The Power of the Platform

by CXOtoday Staff    Aug 25, 2009


The IT pressure cooker

The choices that companies make today will determine how well they weather the current economic storm. Executives in all kinds of companies are contemplating:

* How to go global
* How to gain market share
* How to stay lean and mean
* How to put the house in order

There are thousands of decisions to be made. But the success of each business imperative depends on one thing: timely, holistic, accurate data. And that’s where IT comes in. Companies look to their IT organizations to provide the data the business needs-when and how it needs it.

This task is easier said than done. Data is scattered all over the enterprise-in applications, in databases, and on desktops in PDFs, Excel spreadsheets, and Word documents. It’s housed outside the corporate firewall, too-in applications "in the cloud" with software as a service (SaaS) and business process outsource (BPO) providers and with trading partners.

IT organizations are feeling the heat. Each business imperative begets a new IT initiative. Each new IT initiative creates a new IT project. And each IT project demands data- access to data, movement and consolidation of data, as well as a fundamental understanding of the quality of data.

The old ways aren’t working

Traditional data integration approaches are inadequate. They don’t address the complexity of today’s IT environments, nor can they scale to handle the range of initiatives that IT must execute.

Disparate point solutions that connect hundreds (or thousands) of applications just splinter operational data and lock it up in departmental applications, such as ERP and CRM.

Application-centric approaches to data integration don’t take into account all enterprise data.

Hand-coded data integration approaches don’t work either. Hand coding is time and labor intensive. It’s also prone to errors. As IT strives to manage larger data volumes and more data formats, hand coding often results in more complexity - not less. It drives maintenance costs up and drags IT’s efficiency down.

And what about data quality? Traditional data integration approaches can’t ensure that all
data - customer data, material and asset data, financial data-is complete, consistent, accurate, and current, regardless of where it resides.

If your IT organization continues to take a traditional approach to data integration-in silos, by department, by application, or by database-you will spend more time and money managing the complexity and "keeping the lights on," rather than tackling new business imperatives.

The new economic reality demands a new approach

IT organizations need a robust, new approach to data integration-one that can:

* Integrate all the on-premise data silos within the enterprise including unstructured data
* Integrate off-premise data in cloud computing applications and systems
* Exchange data seamlessly, business to business, with trading partners
* Ensure the quality of all data
* Cost-effectively manage the application life cycle

But just when companies are asking their IT organizations to handle more data integration projects, they’re circling the wagons financially. If they aren’t actively cutting IT budgets, they’re certainly scrutinizing them more closely. Companies are slowing IT purchasing cycles to complete additional due diligence. They are extending time to deployment to assess TCO and analyze potential ROI. And they’re actively looking for ways to control costs and eliminate redundancies.

Caught between these two opposing forces, your IT organization needs to increase ROI while simultaneously reducing TCO. There are three ways you can do so:

1. By increasing operational efficiency
2. By leveraging your existing technology investments
3. By reducing costs of development and deployment, as well as operations and maintenance.

IT organizations can do all these things at once with a data integration platform. A data integration platform is a comprehensive set of technologies for accessing, discovering, cleansing, integrating, and delivering data to the extended enterprise.

Reduce costs

Today’s heavily scrutinized IT budgets make cost a key concern. Individual integration approaches, such as hand coding or point solutions, may seem cost-effective at first, but supporting them soon proves to be costly and time consuming. Changing a single application or system causes ripples across multiple integration points, creating unreliable results that necessitate extra cross-checks and manual cleansing.

By contrast, a data integration platform dramatically decreases the time and resources needed for development, maintenance, and administration. Easy-to-use role-based tools and reusable development assets increase productivity and cut time to deployment. Codified methodologies eliminate variation for more accurate results. High scalability and ease of administration simplify maintenance and upgrades. It adds up to lower IT costs, both up front and over time.

Operate more efficiently

As companies increasingly make data management a business issue rather than just an IT concern, minimizing the complexity of multiple tools, skill sets, and vendors becomes more critical for productivity. Many IT organizations need to learn this important lesson. They try to tackle multiple data integration projects, but they approach each one on an ad hoc basis.

