#2 Mastering E-Commerce Search

Find the e-commerce search that’s right for you

– For the SAP Commerce Cloud or other e-commerce systems –

In the first article of our series “Mastering E-Commerce Search”, we highlighted current trends in search, the various functional depths and the potential of a good search. Now let’s move on to the crucial step: choosing the right search solution for your business. But as so often in life, it’s not that easy. How do you find the e-commerce search that’s right for you?

This article is intended for all e-commerce companies who are thinking about using modern search. Either because you want to transform an existing commerce platform into a powerful sales machine, or because you need an assessment when implementing a commerce system with an integrated search solution. You might question youself:

Is the provided standard search enough for my business or should I replace it, with a more specialized search right from the start or in the near future?

Codeitlabs has its genes in consulting and development around the SAP Commerce Cloud (formerly hybris), so we will occasionally refer to the SAP Commerce Cloud. However, since we are also implementing MACH-based e-commerce systems, this article is written in a generic way and thus offers you to find the e-commerce search that’s right for you.

Is the provided standard search enough for my business or should I replace it, with a more specialized search right from the start or in the near future?

More than just a search field: Modern searches are sales engines and ensure enthusiastic customers

A search solution is a crucial building block of your e-commerce system, if you want to offer your customers a great CX and increase your sales. The search represents the key component between all data and the front office of your systems, and thus controls the behavior of your customers.

A good search not only offers your users the opportunity to search for specific content but can also personalize structured list views such as categories or content such as stage designs. The entire shop can be adapted through a modern search, right down to changing the look & feel. By integrating intelligent, dynamic search filters, the system can further refine the search in order to provide your customers with the perfect search result.

Before we examine the range of functions of various solutions, you should first take a look at your company and your requirements! This is the only way to find the right search for you.

Identify your individual requirements!

“It depends!” The most classic of all consultant phrases. Before you start looking for a solution, you should take the time to understand your requirements and goals. If you have a vision or mission, these can help to define your requirements and derive your goals for a search solution. It is also crucial to know what you want to achieve specifically with an improved search? Which functions are essential for your specific business model? What is the current search behavior of your customers and which KPIs do you want to improve? You should also consider what resources you have available, both in terms of budget and the technical skills and availability of the necessary colleagues.

This is how you find the right e-commerce search for you

Comparable company profiles, in which you may find yourself, can help you as a first aid for approaching the topic:

icon for start-up level/low business requirements

Start-ups or smaller companies

Your company has a low product density and less complex catalog structures. Maybe you have little budget and/or resources? A basic search solution might be enough to provide your customers with a satisfying search experience not to overwhelm your capabilities. If you are not looking for an alternative to the often sufficient search solutions that are provided, you will find those standard solutions on level 1 of the Evolution Pyramid of E-Commerce Search (might also include other solutions than the ones depicted in our pyramid).

icon for mid company level/average business requirements

Medium-sized companies

As a medium-sized company, you have a higher product density and a more complex catalog structure. In this way, your customers can benefit from solutions with a medium range of functions. These provide improved search accuracy and the ability to personalize search results. They also offer other important functions such as autocomplete, autocorrection or synonym searches. You can work with a certain budget and have a small team that covers the internal issues and has the necessary know-how. Solutions from the second and third levels of our pyramid could fit here.

icon for big enterprise level/high business requirements

Large companies or companies with high product density

Your company has (ten) thousands of products and complex catalog structures. You have a large website with a lot of content, so your customers can benefit from the full potential of a very good search solution. These solutions offer, for example, greater personalization based on user preferences and behavior and additional functions that can be used to optimize or personalize your shop in a wide variety of places. So they offer good CX. You have sufficient budget and competence to connect various systems (internal and/or external). Here you will find the right solution on the fourth level of our Evolution Pyramid of E-Commerce Search.

icon for any company with high business requirements

Companies that strive for excellence in search

Regardless of the size of your company, you are a company that is in, or aspires to be, a leader in its market segment. You need the search solution with the best possible range of functions, as offered by high-end AI-supported models. Through machine learning, they constantly adapt their behavior and thus automatically offer the most relevant search results and a wide range of functions and options to turn your shop into a high-performance sales machine. Your customers should not search, but find. You really care about the customer experience (CX) of your customers. Of course, the budget and the effort are important to you, but the expected ROI puts the effort and costs into perspective. Here you consider the value of loyal customers through great CX.

Consider costs

As already noted in the company profiles, the costs for implementing a search solution, as well as the license costs, can vary greatly. You should not underestimate the internal costs involved. These costs depend on the chosen solution and the complexity of your requirements. So it’s important that you not only have a clear understanding of your budget, but that you balance the expected costs against the expected benefits.

Integration and support

The integration of the search solution into your existing technology is an important aspect.

