Categorization of Search Engine Results

Case Study categorization of the relevance of search engine results

clickworker evaluates thousands of search engine results with regard to the relevance of the results produced when a search term /keyword is entered.

The Challenge

Many reference and directory media as well as e-commerce companies integrate their own search engines into their websites.
The search engines must be adapted to meet the needs of the respective user.
If the “human intellect” factor is missing in the algorithms that the search engines are programmed with, the relevance of the search results, the so-called search relevance, is often unsatisfactory. The results do not meet the expectations of the user or disregard the local relevance.
As a result of the large amount of data in search engines the operating company usually cannot check the relevance of the search engine results with internal resources.

The Solution

Our worldwide team of Clickworkers examines thousands of search engine results according to previously determined search terms. They evaluate the results with local market knowledges according to their actual or subjective relevance.
These evaluations flow into the search engine algorithms and optimize the quality of the search results.

Project Data

Clickworker qualifications: Native speakers from the target regions, demographic factors.

Typical number of daily evaluations: > 4,000

Languages: All standard European languages.

Quality assurance: Evaluation per scalable majority decision, injected testing.

Data transfer: CSV file via email or API.

Project / minimum volume: > 20,000 evaluations / categorizations.

Work Flow

  1. The customer determines the order conditions and the evaluation criteria.
  2. Depending on the specifi cations the customer can establish which results and in what form the search engine results will be evaluated.
    a list of search terms, for example a business sector in combination with a place/ a locality + factors potential search engines (details about distance, business sector, etc…) that must be evaluated with regard to their relevance to the search term combination.
  3. clickworker draws up the project, divides it into individual microtasks and puts the project online where it can be accessed by qualifi ed Clickworkers for processing.
  4. The Clickworkers check the quality of the search results and evaluate them on the basis of the (evaluation) criteria according to their relevance.
  5. All of the search results are evaluated by two or more Clickworkers simultaneously and when they are consistent the result of the task is sent to the customer.


  • Human intelligence
  • Local market knowledges
  • Speed – Quick categorization of search engine results
  • Scalable throughput, flexible workforce
  • API integration