• Votes for this article no votes for this yet
  • Dashboard Insight Newsletter Sign Up

Making the ITIL Service Knowledge Management System a Reality

by Mike Urbonas, AttivioTuesday, February 21, 2012


The Cash Management division of a leading North American financial services firm handles some 1,500 IT incidents per month from staff seeking support of 80 customer-facing internal enterprise applications. The applications have dependencies with more than 170 servers within the division, as well as nearly 40,000 servers company-wide. Major incidents must be resolved within minutes or incur such consequences as higher costs from violated service level agreements (SLAs) on up to what a company executive described as "a major disaster," such as failing to transmit daily information to financial regulators before their deadline. The company recognized the need to reduce the high turnover of support technicians – and the related high costs of constantly training new hires – as well as improve service, as evidenced by unacceptably high mean time to resolution (MTTR) and service escalation rates. The company had made major progress implementing the ITIL framework and a leading IT services management application, but that addressed just a piece of the problem because it only included structured data sources. There was still an unacceptable bottleneck because system analysts had to spend excessive time accessing and manually correlating other essential information sources, including application log files, wikis, application manuals, programming documentation, e-mail, EDMS (PDF, Excel, PowerPoint, Visio, etc.), external sources, and more than 800 SharePoint sites. Read on to learn how they achieved breakthrough results, including decreasing MTTR by more than 40 percent.

As the above real-world scenario illustrates, knowledge management – timely, on-demand access to all relevant enterprise information – is a vital, pervasive process affecting all areas of IT operations.  Effective service management cannot be provided without effective knowledge management.

Recognizing the widespread impact that effective knowledge management can deliver, the IT Infrastructure Library (ITIL®) now recognizes it as a service management Best Practice in ITIL 3, providing guidelines that did not exist in ITIL 2.

The most significant of these recommendations is the creation of a Service Knowledge Management System (SKMS), designed to capture and present knowledge from sources ranging from one end of the service management process lifecycle to the other.[1] An effective SKMS resolves the vexing problems and hefty costs noted above, providing IT and business benefits that are easy to understand, measure and translate into return on investment (ROI). Even better, by helping to end a constant state of reactive 'fire drills,' the SKMS enables service management leaders to proactively focus on strategic planning and improved decision-making in such areas as IT infrastructure and new technology adoption, to cultivate a new level of customer service and competitiveness.

As ITIL provides guidelines and not specific instructions, there are some misunderstandings as to what constitutes an SKMS, whether creating one is an attainable goal, and how to get there.  This paper will:

  • Address misconceptions regarding the SKMS
  • Explore key technical challenges to creating an SKMS
  • Present unified information access (UIA) as the best technology to fulfill these challenges
  • Conclude with a full case study highlighting the successful SKMS implementation at the financial services company profiled above, using a leading UIA enterprise platform, and the business benefits achieved

Overcoming SKMS Misconceptions and Technical Challenges

Most ITIL defines an SKMS as "…a set of tools and databases that are used to manage knowledge and information. The SKMS includes the Configuration Management System (CMS), as well as other tools and databases. The SKMS stores, manages, updates, and presents all information…to manage the full lifecycle of IT services."[2]

This 'guideline' definition is subject to interpretation, which has led to some misconceptions.

One misconception is that deploying an SKMS requires, by definition, the stitching together of various service information sources in the form of a custom systems integration project:

The vision of SKMS is a good goal to shoot for… But with the heterogeneity of data, the diversity of tools used in most organizations…, and the fact that most large ITSM tool vendors [have yet to integrate] products that have been merged together through acquisition, [the SKMS] still remains the one of the largest…ITSM challenges.[3]

ITIL guidance clearly encourages companies to envision the SKMS as an enterprise knowledge platform, not a point solution for problem resolution.[4]  This is important: an SKMS platform, rather than a customized, purpose-built integration of a finite set of data sources, will prove far more effective in overcoming IT information silos, empowering service professionals to effectively manage incidents, problems and system changes, and easily integrating new information sources as required.

