Editorial
Over the last two decades, academics have produced an incredible body of research into how we search the web. Thousands of research papers, hundreds of books and reports, and of course, many conferences have been devoted to the topic. Yet when it comes to research into enterprise search, it has only been this year that papers have been published that begin to shed light on how employees use enterprise search applications.
Undertaking research into enterprise search use is fraught with challenges. A group of 10 employees might end up with 10 different sets of results as a result of security trimming and personalization, both of which they — and the research team — will have no knowledge about. Issues of confidentiality and timing also thwart research, as search use patterns cannot be derived from click data when there is no context for the query.
In the face of such a paucity of research into enterprise search, how is it that the search industry professes to understand what users want?
The Factors Impacting Search Satisfaction
Paul Cleverley and Simon Burnett undertook a two-year research project which revealed the origins of search dissatisfaction within a single multinational company in the oil and gas sector. The technology factors included search tool reliability, search ranking and query syntax handling. In total these factors were the largest single group (38 percent), which some could see as a justification for investing further in search technology. However information factors (36 percent) and literacy factors (26 percent) combined accounted for 62 percent of the reasons for dissatisfaction, which to me indicates that technology investment on its own will not make a significant difference to search satisfaction.
The search application was used by around 70,000 staff each month and generated over 450,000 search queries. The average query length was 1.89 words and the top 30 most frequent searches fell from 14 percent of all search queries at the start of the project to just 8 percent by the end of the project two years later, when users had gained more experience with the application. This reaffirms my anecdotal evidence of the long tail of low frequency queries in the enterprise environment. This has clear implications for "cognitive search." When the majority of queries produce such low levels of use data, predicting optimal results will be challenging at best. Although this case study was in a particular sector, I see no reason for not scaling it to most multinational corporations outside of professional services.
Related Article: The Next Generation of Enterprise Search Is in Sight
The Spontaneous Habits of Search Tasks and Work Tasks
Miamaria Saastamoinen and Kalervo Jarvelin at the School of Information Sciences, University of Tampere have been considering how work tasks (WT) in an organization give rise to search tasks (ST). In a recent paper they note the concept of task-based search dates back at least a decade, but the research has not been carried out within the environment of an organization at work. They published a paper in 2016, at an earlier stage of their research, which stated:
“One of the main points was that the researcher identified the endogenous, spontaneous STs rather than setting an experiment. IR [information retrieval] is an integral part of the flow of WTs. People do not normally start working search-minded. It may be that STs could be afterwards clearly identified by researchers or even by the participants themselves; but IR is seldom a part of a work plan. Without a doubt, this is a feature dependent on work role, profession and task. However, the search actions in the present data were mainly quick (often nearly inadvertent) choices made between, for example, browsing, calling a colleague, looking up in a book, delegating, etc.”
Related Article: The Scent of a Good Search
Learning Opportunities
Webinar
May
30
ChatGPT and the Future of Conversational AI
This discussion explores various aspects of AI implementation, ethical considerations, and challenges for your organization.
Webinar
May
31
The Evolution of Employee Listening - 2023 Trends & Best Practices
Learn proven effective methods and solutions to supporting an inspirational future of Employee Voice
Webinar
Jun
1
How organizations can design, build and run Generative AI into the Customer Experience
Discover how to drive enhanced CX through Generative AI
Webinar
Jun
6
The Future of Social Media: Emerging Trends and How to Stay Ahead of the Curve
The impact of emerging technologies like AI on social media
Webinar
Jun
7
The Building Blocks of Personalization
This webinar will explore how personalization can create powerful ecommerce shopping experience.
Webinar
Jun
13
The Future of Contact Centers with AI at the Helm
Learn how to balance human intuition and interaction with the potential uses of AI to create delightful and lasting CX
Webinar
May
30
ChatGPT and the Future of Conversational AI
This discussion explores various aspects of AI implementation, ethical considerations, and challenges for your organization.
Webinar
May
31
The Evolution of Employee Listening - 2023 Trends & Best Practices
Learn proven effective methods and solutions to supporting an inspirational future of Employee Voice
Different Levels of Search Skills: Professional vs. Operational
Over the last few months, perhaps as a result of a new program on the topic at SIGIR in July, attention is being focused on how professional searchers work. Professional searchers work within specialist organizations, but also within a corporate setting. Think of lawyers or patent agents as an example. They both spend a significant part of their working day seeking information, and possess search skills as well as subject expertise. One of the innovators in this area is Tony Russell Rose and it is well worth reading his introduction to the workshop. Sadly I suspect few professional searchers will make the trip to Ann Arbor, Mich. for the conference.
What should we call everyone else who uses enterprise search applications? I would call them Operational Searchers. They use search on a mainly ad hoc basis, within the context of a work task and may have poor search skills and perhaps limited domain expertise.
Related Article: Diagnosing Enterprise Search Failures
'We Know What Users Want'
The search industry is very confident in its knowledge that users want one precise piece of information in response to a query, driven by AI and machine learning within a personalized search experience. To my knowledge, no published research exists that supports this supposition, which also seems to be one of the drivers of the chatbot business. Vendors may have conducted their own research and succeeded at overcoming the immense methodological challenges identified by Saastamoinen and Jarvelin. If so, it would be a service to the search community to publish such research.
In the meantime, we are fortunate that the research community recognizes the intellectual challenges and business value of looking in detail about the process of enterprise search.
Learn how you can join our contributor community.
