Taxonomies for business. (Software World).
A report published by TFPL analyses the Taxonomies practiced by some twenty one major companies, including Microsoft Inc, ICL, DERA, BBC, and others in a case history forum. The work funded by TFPL and Smartlogik provides a useful illustration to the manner in which Taxonomies are applied in a variety of different industries. Considerabk discussion is applied to defining a Taxonomy by a process of argument rather than by definition of a set of rules. The following analysis also includes our interpretation of the characteristics of the Taxonomy, prompted by this report. These are included in italics to avoid confusion with the authors statement as they may not be wholly exact.
Defining Taxonomy (Authors) The word Taxonomy is derived from two Greek roots: 'taxis' meaning arrangement, primarily the distribution of forces before a battle and 'onoma' meaning name. Whilst taxonomy development may sometimes feel like a battle, in our view it is the creation of structure (arrangement)and labels (name) to aid the location of relevant information. This is a very broad definition. A closer definition might be the arrangement and labelling of metadata(data about the data being handled) which allows the primary data or information to be systematically managed and manipulated.
The Taxonomy Problem (Editor)
Comment: For practical purposes a taxonomy should classify and position knowledge in such a manner that its retrieval can effectively match a specific demand to a high degree of relevance. However, the impression given by naming a business system a taxonomy, is that it is defined as a deterministic entity, from which a repeated query will be given a repeated identical solution. This may be difficult as knowledge terms may have varying] meanings, depending on the state of the organisational culture of the organisation at thai time or moment. In particular, the very meaning of words, particularly in the IT industry, can change radically in a short time as they are adapted to describe quite new soft ware processes. This has to be taken into account in the development of the taxonomy.
As an example the Accenture case study in the report says: In an organisation as diverse as Accenture, there are a huge number of different technical views of what staff do, and the information they use: the same material would need to be tagged very differently in say tax optimisation or risk- management contexts. What Accenture staff have in common are their cultural norms-business principles and the way they divide the world, whether for geography or to differentiate technical areas-and this is what the global taxonomy expresses. Comment: Going back to the original definition expressed in confrontational terms. The function of taxonomy in a business competative situation is subject to the difference between a companies market disposition and its strategic or tactical intentions. The genius of Rommel in the Western Desert was his ability to expoit limited forces which were not pre-determined by his original dispositions, particularly in the fluid battle situation which characterised desert warfare. This con fused the British who were accustomed to relate enemy dispositions to future intentions.
Taking all these points and others into account the authors developed a comprehensive list of challenges listed below together with a definition of the nature of a taxonomy.
Taxonomies for Business
The development of structure around information is by no means a new concept. Society tends to organise itself by subject, and we see both government, academia, business, industry, and people, clustering concepts to create both order and structure to information in a way that makes sense for their thinking and their interactions. But, as society has to cope with larger quantities of information through sources such as the Internet, it becomes progressively more challenging for people to identify and exploit the specific information that will help them in their lives and in their jobs.
As we move into the era of the knowledge economy, with its increased reliance on the effective use and exploitation of human intellectual capital to gain competitive advantage, it becomes understandable that organisations want to place some basic structures behind the information that people need. Structuring access to information not only helps prevent overload, but can help ease the identification of information that may prove Critical to the competitive position of an organisation in the marketplace
Defining a Taxonomy
At the start of the research our definitions for the term were both a process and a product:
* The creation of structure (arrangements) and labels (names) to aid location of relevant information.
* A high level device constructed to enable the user to gain an understanding of, and navigate around, available information.
As we explored how organisations were approaching these challenges, we found that:
* Businesses are now expending significant human, financial and technical effort to resource and develop taxonomic approaches that will support exploitation of their disparate information and knowledge resources well into the future.
* The majority of enterprises surveyed were looking for greater business effectiveness through improved information access; or efficiencies in information retrieval, which free time for more productive use. Intranets and enterprise-wide information portals are likely to fail if the issue of information content structure is not addressed.
* The corporate intranet has provided a common technological platform, and XML has the potential to provide a common data platform. The challenge now is to devise and deliver common information platforms.
* Many approaches to information structuring (e.g. classifications and thesauri) continue to be valid; but enterprises are likely to find that these cannot provide the complete answer to information overload as may be experienced, for instance, at the level of the enterprise portal, and that the 'corporate taxonomy' may be an answer.
* Organisations are, for the most part, taking an enterprise-wide approach to establishing taxonomies although the tactics for doing so vary and the scope of the taxonomy may well focus specifically on critical areas of information and knowledge.
* Few organisations had produced business cases for the taxonomy at the detailed level, but the great majority had backing from senior management to tackle the problem. This is a critical factor for success.
* Those responsible for co-ordinating the building of taxonomies within an organisation must facilitate (through shared understanding of needs and concepts) the creation of a structure that reflects the needs of the organisation and the sector within which it competes.
* In building a taxonomy there needs to be a balance between top-down (purpose-driven) and bottom-up (content-driven) development. By involving the user in the decisions on what constitutes a valid concept for inclusion in the taxonomy there will be greater buy-in to the use of the taxonomy across the organisation.
* A good taxonomy will always make sure that the information user knows where they are within the information structure even where the taxonomy itself is not directly visible. Good taxonomies ensure that a user can expand or refine their search - upward, downward or across important terms and concepts.
* Most of the enterprises felt that over-reliance on software solutions was dangerous; and were, in consequence, prepared to invest labour-intensive time for human intervention in building, applying and maintaining taxonomies.
