Pre-Packaged Data Warehousing – A Solution or an Oxymoron?March 26, 2014 No Comments
Featured Article by Bob Eilbacher, Vice President Marketing & Sales at Caserta Concepts
A critical component of a successful Enterprise Data Warehouse (EDW) initiative is the ability to accurately reflect a company’s core structures and processes. No two companies are alike – and all companies are likely to grow and evolve as the business environment changes. Architecting an EDW that reflects your business today as well as tomorrow is a challenge but also a key to success.
Is it possible to create an Enterprise Data Warehouse solution with a commercial, off-the-shelf Data Warehouse package? That’s a question that has been raised many times in recent conversations. “Off-the-Shelf” or “Out-of–the-Box“ Data Warehousing solutions have been around for some time, and to their credit these offerings have had some success in the Data Warehouse market. The packages are typically focused on a narrow slice of a particular industry and there are more than a few options. For example, in the Higher Education space there are solutions that offer reporting and analytics on everything from enrollment to financial aid to faculty and staffing. The lower-cost model typically appeals to highly budget-conscious organizations, but there’s a significant price to pay with an off-the-shelf approach. The challenge for organizations that are about to make a Data Warehouse strategy decision is understanding what the pre-packaged constraints mean to them and considering the longer term consequences of an out-of-the-box approach.
How do they get it “in-the-box” to begin with?
Ask any Product Manager who has been responsible for bringing an off-the-shelf product solution to market and they will agree that the more you want to put a box around a solution the more constraints you have to put in place for how that product interacts with real-world environments. That’s the nature of the beast. The real world however, demands flexibility. This is certainly true of Enterprise Data Warehouses. There aren’t too many enterprise data systems that have to touch every other system and subsystem in every department across the entire organization like a Data Warehouse solution does. This requires the ultimate in solution design flexibility. Every product has a certain amount of configurability built into it, but to keep the implementation low-touch and in line with the off-the-shelf paradigm, you have to keep configurability from getting too complex. Additionally, many of the pre-packaged solutions available are extensions of, or very tightly coupled to, a particular source system. They might do a good job of integrating with the variability in that source data, but all of your other source systems will require “external” integrations. Constraints on configurability will invariably present limitations on the extent to which a pre-packaged solution can integrate into your environment. The result to your Data Warehouse can typically show up as delayed data or missing information.
Whose Enterprise is it anyway?
A critical component of a successful Enterprise Data Warehouse initiative is the ability to accurately model an organization’s core structures and processes. There are no two companies or institutions built the same, and no two whose data environments look the same. Organizations may have similar transactional systems but those systems have been configured to support each company’s unique processes. Likewise, the data stored in those data systems, or shared between them, is wholly unique to that organization. As a simple example, if you were to try and compare organizational charts between two companies of equivalent size serving the same industry segment, you would be amazed at how different they are. Companies compete by differentiating themselves from their competition; it transcends their entire operation and consequently permeates throughout their data systems. In this kind of market environment you simply can’t handle every possible enterprise model in a prepackaged solution approach — at least not without adding the level of complexity that would require high touch implementations, something that runs contrary to the approach. You need to live within the constraints afforded to you by the package, which means at some level you will need to adapt to that enterprise model. The result to your Data Warehouse is a picture – delivered through reports, charts, graphs, dashboards, etc. – of an enterprise that is somewhat misaligned with your reality.
What happens when you don’t fit in the box any longer?
Out-of-the-box Data warehousing solutions can serve an immediate need, but because of the limitations they typically present, they are often only a temporary solution to an evolving need. Enterprise Data Warehouses are living, growing entities. They should be designed to grow and change as the enterprise they reflect evolves, whether that’s slowly with organic change or rapidly from acquisitive growth. In addition, growing organizations typically become more complex; think legacy systems and data, plus new data systems, larger volumes, more customers, etc. At some point most companies faced with these new growth challenges hit the wall with pre-packaged Data Warehouse solutions. The enterprise they could model before, at least to some extent, just can’t be accommodated going forward. Likewise, if an organization using a Data Warehouse that came pre-packaged with one of their source systems decides to change that system, they have essentially obsoleted their Data Warehouse as well. Unfortunately, in all of these cases, those enterprises are now faced with a new Data Warehouse initiative, often starting again from scratch, to create an enterprise data model that reflects their new reality. If they’ve learned their lesson, it will also be one that can scale to support their organization in the future.
Stay out of the box
Data Warehouses help you fundamentally better understand and run your unique business. By its very nature, designing an Enterprise Data Warehouse is an intimate affair. It’s “custom” by definition and not well suited to generic, off-the-shelf implementation. It’s also not trivial. It should be designed by seasoned Data Warehousing Specialists who take the time to understand your business, your organization, your technology landscape and your data environment, and who have the skills required to architect an Enterprise Data Warehouse platform with your growth in mind. Most companies don’t have all of the skills required in-house, partnering with a well-regarded Data Warehouse consulting/services partner can help ensure success by following proven methodologies and best practices that have been developed within the industry over the past 20+ years.
Bio for Bob Eilbacher, VP Marketing & Sales for Caserta Concepts
Bob is an experienced executive in the technology solutions and services field. He has focused on developing and executing strategic corporate initiatives to provide operational insight and drive efficiencies across the enterprise.
In his role with Caserta Concepts, Bob works with clients to develop information management solutions that transform structured and unstructured data into analytics-driven business insights. His strong technology background, coupled with a deep appreciation for how solutions provide value to the enterprise, allows Bob to engage as a trusted advisor.
Leading new business development initiatives and ventures with a demonstrated track record of success, Bob helps analyze the viability of proposed solutions, develops creative strategies to achieve success, and implements processes that ensure results are on time, under budget, and beyond expectations.
DATA and ANALYTICS , Fresh Ink