Learn from Business Intelligence and Analytics Mistakes

By : |June 11, 2019 0

Analytics data from enterprise and external sources can bring insight which can be applied to improve business operations and customer satisfaction and speed the launch of new products and services. However, the journey to implement successful Business Intelligence and Analytics programs may have some roadblocks. If we pay attention to common mistakes like the ones described here, we can avoid repeating these mistakes and can add value to the business.

Organizational mistakes

Lack of strategy and governance

= Lack of organizational strategy for business intelligence and analytics can lead to duplication of efforts and also result in data silos without proper ownership and accountability.

= Without clear identification of data owners who can manage and define organizational KPIs, it becomes challenging to drive business intelligence data analytics outcomes as well as derive value from the data to benefit business/

Dependency on technical team

                                 

___________________________________________________________________________________________________________

= It would be a misplaced expectation on the technology team to create or lead the strategic road map. This may lead to underutilization of applications or investing in applications that are not aligned to business objectives.

= It can also result in the wrong prioritization and cause delays in the implementation of critical analytical applications, and deliver a negative impact for the organization.

Lack of knowledge in Compliance and Data Privacy Risks

= Data privacy and compliance for data security should be a priority focus right from the beginning, as violating these requirements can have a huge cost implication, apart from the waste of resource bandwidth and time invested in building analytics programs that violate customer privacy.

Waiting too long to get all your ducks in a row

= Analytics can be applied even for small use cases with existing data and tools or readily available services on the cloud – waiting for a big project to have everything in place may lead to a loss of market opportunities.

Technology Mistakes

Not focusing on data

= Data quality and security should not be an afterthought and need to be built as part of the process to create data that we can trust, and that meets compliance regulations.

Not agile enough to address business needs

= By not being aware of the business needs which are sensitive to market requirements, you may end up with multiple siloed projects and platforms across different departments.

= Buy versus build – It can be detrimental for an organization’s success if they are not open to simple and efficient, time-tested platforms that are available which may solve the problem quickly.

Thinking tools and technology equal analytics

= Technology teams tend to get carried away with the opinion that the tools and technologies are analytics rather than the business outcome that result from these tools. It must be understood, that these platforms create value only by addressing a business need or challenge.

= User Experience – The tools selected might have low adaptability due to a steep learning curve, and some basic features may be unavailable for the users. Getting defensive about the tools and technologies and forcing its use will lead to the failure of these platforms.

Not popularizing (democratization of) data assets

= Having wonderful data platforms with certified valuable data which is not popularized or socialized with business user communities and forums will lead to underutilization of those assets.

People Mistakes:

Lack of a team with domain and technical knowledge

Most analytics programs fail due to lack of understanding of the requirements by the implementation team, making it imperative that the team has members with both the business domain expertise, and the technical feasibility knowledge, and an experienced leader to avoid imbalances between technology and business.

Not keeping up to date

= The landscape is forever evolving, and not keeping up with technology advancements such as big data analytics platforms like Python or numerous cloud platforms will not only lead to failure to meet business objectives, it will also make the team obsolete.

Dependent on vendors

= While it may be prudent in some cases to leverage third party expertise, especially in the case of new technologies to gain a head-start, too much dependency on vendors must be avoided as it is not a sustainable approach. Instead, investments should be made in developing a team that is aware of organizational goals, the business model, processes and the tools and technology platforms available within the enterprise. Teams with this kind of vision can offer more consistent and business aligned support.

 

Implementation mistakes

Managing Expectations

= Clear goals for any analytics program are necessary to set achievable expectations and make tangible contributions to business transformation.

= It is equally important to set judicious implementation timelines to a program or a platform with all the required data, or else it can result in abandoned programs.

= Implementation Feasibility – The data availability must be keenly assessed, as they may be have external sources, carry sensitive information, may need a business process change.

Scalability and Performance

= While some programs can create interest at the prototype level with sample data, when it comes to production level implementation with a larger user base and a large volume of data, these applications if not assessed at the onset, may not scale and can lead to an abandoned platform.

 

End User Training

= Without end user training, products rolled out may not be leveraged or used according to expectations.

 

By Ravi Venkatesh, Assistant General Manager, Business Intelligence & Data Analytics, HTC Global Services

No Comments so fars

Jump into a conversation

No Comments Yet!

You can be the one to start a conversation.