Businesses that fail to provide effective ecommerce tools – including site search – risk a loss of sales to competitors. There is often a disconnect between those who maintain site search and those who understand customer needs and behaviour. This article describes a ‘measurement, experiment, repeat’ approach and introduces a suite of open source tools (‘Chorus’) that supports rapid experimentation and implementation and which gives control of site search back to search teams.
Search data is key to understanding customers. This article explores the competencies and roles required to gain actionable insights from customer search data. Four broad competencies are required: data, engineering, product and business. The roles required for each competency are described. Data from successful search teams is used to show levels of involvement from each competency.
Taxonomies are business assets, built to help the organisation achieve beneficial outcomes. This article explores the key stages of taxonomy projects, describes what can go wrong in each stage and suggests ways to avoid taxonomy project failure. This includes the critical importance of ensuring stakeholders understand the value and impact of a well-designed and managed taxonomy, and the need to take a long-term view of maintenance.
Knowledge workers (e.g. healthcare information professionals, legal researchers, librarians) need to create and execute search strategies that are comprehensive, transparent, and reproducible. The traditional solution (command-line query builders offered by proprietary database vendors) dates from the days when users could access databases only via text-based terminals and command-line syntax. This article presents a new approach in which users express concepts as objects on a visual canvas and manipulate them to articulate relationships. This offers a more intuitive user experience (UX) that eliminates many sources of error, makes the query semantics more transparent, and offers new ways to collaborate, share, and optimise search strategies and best practices.
While research shows that different types of searchers interact differently with search applications, vendors are unlikely to offer an interface that is optimised for specific tasks or customisable by the user. Examples of how the user experience can be improved include ensuring results are presented for ease of scanning and reviewing and providing a well structured and presented snippet. Vendors that focus on customisable interfaces will find it easier to differentiate their products in the marketplace.
Search auto-suggest -- the drop-down menu of suggestions displayed while typing into a search box -- is the first and often the only search interaction that users encounter. The goal of auto-suggest is to show users possibilities related to their typing, and provide them dynamic context information about what is available. This article explores creating auto-suggest indexes, matching, retrieval and ranking, handling zero matches, and user interface options, with considerations and trade-offs for various approaches.
Effective search requires more than a technology solution. This case study shows that poor quality content reduces findability and negatively impacts user satisfaction. Although there is no quick fix and each organisation is different, a content inventory followed by work on content type and lifecycle, metadata, permission and accessibility, information architecture and information curation will form part of the project plan.
This article explores the key roles and skills required in an effective in-house search team. The key roles to cover include search relevance strategist, search relevance engineer, data analyst/scientist, content owner, metadata owner. Several approaches to skills development are outlined, including external partnering. Your ultimate goal is to build your organisational capability so that your search function is owned and managed in-house.