With different tools and methodologies for every project and no ability to leverage what has been developed or learned in past projects, the results all too often end up being costly, complicated, redundant, and unreliable.

A data integration platform helps IT organizations operate more efficiently by increasing productivity. A platform keeps IT from having to reinvent the wheel for every project. Instead, IT can share methodology, technology, and assets, such as logic and metadata, across all projects.

A unified data integration platform enables IT and the business to collaborate more effectively. A platform provides toolsets that share a common look and feel and are designed to work seamlessly with every other tool across multiple projects. These tools are tailored for each function, so each role can focus on their areas of expertise and develop their skills faster. Each person involved in data integration spends less time learning the platform and more time putting it to work.

Maximize the value of current technology

In this economic environment, every single technology investment is under intense scrutiny. IT organizations need to make the most of the technology they have now. With a data integration platform, IT organizations can continue to use legacy systems and applications while avoiding the expense and risk of "rip and replace." In addition, a data integration platform enables IT teams to reuse assets from one project to the next, thus reducing TCO as well as the expense of training people and developing their skill sets.

The ideal data integration platform

A data integration platform must solve the problem of data fragmentation across the enterprise to allow data-driven business decisions to be made more quickly and business operations to be run more efficiently and effectively. It must serve as your company’s technology foundation, providing a managed approach to data integration.

To meet these requirements, a data integration platform must be four things: comprehensive, unified, open, and economical.

Comprehensive

The ideal data integration platform must feature a comprehensive set of capabilities that enable your IT organization to provide data that the business can trust, when it’s needed, and where it’s needed.

Supporting the complete data integration life cycle

A data integration platform must support all five key steps in the data integration life cycle: access, discover, cleanse, integrate, and deliver.

Step 1: Access - Most organizations have data in thousands of places, not just within the enterprise but also beyond corporate firewalls, residing with business partners or "in the cloud" with SaaS vendors. All data must be accessible, regardless of its source or structure.

Step 2: Discover - Data sources-in particular, poorly documented or unknown sources-must be profiled to understand their content and structure. Patterns and rules implicit in the data must be inferred. Potential data quality issues must be flagged.

Step 3: Cleanse - Data must be cleansed to ensure its quality, accuracy, and completeness.
Errors or omissions must be addressed. Data standards must be enforced, and values must be validated. Duplicate data entries must be eliminated.

Step 4: Integrate - To maintain a consistent view of data across all systems, data must be integrated and transformed to reconcile discrepancies in the way different systems define and structure various data elements.

Step 5: Deliver - The right data must be delivered in the right format, at the right time, to all the applications and users that need it. Data must be both highly available and secure in its delivery.

Furthermore, a data integration platform must also:

Audit, manage, and monitor — Data stewards and IT administrators need to collaborate to audit, manage, and monitor data. Key metrics, such as data quality, are constantly measured with an eye toward steady improvement over time. The goal is to track progress on key data attributes and flag any new issues for resolution and continual improvement once data is fed back into the data integration life cycle.

Define, design, and develop — Business analysts, data architects, and IT developers need a powerful set of tools to help them collaborate on defining, designing, and developing data integration rules and processes. A data integration platform should include a common set of integrated tools to make sure all people are working together effectively.

Enabling any data integration project

A data integration platform must be robust, flexible, and scalable enough to handle any type of data integration project, including data warehousing, data migration, test data management and archiving, data consolidation, master data management, data synchronization, B2B data exchange.

Your IT organization may be conducting several types of data integration projects at once, from a single department data warehousing project to a global data migration project. Your team needs to be able to start small with one project type and then reuse the same skills and assets-enabled by shared metadata-for follow-on projects.

And a data integration platform needs to be able to handle both analytical data integration as well as operational data integration.

Delivering data at any latency

Depending on the application and use case, there is a wide spectrum of time frames and latency requirements for data integration. Some projects require data to be integrated monthly or weekly; others need integrated data available in seconds. And IT organizations need the flexibility to change latency requirements without having to rearchitect the entire infrastructure. The ideal data integration platform must furnish support across the entire latency spectrum, delivering trusted data whenever applications or users need it-whether in real time, batch, or changed data capture (CDC).

