Elevating business intelligence
There are critical success factors on the road to the cloud for business intelligence.
In my previous Industry Insight, I outlined certain important points to consider when defining a cloud business intelligence (BI) strategy. In summary, this came down to having a clear understanding of the business case (and associated benefits), as well as the business use cases (to understand where this would be applied).
At this point, if you had gone through this process, hopefully as part of a coordinated, managed project, you would most typically be right in the middle of the hype cycle of the project. At this stage of the process, everyone is really excited about the expected benefits identified in the strategy, such as improved performance, reduced costs, scalability, agility, improved planning, financial forecasting and costing.
As a result, there is tremendous energy and pressure to proceed rapidly forwards. However, what about the potential downside? And, what about the risks? It is crucial that not just the pros of a cloud BI solution are evaluated, but also the cons. Knowing the potential challenges upfront will allow us to implement mitigating steps, or worst case, delay the implementation until potential issues/risks can be resolved.
While there is certainly no limit to the number of unique challenges that can be encountered across diverse, unique organisations, I believe there are a handful of standard, common pitfalls that organisations should be aware of, and proactively acknowledge and manage.
Data traffic: Initial and/or recurring uploads of data into the cloud - can be costly and time-consuming. This can have a significant impact on the batch windows and overall cost of ownership. Very often, additional, dedicated lines need to be set up to the relevant service provider to ensure optimal performance. This is especially true in the African context, where data bandwidth is costly. It is critical that a realistic assessment of these costs is included in the overall estimates and planning.
Security and compliance: Moving data into the cloud, especially personal information, can be difficult to obtain approval for. There is a great deal of caution in South Africa right now with regards to personal information. This can often result in a hybrid solution requirement, where certain data has to remain onsite, while non-restricted data is moved into the cloud. At the very least, additional due diligence needs to be performed to ensure regulations are not violated when adopting a cloud BI solution.
Pricing models: Pricing models for cloud BI solutions can at times lack a certain level of transparency. It is often initially easy to consider a cloud BI solution to be cheaper, but on closer reflection, based on pricing/usage, cloud BI solutions can sometimes end up being more expensive over the long term.
Governance: Adopting the cloud as a platform implies relinquishing a certain level of control and governance. On the whole, we are comfortable with this, as most of the vendors and platforms have demonstrated their ability to manage this on our behalf very well. However, it would be foolish to assume all organisations will appreciate this point. This is an important human element to consider and be mindful of.
It is crucial that not just the pros of a cloud BI solution are evaluated, but also the cons.
Having understood the potential cons, it is now important to define the critical success factors that the potential cloud BI solution needs to be measured on. These can once again include numerous points, but there are certain core principles that need to be clearly understood, quantified and measured, before continuing any further.
Data privacy and security: Carefully consider the minimum requirements of local and international regulations on privacy and security of data, which can limit what data is stored in the cloud, or hosted in specific countries.
Data transfer rate: Define the acceptable speed at which data needs be uploaded/downloaded in order to meet batch window and end-user requirements.
Data transfer volume: Define the expected data transfer volume and frequency, and evaluate within the context of existing bandwidth.
Data transfer costs: Define an acceptable cost per gigabyte of data transfer, taking into consideration any potential price escalation clauses based on volume uploaded or downloaded, etc.
Local availability: The importance of reliable Internet connectivity needs to be clearly understood and defined, particularly with regards to the impact that lack of Internet access can have on the solution and the business as a whole.
Cloud availability: The availability of the cloud BI service provider obviously has a huge impact on the success of these solutions. We expect cloud BI service providers to have stable platforms, but what are the organisation's requirements and expectations regarding this?
Disaster recovery: Appropriate disaster recovery needs to be in place to protect data and solutions, as well as to meet regulatory requirements.
Suitable redundancy: This speaks to the ability of the solution to configure/select the level of redundancy to suit the nature, importance and usage of the data being stored in the cloud.
Change management: This speaks to the internal organisation's capability to adopt the new paradigm. This is an important part of the successful implementation of the solution.
Understanding these points in the beginning of the cloud BI journey will yield great dividends in the future, as it lays the groundwork for all subsequent decisions around vendor and platform selection, and solution implementation options.
Julian Thomas is principal consultant at PBT Group, specialising in delivering solutions in data warehousing, business intelligence, master data management and data quality control. In addition, he assists clients in defining strategies for the implementation of business intelligence competency centres, and implementation roadmaps for a wide range of information management solutions. Thomas has spent most of his career as a consultant in South Africa, and has implemented information management solutions across the continent, using a wide range of technologies. His experience in the industry has convinced him of the importance of hybrid disciplines, in both solution delivery and development. In addition, he has learned the value of robust and flexible ETL frameworks, and has successfully built and implemented complementary frameworks across multiple technologies.