Some solutions are easier to integrate, e.g. into the SAP Commerce Cloud, than others. You should also take into account the available support and the update policy of the provider. Speaking of integration…

How does the Solr-based search of the SAP Commerce Cloud differ from the standard Apache Solr search

SAP has implemented the search of the SAP Commerce Cloud based on Apache Solr and optimized it with extensions typical of e-commerce. The most important points that speak for the SAP CC Solr search are:

Integration into the SAP Commerce Cloud:

This is probably one of the most important reasons why many SAP Commerce Cloud customers choose this variant in the first place. SAP has seamlessly integrated Apache Solr into the Commerce Cloud. This means that data can be exchanged efficiently between the two systems, resulting in quite fast seek times and a good CX. First of all, there are only small (specific) expenses for the search. And there is no need to evaluate an additional solution or conclude an additional contract. Furthermore, further developments and updates of the Solr core – Apache Solr is an open source project – are regularly transferred to the SAP Commerce Cloud.

Compared to the “standard” Apache Solr search, the SAP CC search offers various optimized e-commerce and business features. The most important are:

Configure search profiles with relevance adjustment, e.g. rank certain categories or products higher, but also rule-based through scoring. For example: Show available items (Score 1) that are on sale (Score 2) higher up in the results list. Search profiles can also be configured user-specifically.
Configuration of search filters (facets). Users can filter the search using these parameters.
Configuration of auto-completion, users get letter-by-letter suggestions and also preliminary results for the search term
Suggestions for “Did you mean”
Configurable fault tolerance. Typing errors still lead to the right search results

Configuration of search filters (facets)

Another important advantage can be found in support and maintenance: Thanks to the integration with the SAP Commerce Cloud, companies receive support and maintenance from SAP for the search itself. This means that a professional support team is available in case of problems or questions regarding the search stands.
Last but not least, the Commerce Cloud largely frees the interface to Solr – so that individual extensions – given a solid Solr know-how – can be made in the project.

Taking these advantages into account, the SAP CC SOLR search can even be classified on level 3-4 of our pyramid. For a higher level, you lack advanced personalization functions and AI or ML functions. However, it must also be mentioned, that all functions in the SAP Commerce Cloud – and thus also the search – can be optimized. In this context, SAP customers have to ask themselves the classic “make or buy” question.

Incidentally, SAP announced COVEO as an “endorsed SAP partner for search” in Q1 2023. We currently rate Coveo as the most advanced search solution. That’s why we sort Coveo into the 5th level in our pyramid as the only solution. Closely followed by Algolia.

Note: We will deal with the integration of search solutions in the SAP Commerce Cloud in a separate article in our series.

At a glance: Overview of current search solutions in our Evolution Pyramid of E-Commerce Search (EPoES)

The following graphic sorts common search solutions by functionality, relevance of the search results, possibilities for personalization and learning ability, NLP and adaptability of the search solutions. Please note that this is not a linear sorting, but that the solutions can only be sorted approximately.

With the appropriate effort, many of the searches can be optimized. When sorting, we assume the highest quality variant with normal adjustment effort.

The systems considered here continue to represent only a selection for the various scenarios. For most of the levels of our pyramid, there are other solutions that can be evaluated.

As already mentioned, in the end your requirements and possibilities are the decisive factors on which your decision for your search must be based. This is the only way to find the right e-commerce search for you.

Sorting examples of search solutions into the different levels of E-Commerce search

Functional overview of 5 exemplary search solutions:

The following overview will help you to review and evaluate individual functions and criteria for your evaluation. *Note that the individual solutions have different characteristics (Basic, Advanced and Pro, etc.). As already mentioned above, some functions can be further developed in different solutions or have to be configured first. Each software provider may have different features for the following functions.

Accordingly, we can only give a rough guide:

AI-powered search and ML

	Elasticsearch	Factfinder	SAP CC Solr	Fredhopper	Algolia	Coveo
EPoES Level	2	3	3+	4	4+	5
Function
Simple full-text search (filter/sort)	X	X	X	X	X	X
Correct spelling mistakes	X	X	X	X	X	X
Recognize synonyms	X	X	X	X	X dynamic	X
Faceted Search	X	X	X	X	X	X
Autocomplete search query	X	X	X	X	X configurable built-in features/AI-powered	X configurable built-in features/AI-powered
Simple personalization functions (relevance/ranking depending on user preferences)		X	X via SAP`s adaptive search module	X	X	X
Advanced personalization features (depending on user behavior)			X rule-based	X rule-based	X rule-based with suggestions	X rule-based & AI-powered
Natural Language Processing NLP					X	X
AI-powered search and ML				X	X	X

Also note that, except for Apache Solr, all the solutions listed here are Search as a Service software. Thanks to the simple user interface, automation and configuration options, even smaller companies or companies with lower search volumes can deliver better, more relevant search results for their end customers in a short time and with little effort.

If you have any questions about search solutions in general or you are about to evaluate a search solution and need some further advice, please do not hesitate to contact us.

In this article we hope we could help you to find the e-commerce search that’s right for you.

The next article in our series “Mastering E-Commerce Search” continues with a deep dive into the SAP CC search based on Apache Solr, followed by deep dives into Fredhopper and Algolia/Coveo. Later we will continue to publish reference interviews and best practices for integrating search solutions.

So stay tuned!

Good luck with your search!