Unfortunately though, the importance of one’s own product search often takes a backseat. Although companies often have specialists for product maintenance, marketing measures and pricing, they rarely have an expert in search and results design.
Therefore, when it comes to areas of improvement, many other disciplines are thought of before searching: “Is the product data extensive enough?”, “Is the home page perfect down to the last pixel?”, “Do we collect enough data about purchasing behaviour?” … But hardly anyone asks: “Are our search results relevant?”.
With Apache Solr, SAP Commerce offers a powerful and fully integrated tool to deliver not only fast, but also meaningful search results. Apache Solr is a de facto standard on the market for the implementation of faceted free text searches. Based on the Lucene engine, Solr offers important functions from the get-go: fuzzy search, result scoring with freely configurable boost option, adaptation of terms (e.g. reduction to word stems) at runtime or preparation of the search result and much more.
Correct handling and understanding of SAP Commerce search integration are essential to quickly guiding the customer to their objective with their search request. Unfortunately, a lack of expertise among the implementation partners usually leads to the purchase of expensive, third-party solutions that often cause problems with data recency and integration.
For example, this means that adaptive search is not used and there is no option to personalise searches according to customer segments, with all of its subtleties from the SAP Commerce portfolio.
The planning and ongoing configuration of the search is characterised by the following questions:
- Regardless of what buyers and product managers think, what kind of products does the customer want to find?
- What is expected of the search result: lots to look through or a few precise results?
- Is the relevant data on the product also part of the search and is it more suitable for free text or faceting?
- What alterations do I have to make over time for fine adjustment?
Answering these questions can be a difficult task, as the procedure for establishing and optimising the search requires experience and expertise in dealing with SAP Commerce, Solr and Adaptive Search. It is also important to build up the necessary expertise in your own company, which requires assistance from a partner.
For example, the quality of search results is often criticised: there are too many products in the results or products that are of no relevance end up too high in the results, while metadata seems to have no value in the weighting. However, this is not due to an inadequacy of the search engine used, but is because the metadata and the documents prepared in the search index have not been fine-tuned to the product range.
Often, the name of a product is given far too much relevance in the search results, while attributes which can take time to maintain (e.g. metadata on colour) are not even made known to the search engine or have a negligible factor in measuring the importance of a result. The goal here is to carry out an inventory of the current configurations and to work together on optimising the data that is going to be used and to work on measuring its relevance or make individual changes.
In addition, the challenge is to break down metadata in such a way that makes it searchable in a form that is more like natural language and not technical and convoluted. Here, too, it is important to have a partner who can tailor the search engine by making individual changes.
We are happy to support you in making your search not just a tool, but a central hub for your ecommerce using our tried and tested procedure. With the best possible searchandising functions in your SAP Commerce platform, we get the best results for your customers!