Is Data Warehouse Necessary For Dashboards Or Insights ? Well, A Data Warehouse Is Needed In The Long Term, But Not A Must For BI In Short Term.
<p>A business or an enterprise runs it operations on a continuous feed of information that comes from various sources, platforms, and systems. Collectively such sources make up the components for enterprise data.</p><p>The business takes decisions based on such information usually in the form of reports and/or dashboards from time to time. It is important to secure this flow of information. The information from enterprise data sources needs to be trustworthy, relevant, accurate and accessible. In a business of any scale there are various forms/sources/system that generate data from which reports are created for decision making.</p><p>Reports used by the business users for decision making are critical to the business and have a long life – typically for as long as the need remains e.g., consider the following reports used by various departments in a business:</p><p><img style="display: block; margin-left: auto; margin-right: auto;" src="https://kradminasset.s3.ap-south-1.amazonaws.com/ExpertViews/sangharsh+img.PNG" /></p><p>If we consider an organization manufacturing and selling goods, various departments will typically need a lot of reports at different times with varied levels of details. Most such reports are derived from transactional data residing in various systems. (see an example below) :</p><p style="text-align: center;"><img src="https://kradminasset.s3.ap-south-1.amazonaws.com/ExpertViews/sangharsh+img+2.PNG" /></p><p><em>Exhibit 1</em></p><p>In this case, data is derived directly from feeding independent systems and reports are provided to the organization, however, there are limitations and constraints in this mode of information extraction for decision making</p><ul><li>Reports are derived from transactional systems that are designed to handle transactions</li><li>The data from transaction systems is continuously refreshed and updated</li><li>Any change in feeding transactional systems will necessitate a change in reports and could have a cascading impact on consumers of that information in the organization</li><li>There usually is a limit on storing data in the transactional system</li></ul><p>However, realities of business operations demand that transactional systems be flexible to meet new business demands, bring in efficiency, incorporate changes in policies or business rules. Sometimes new systems are identified and/or older systems are replaced to feed information. </p><p>In summary changes in data, design, system to feeding transactional system would have a knockdown impact on reports.</p><p>To address this challenge on preventing knockdown impact of the transactional system on reports for decision making, <a href="https://www.bibirbal.com/"><strong>B</strong></a><strong>IBirbal</strong> proposes to build a repository of that collects data from different transactional systems, is well organized to cater to the demands of organization</p><p><a href="https://www.bibirbal.com/"><strong>BIBirbal</strong></a> proposes an approach to build a model that will transform and store data from transactional system for generating reports for decision making. The model is depicted below:</p><p style="text-align: center;"><img src="https://kradminasset.s3.ap-south-1.amazonaws.com/ExpertViews/sangharsh+img+3.PNG" /></p><p><em>Exhibit 2</em></p><p>This approach introduces a data store and common object model. The data store serves as the primary source of data for all decision-making reports consumed in the organization.</p><p>The benefits of this approach allow:</p><ul><li>To refresh data in data store from transactional systems</li><li>Reports are run on data from data store; any changes in transactional systems does not impact reports for decision making</li><li>Data in the data store can be maintained for a longer period</li><li>Technology and design of data store is different from those of transactional systems, a data store allows gives greater flexibility and speed to fetch information AUXILLARY</li></ul><p>We propose to offer services that analyse your reporting needs, study your report design architecture, and help you build a data store and reports that give you more flexibility and longevity.</p><p> </p>