♦  Business Software Research and Reviews

Software-as-a-Service

Although the application will be maintained and supported by a service provider, IT must still maintain control over it. IT managers must make plans to adjust the following existing IT processes:

• Change management. This process must have clear guidelines and instructions as to how to submit, prioritize, and implement changes to the new SaaS application. This isn’t different from any traditional deployment. The part that changes is that under SaaS, this process may involve new stakeholders as any customization that requires development work will be performed by the vendor.
• Support processes. The new SaaS application must be included in the support process. With SaaS, IT will be responsible for the first two levels of support (connectivity, functionality, or process issues), and the SaaS vendor will be responsible for the third level (technical issues with the application or infrastructure, and unresolved functional issues).
• Service level management. Because of the large reliance of the enterprise on the SaaS vendor, a solid SLA is absolutely mandatory. The service level management process must be adjusted to ensure proper creation and monitoring of the SLA.

from Software-as-a-Service: Can Your Software Lifecycle Take the Change? (IN-6453)

Document Collaboration - Microsoft Office

The de facto document collaboration tool is, of course, e-mail, and here Microsoft Exchange continues to dominate the corporate market along with IBM’s Lotus Notes. Although described by Microsoft as a complementary product, Exchange Server sits outside of Microsoft’s ‘Office system’. This, Butler Group believes is a weakness in Microsoft’s overall collaboration strategy, as e-mail is at the centre of just about every user’s world. Indeed, 9 times out of 10, document collaboration relies totally upon the services of email, and to consider it as an adjunct to the main Office system is a mistake. Although still in a very dominant position, we are starting to see organisations wavering over their Microsoft Exchange investments, and if Microsoft were to lose its grip on this most important market, then surely we would start to see the whole Microsoft edifice collapsing like a house of cards. Microsoft needs to re-invent the way that users collaborate - especially across different organisations, and to do this it must re-invent the way we all use e-mail.

Whilst IBM has been struggling to bring its Workplace strategy to market, Microsoft has been steadily building market share (and mind share) through deployments of SharePoint. Now, as we head into 2007, Microsoft is trying to deliver a knock-out punch with the simultaneous release of Windows Vista and 2007 Microsoft Office system; and whilst the effects of this mighty blow might not be felt for a couple of years or so, Butler Group believes that Microsoft’s combination of programs, servers, and services will be hard to beat in the long-run. Although there are one or two Enterprise Content Management (ECM) vendors showing promise, only IBM has the products, technology, and market presence to offer Microsoft a real fight. But from an end-users perspective, the continued domination of the desktop by Microsoft might not be a bad thing, as it provides a set of standards and a level playing field on which businesses can co-operate at that most important level: the human level.

from Document Collaboration: Linking People, Processes and Content (BG-0037)

Web Content Management

Determine the costs of the existing Web management infrastructure. The benefits of WCM are soft. To establish the business case and justify the project it is essential to baseline Web production costs. Establish a base value for the following metrics before considering a WCM solution:

• Total production costs
• Administrative overhead
• Employee training costs
• Help desk costs
• Site maintenance costs
• Graphic design costs
• Customer satisfaction levels
• Content quality ratings

from Web Content Management Crucial at 1,000 Documents (IN-6433)

Enterprise Search and Retrieval

ESR technologies are maturing rapidly, and there has been both innovation and consolidation from the vendor community. Embedded search tools within specific enterprise applications such as ECM, whilst useful and optimised, may not deliver navigation or personalisation functionality. Furthermore, they often require large amounts of bespoke work to connect with other enterprise applications, or legacy systems.

Enterprise applications will change, and new ones will be added for a specific business purpose or potentially after a merger or acquisition: Butler Group, therefore, believes that ESR technologies should be abstracted from, but integrated with, those applications, so that it should be possible to initiate the search from within enterprise applications such as CRM or BI.

However, integration of the ESR solution with internal, and potentially also with external applications, needs to go beyond the positioning of a search box on a screen. As an abstracted layer in the enterprise technology stack, Butler Group believes that ESR technologies should be tightly integrated with BPM technology to drive automated actions that utilise the information retrieved by the solution, and just as importantly to provide the tracking and audit functions. Integration with Workflow functionality in applications such as ECM is a good first-step, but ideally search and discovery at the enterprise level should be regarded as a universal ‘service’ to be called as required, from within process-driven applications working within a SOA. ESR technologies delivered as services are also an integral part of Enterprise Portals.

from Enterprise Search and Retrieval - Unlocking the Organization’s Potential (BG-0035)

Information Lifecycle Management

The starting point for a company is to understand its business needs, and then to ask the question: "What have we got in our data?" in order to start matching these up. Few organisations even know the nature of the data that is needed even if they have some idea about what they have stored; they need to sit down and think about this. From this exercise should come a defining of some categories of information.

For instance, value may be looked at on the basis of how much money the organisation will lose if the piece of data or an application becomes unavailable or, more subjectively, the risk or loss of company image associated with loss of the data. Applications could, for the sake of argument, simply be placed in a hierarchy class numbered 1-7 along with the information they produce.

In total, ILM needs also to handle both structured and unstructured data (in any mix). Most internet-based report items and e-mails are unstructured or semi-unstructured whereas most internal database software such as management accounting, ERP and CRM systems that access a database have structured information; the latter can more easily be identified, tagged and indexed by category or class - once the categories needed have been resolved. Any internally created file could be accompanied by internally created metadata about them, from the time of creation. Last used/number of times used information can be then be captured over time. However, this highlights the need to define the format of the metadata.

from Information Lifecycle Management (BL-5553)

Business Intelligence

The main drivers for strategic BI all have a direct impact on the data warehouse; for example, making effective use of information results from significantly increased data volumes, and the wider exploitation of BI services means that more applications will be accessing the warehouse. It is certainly the issue of business velocity that will have the most dramatic effect. Because data warehouses have traditionally been designed for periodic update, they are not as well suited to an increasing number of applications with a need for near real-time information, and as this requirement grows it will lead to an increased emphasis on integrated links with live data sources.

It is certainly true that maintaining the data warehouse and associated BI data repositories will become more of a complex issue when looking at the enterprise use of BI services than was the case when a data warehouse was being used to support limited and often more restricted departmental functions. In the past the data warehouse has normally been used to pick up what is in effect archive material and store it in a manner that provides useful information, typically supporting management decision-making.

In this role the fundamental difference between a live transaction database and a data warehouse would be that the live database involved data additions - updates and deletions - as well as read operations, whereas the data warehouse was principally used for reading only, with updates having taken place at a scheduled point in time. Also, unlike a transaction database, which is very specific to the workflow and function of the business and is therefore static in its nature, the data warehouse needs to be versatile, reactive, and easy to change. This situation will be added to as the use made of such data repositories is changed by the dynamic nature of strategic, rather than tactical, BI. This position will become even more of an issue as across-the-operation management learn how to ask new and more pertinent questions of the data, all of which may require changes right from the ETL level and add to the possibility of extending or adding to the use of existing data marts.

from Business Intelligence: A Strategic Approach to Extending and Standardizing the Use of BI (BG-0032)