope of work
Search analysis mechanism implementation (phrases registration and normalization, reporting, and export)
What Challenges Did the Client Encounter?
The website users searched countless phrases daily. However, the lack of search query analysis hindered access to a key knowledge source about customer needs and interests. Limited insight into user expectations led to constrained offer optimization.
This also meant that key questions remained unanswered:
- what users are actually looking for,
- which topics evoke the greatest interest,
- which query results are suboptimal or non-existent.
How Did We Help the Client Understand Their Users?
To solve the problem, we designed a special mechanism on Drupal 10. It enables data analysis, which reveals users’ real needs without invading their privacy.
The solution saves each phrase in its original form while also ensuring proper normalization. They are standardized as well, so different variants (e.g., Polish characters and declension) are treated as a single query, rather than several separate ones. This provides the client with clear, organized data. Importantly, users do not have to worry about their data – it is fully anonymized.
In this project, we used quite standard technologies such as Elasticsearch and Analyzer. It is the combination that is unique. The solution enabled us to provide the client with precisely tailored phrase normalization, usage records, and reporting.
How does it work in practice?
- The user enters a phrase, e.g. “insurance”, into the search engine.
- The system saves it along with the search date – anonymously, without the user’s data.
- Different variants of the same phrase are automatically unified.
- Data is sent to a legible report, so the client can use it to analyze user expectations.
Phrases can be filtered, among others, by date (from-to) and by version (normalized and original).
How Did the Client Benefit?
Our solution provided the client with access to key data that was previously unavailable. It enabled:
- monitoring the most frequently searched queries and filtering them by date and phrase version,
- a better understanding of user needs and expectations,
- easier trend analysis,
- optimizing search results and offering efficiently,
- adapting website content to users’ interests.
We wanted our solution to simplify administrators' work. We implemented automatic creation of phrase and usage entities. It allows everything to run in the background. Most important from the client's perspective, however, was the visual presentation of the collected data, i.e., the most frequently searched phrases view. We ensured it included full information about the searched phrase and its standardized version. This gives the client full control – for example, they can delete a single phrase usage.
When Is It Worth Using This Solution?
When you want to understand seasonality and trends in your industry better.
When you need to adapt menus, categorization, and filters to user behavior.
When the search result accuracy needs refinement.
Want to know what users look for on your website?