* At the heart of the taxonomy debate is the need to achieve a balance between the talent of the taxonomy designer, the cost of the system to implement the taxonomy and the familiarity of the users both with the system and the structure of the information itself.
* The corporate taxonomy works at the level of information management by connecting people to documents (pull technology) and people to documents (push technology), but also at the knowledge management level by connecting people to people.
* Organisations may have to look at developing an interna 'metalanguage' to deal with the problem of different languages within a global business.
* The corporate taxonomy provides a 'knowledge map' to facilitate navigation of, and access to, the intellectual capital of the enterprise.
It is these last points that lead to the definition of taxonomy that we have developed as a result of this research:
"A taxonomy aspires to be:
* A a correlation of the different functional languages used by the enterprise
* To support a mechanism for navigating, and gaining access to the intellectual capital of an enterprise
* By providing such tools as portal navigation aids, authority for tagging documents and other information objects, support for search engines, and knowledge maps and possibly, a knowledge base in its own right."
Realising a Taxonomy in numbers. (Editor)
Lord Kelvin, it is said, made the somewhat provocative statement that nothing is known about a phenomena unless it can be expressed in numbers. As far as taxonomies are concerned the authors have the following to say: Measuring costs and benefits in information systems is notoriously difficult, and this was true of these case studies. Some of the organisations were able to give figures for development and maintenance costs, but these are impossible to compare without far more knowledge of the parameters obtaining in each case. We found little experience of systematic measurement techniques to monitor the effectiveness of the taxonomy development. One of the few to produce some objective figures (after the interviews were completed, and consequently not in the text of the case study) was Microsoft:
* 78,818 staff worldwide;
* 2.2 million documents available on the Microsoft intranet;
* 2% reduction in the number of clicks;
* An average 16 seconds saved per task;
* 11 % increase in task success rate;
* satisfaction rate:42% very satisfied and only 4% dissatisfied.
Another useful benchmark was produced recently by Forrester Research, who estimate the average cost of indexing web pages or documents (within a large project) at $4 each. The majority of case studies assigned at least one FTE member of staff to taxonomy development alone.
From these comments it is clear that there is no simple standard manner in which a taxonomy can be constructed, or indeed, a series of formal rules. When the idea was conceived, armies fought in simple formats, eg. Social Mobs, Lines divided into Left Wing, Centre, Right Wing, or organised groups with a disciplined command structure. Also battles lasted for only one or two hours before one taxonomy or the other prevailed As the above comments indicate even the study of a taxonomy does not reveal a lot.
This report is 174 pages long most of which consists of case histories of particular companies taxonomy structures. A close study of these histories is probably the only way that the operation of a taxonomy in a particular application can be appreciated since the number of variables, together with their significance, vary from company to company. Copies of the report are available from TFPL Ltd, 17-18 Britton Street, London ECiM SIL, and 55 Broad Street # 20c, New York NY10004.
RELATED ARTICLE: Managed data storage services
Interxion, providers of independent Internet Exchange Centres (IECS) has broadened its service concept to manage data storage, allowing Application Providers, Content and Media companies to manage, move, share, and protect data as if it were residing at their own physical location. Customers can buy as much data storage as needed at anytime on a pay-as-you-go basis. For data storage services, Interexion has an agreement with EMC Corporation, who will provide the customers of Interxion's IECs with EMC's storage systems for round-the-clock access to the data needed to run their businesses. EMC are specialists in networked storage solutions and allow organisations to effectively store and access the large quantities of data, in all its different forms, generated by the information economy. For storage, customers have three service options: storage on demand, back-up and restore and professional services. Note: Storage hosting revenues in the Internet infrastructure industry are expected to expand to $10bn busine ss by 2005, according to the Internet Hosting Report Nov. 2000 by Morgan Stanley Dean Witter the Investment Bank. www.interxion.com
Shortfall in Customer Information
Research by International Consulting across marketing directors of large business to consumer organisations from the banking, insurance, mail order, telecoms and utilities sectors, has uncovered a significant shortfall in the capture and accessibility of customer information within large organisations. Almost 50% of marketing directors said that they did not have access to what they view as their 'top five' most important items of customer information.
1. Customers personal information - includes demographic profile, income status, occupation
2. Customer address
3. Customer history: includes the tracking of products purchased and spend and value of customers
4. Customer financials : includes turnover, credit rating and payment method
5. Customer Contact information: includes more detailed information on how to contact a customer apart from address, such as telephone and email
Surprisingly, much of the information that marketing directors do not have access to is basic customer information. Almost 40% of marketing directors that do not have access to the important customer information they need, said it does already reside within their company but is inaccessible. This high-lights a significant opportunity for organisations to utilise the customer information they already have by using business intelligence (BI) and analytical tools to unlock the data and deliver it to the various functional areas of the business.
Over 60% of marketing directors that do not have access to the important customer information they need said that it is not currently being captured by the business. This suggests that a large number of organisations have yet to implement CRM systems that enable them to automatically capture and retain customer information.
These findings indicate that a significant number of organisations are missing a huge opportunity for competitive advantage through targeted customer marketing because they lack access to fundamental customer information which drives the success of any CRM strategy.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||TFPL report|
|Date:||Mar 1, 2001|
|Previous Article:||Business intelligence to trigger fundamental restructuring of organisations. (Storage Developments).|
|Next Article:||Too many hits bad for the web site. (Viewpoint).|