man looking around the corner at the top of a stairwell
PHOTO: Joseph Akbrud

Throughout our working lives there will be many occasions when we realize there is a gap between the information we have and the information we need. Our search for that information doesn't always mean we turn to an enterprise search application. We could contact a colleague, browse through the intranet, post a request on social media, track down an expert, look through documents and data or put it on the agenda at the next team meeting. These are all examples of information seeking, of which search is just one option.

Any attempt to follow how employees go about that process of seeking, and the role search plays in it, have always come up against a memory challenge. If asked at the end of a working day how they found information, people will never be able to recall every instance and its outcomes. Checking the post code for the office in Budapest might have taken no more than 20 seconds using the corporate website and becomes almost a subliminal act.  

Related Article: Enterprise Search: An Invaluable, If Sporadically Used Application 

Computational Ethnography Sheds Light on Work Behaviors

The term ethnography originates from Greek ἔθνoς ethnos (“folk, people, nation”) and γράφω grapho (“I write”). It describes a method initially used by social science researchers to closely examine the meaning in the lives of a cultural group. Carrying out ethnographic research brings with it such challenges as the amount of time and effort needed to undertake the observations and interviews, the possibility of a bias being introduced by the research team and the scale of the analysis of the outcomes.

Over the last few years computational ethnography has emerged as a powerful means of overcoming, or at least reducing, the challenges of interview-based ethnography.  Computational ethnography leverages computer or sensor-based applications to unobtrusively record end users’ routines as they undertake work tasks. It can provide higher objectivity and less intrusion when direct observation by human observers is not possible, and better scalability for data collection, aggregation and analysis. Among these applications are keyboard and screen-logging software developed to mitigate the loss of sensitive data.

A critical limitation of computational ethnographical methods is that although they may tell what happened, they are often incapable of shedding light on why employees demonstrated the observed behaviors. Computational ethnography needs to be integrated with qualitative research methods such as interviews, diaries and observations.

Related Article: Unpacking the Complexities of Enterprise Search Behavior

Differentiating Work Tasks and Search Tasks

Professor Kalervo Jarvelin of the University of Tampere in Finland one of his PhD students, Miamaria Saastamoinen, pioneered the use of computational ethnography in the quest to understand how people perform information seeking in the working day. They logged the use of digital resources in the workplace, including email, website search, a PC and internal search, rather than relying solely on diaries or self-completed surveys. The analysis also captured the time spent searching. The figure they arrived at was far lower than the two hour 30 minute figure that IDC came up with almost 20 years ago which is still used as a core business case for time saving through "faster search."

A very important outcome of this work is the differentiation of work tasks and search tasks. Each work task may involve many search tasks. A lot of research has been carried out on search tasks over the last couple of decades but what Jarvelin and Saastamoinen highlight is the importance of understanding the context of a search task.

Related Article: Diagnosing Enterprise Search Failures

Search in Context: Professional Search

At an extreme end of the spectrum the work task is a search task. Lawyers, medical researchers, patent agents and recruitment agents all spend a significant amount of time undertaking search tasks which generate revenue. A comparative study of how these four professions go about their work has now been published, presenting a very detailed picture of the search functionality that each profession regards as important to their work. One interesting outcome is the heavy reliance on Boolean query development. As with the Finnish research, the study tracked the time spent on a search, illustrating a very wide range of values.

Related Article: The Clue to Your Digital Workplace Problems Lies in Your Employees' Work-Arounds

It Comes Down to Contextual Understanding 

In my opinion, the low levels of search satisfaction revealed in surveys from AIIM, Findwise and NetJMC is in large part due to the focus of both search vendors and IT departments on the process of searching alone, without including a concomitant understanding of how users are seeking information and the context of why they are then using search. For the same reason, the vision of providing personalized search based on tracking the information employees create is fundamentally flawed because the relevance of a document depends on so many factors that are not capable of being tracked digitally. 

The promise of computational ethnography will only be achieved by integrating the outcomes with a wide range of other research methods, and that is why you need a search center of excellence. As Daniel Tunkelang has observed, "Ultimately, your search will only be as good as the people who work on it."  

Related Article: Does Your Business Need a Search Center of Excellence?