Unified

A single, unified data integration platform greatly simplifies the lives of your IT resources. When you have all the data integration capabilities you need for the extended enterprise from a single vendor, you maximize productivity with role-based collaboration, shared metadata, and a single, unified run-time engine.

Role-based collaboration

Data integration projects involve both IT and business people in multiple roles. They all have very different tasks to accomplish, and all bring different skills. Each role needs a different set of tools designed specifically for it. At the same time, project team members must work together, sharing artefacts and tasks, to increase cross-team productivity and ensure that IT and the business are aligned.

The ideal data integration platform provides role-specific tools, specifically designed for each person’s skills and tasks. These role-specific tools share consistent interfaces. They have a common look and feel and are integrated with each other. As a result, they are easy to learn and easy to use. Team members can get up and running quickly and stay productive by reusing assets across different data integration projects.

Shared metadata

A data integration platform must provide shared metadata. Each tool within the platform must be able to access relevant metadata about where data is stored, as well as the business rules and logic associated with it. With shared metadata, everyone can be working on the same thing. The metadata stays consistent, and every user can easily see the impact of potential changes.

Unified run-time engine

The heart of a data integration platform is a single run-time engine. The individual products that make up the platform should all run on the same engine, which simplifies implementation, administration, and maintenance. A single engine ensures easier upgrades over multiple releases.

The platform must be designed for enterprise-grade deployments, with proven scalability, availability, and security, so that you can bet your business on the platform.

Open

An open, neutral data integration platform is designed to work with everything-your hardware, software, technology standards-in your current IT environment, as well as anything you may add in the future. An open platform can protect your company from the risk associated with vendor lock-in.

Accessing data from any source

Most organizations store data in hundreds of different formats: enterprise applications, databases, flat files, message queues, spreadsheets, and other documents. A data integration platform must handle any datatype or format, including structured and unstructured data, and all master datatypes

More and more, data is moving beyond corporate firewalls and "into the clouds." Cloud computing has become more mainstream, with more companies relying on SaaS providers of human resources and CRM applications. A data integration platform must be able to access data residing outside the enterprise. This includes data that may come from multiple business entities and may be spread across many different geographies or countries.

Mitigating Risk

The IT landscape is changing. This causes uncertainty. IT organizations need a strategy to mitigate the risk of this change. You need a data integration platform that supports all current technology standards, from operating systems to databases. It must be open, ensuring that it works with everything you already have today or that you might have in the future. This includes all the different applications and data sources in your enterprise, as well as in the cloud or with your partners.

Economical

An economical data integration platform is one that delivers the lowest possible total cost of ownership (TCO) and delivers the fastest and highest return on investment (ROI). These factors are particularly important in today’s tough economic climate where every single technology investment, current and future, is scrutinized for its ability to help the IT organization and the business:

* Reduce costs
* Operate more efficiently
* Deliver value quickly

Lower TCO

A data integration platform must provide easy-to-use tools and proven scalability and performance to reduce up-front expenses, cut ongoing maintenance and administration costs, and deliver value rapidly. Companies can deploy the platform for a particular data integration project and then scale up to tackle additional projects without spending money on further tools or training. In short, a data integration platform enables your IT organization to do more with less.

Faster ROI

Achieving a quick return on your investment in a data integration platform depends on your ability to ramp up quickly and put it to use. You need to augment your IT resources.

Conclusion

The companies that successfully weather economic slowdowns are those that can sense and respond to change. They’re the companies that can act quickly and take advantage of opportunity as shifts occur in the competitive landscape, in the market, and in the economy.

They need data - The right data at the right time with unquestionable quality. According to Gartner, "Strategic use of information determines the ability of enterprises to compete and win.

These companies will rely heavily on their IT organizations. IT has a critical role to play to help their companies become data driven. A comprehensive, unified, open, economical data integration platform allows IT to rise to the occasion. It provides a firm foundation for more efficient, effective, affordable access to data. It keeps the lifeblood of timely, trusted data fl owing. And that enables IT organizations to support the business through hard times as well as position the business as stronger, more agile, and more competitive when the economy improves.