Clearly, a key barrier to implementing an SKMS is the fact that IT organizations tend to operate in distinct "silos" each with its own disparate information and based on technology rather than how IT actually supports the business.[5] In fact, organizations that built out their IT service management based on the prior ITIL 2 framework often found the dozen distinct service processes owned by separate teams and isolated in silos.[6]

Another mistaken notion is to think of an SKMS purely in terms of databases (or structured data).  For example, the book IT Service Management Global Best Practices presents the SKMS as consisting solely of databases; specifically, one or more configuration management databases (CMDB)[7] along with other databases: service desk data, response time data, etc.[8]

However, to be truly effective, the SKMS must include a significantly wider array of knowledge artifacts, drawn from diverse information sources throughout the organization, spanning well beyond databases to include other types of data: 

  • Semi-structured data ('machine-readable' text information with tags or other markers enabling the parsing and identification of useful data; e.g., log files, XML, etc.)
  • Unstructured content ('human-readable' text information; e.g., documents, web content, email, wikis, etc., residing in enterprise content management systems; websites, file servers, etc.). 

Unstructured information sources are typically far more prevalent and plentiful – often estimated as comprising 80 percent of all enterprise information – than structured data sources (databases).  It is the unstructured, text-based sources of service information that will comprise most the SKMS' wide body of knowledge as envisioned by ITIL.

The financial services company profiled at the beginning of this paper recognized early on the need to integrate knowledge well beyond its service management databases.

"We needed to go beyond the data sources specifically named in the ITIL framework," an executive of the company said.  "We need the ability to refer to documents, wikis, application log files, SharePoint and other content as well as databases.  If we only needed structured data sources, we could have built that in-house, but that’s just a small part of our wide variety of essential information sources."

Additionally, the financial services company acknowledged the challenge of integrating enterprise information silos.

"The data repositories and unstructured information related to each major process – incident management, problem management, change management and our CMDB - each involve different groups of people," the executive said.

Viewing an SKMS Implementation as an Extreme Information Challenge

Text Box: Figure 1.  The simplified illustration of the SKMS from ITIL (Service Transition book, p. 147), expanded to convey the financial services company’s specific, and substantial, informational needs spanning well beyond service databases. This is in keeping with ITIL’s description of the SKMS as a  "broader concept" that "covers a much wider base of knowledge" (p. 147).

It will be helpful to think of the task of deploying an SKMS platform as an extreme information challenge.  Extreme information is the term Gartner uses to more accurately describe the full picture of what many refer to as Big Data: the convergence, joining and presentation of enterprise information with high levels of volume, velocity, variety and complexity:

  • Variety – The full complement of enterprise data sources and multiple data types as defined above:  structured data, semi-structured data and unstructured content
  • Velocity – Often time-sensitive, Big Data must be analyzed as it is streaming into the enterprise in order to maximize its value to the business
  • Volume – Certain information sources can approach "Big Data" levels of volume
  • Complexity of individual data types, including ensuring that all related information, regardless of format, can be easily joined and presented together in response to a service professional's query

An effective SKMS must integrate, correlate and present an organization's entire body of service knowledge for easy on-demand access by end users via a wide variety of end user tools, interfaces and applications.

Unified Information Access as an SKMS-Enabling Platform

Beyond integrating, the extreme information described above falls outside the capabilities of available technologies.  A different approach, unified information access (UIA), acquires, integrates and presents all sources of information – structured, semi-structured and unstructured – in ways that enable the integration of seemingly unconnected information.  UIA provides flexible SKMS-enabling capabilities, including:

  • Unlike a data warehouse, UIA does not require relationships between data and/or content to be defined in advance. UIA is built around schema-less indexing so that users can dynamically join and present related information, regardless of data type, as well as add new information sources in the future.
  • UIA provides secure, high-performance access controls regardless of what information a user is accessing (structured data and/or unstructured content) or how they are accessing it.
  • UIA provides broad information enrichment capabilities with advanced text analytics that analyze incoming text for key phrases and entities (people, places, companies, etc.) that are extracted as additional metadata.  They are also used to reconcile index entries so that similar or related concepts within disparate information sources are unified to ensure all related information is joined together on demand. Content classification provides context and information organization, including categories, added to the content's metadata.
  • Best of breed UIA platforms can support a wide range of user interfaces, including:
    • Purpose-built applications using APIs. UIA is not a solution; it is a platform that enables rapidly deployed applications, including multiple custom views matching the needs of service professionals: incident management, problem management, change management, etc., as well as analysis of performance trends over time.
    • Simple search user interface providing keyword querying with the added benefit of navigation and discovery of related structured and unstructured information.
    • Business Intelligence (BI) integration with tools via SQL support and ODBC/JDBC connectivity.  UIA joins and exposes structured data and unstructured content for BI tools.

Unified Information Access for SKMS in Action

To fulfill its needs for a robust SKMS, the financial services company profiled in this paper implemented an IT Knowledge Expert solution which provided the essential information acquisition, integration and presentation capabilities described above. With this, the company has already reduced its incident mean time to resolution (MTTR) by 40 percent. "That's a dramatic improvement in a very short period of time," the company's executive said.

The company also reported substantial reductions in service technician turnover since implementing the solution. While the entire reduction in turnover is the combined result of a number of actions, the solution clearly played a major role. "Our systems analysts feel far more empowered now," the executive said. "They are no longer caught in situations where they don't have the information they need and feel helpless."

Prior to implementation, support specialists had to manually reference multiple databases, log files, knowledge bases, wikis, documents, and other repositories across the organization, each with its own login and methods for finding and accessing relevant support materials.  Often such manual inquiries were abandoned due to the excessive amount of time required. As a result, issues were frequently escalated before being adequately triaged.

The company also resolved other issues, including aggregating all relevant content from its ECM, EDMS, wikis and log files sourced from over 90 applications, as well as data from IT support management software and company-wide personnel databases identifying subject matter experts and executive stakeholders for each enterprise application.  They can now converge, join and present all relevant information from these disparate structured and unstructured information sources into a single search-centric service support portal.

The new unified views of all relevant data and knowledge enable service specialists to more quickly and efficiently access a large number of support knowledge repositories, without prior knowledge of where the answers might exist.

Similarly, they can now “connect the dots” between the multiple databases and repositories used by IT support specialists to resolve incidents, identify problem-solving root causes, manage system changes and 'system collisions' – incidents caused by simultaneous changes to dependent systems – and more. They also now have an operational management view into service management activity over time, including performance by support specialist, which systems incur the most incidents, which incidents impacted SLAs and more.


1. Peter Dorfman, Knowledge Management and the New ITIL Framework, May 12, 2007.

2. Office of Government Commerce, ITIL: Service Transition, page 244.  London: The Stationary Office (2007).

3. Compliance Process Partners blog (cppit.com), Service Knowledge Management System Nirvana, January 3, 2011.

4. Dorfman.

5. Dunn and Ho.

6. Dorfman.

7. A configuration management database (CMDB) contains relevant attributes of IT assets, or configuration items (CI) along with relationships and dependencies with other CIs. One or more CMDBs may be managed by a configuration management system (CMS). Source: see endnote 4, below.

8. Bryce Dunn and Linh C. Ho, "Bringing wisdom to ITSM with the Service Knowledge Management System," p. 415-424, IT Service Management Global Best Practices, Norwich, UK: Van Haren Publishing (2008).

Tweet article    Stumble article    Digg article    Buzz article    Delicious bookmark      Dashboard Insight RSS Feed
Other articles by this author


No comments have been posted yet.

Site Map | Contribute | Privacy Policy | Contact Us | Dashboard Insight © 2018