If you're wondering how organizations can get started with advanced analytics, it would first help to know how they typically identify opportunities to put big data analytics to use. When identifying new opportunities for your analytics team, start with your client and/or stakeholders. Begin with what is top of mind to them. Often, analytics projects start from a business case with an information problem that must be solved. Data scientists apply analytics tools to business processes to solve information problems - this is why the work is often called "business intelligence." In some cases, information problems result from a failure to take advantage of a data source or sources. Data collection is most impactful when there is an underlying analytics strategy that explains how it will be used and why this makes sense from a business standpoint.
Quarterly or monthly meetings should be held with clients and stakeholders to discuss your project queue (or projects that are in flight) as well as any new opportunities. These calls should allow the engagement necessary to confirm the impact of your data science projects. The use of modern collaboration and communication tools will enhance cooperation throughout the organization. When a new opportunity is identified, it should be evaluated to make sure it fits your defined vision, and roles and responsibilities as an analytics team, as well as its priorities relative to the other projects already in your analytics portfolio. Further, opportunities should be prioritized based on an agreed upon approach for determining how much value the opportunities will enable for your client/stakeholder and the feasibility of implementation. As a result of these opportunity identification/prioritization discussions, it is important to identify who will be accountable for the project and champion its value. It is critical to determine how the value will be measured prior to beginning the project and establish a baseline. It is only in doing so that the value can be captured throughout and measured once the project is completed.
The value of analytics projects can be categorized into soft and hard value generation. Hard value generation can be further divided into three buckets:
- Value generated by having more efficient systems and processes;
- Revenue generated from uncovering hidden opportunities; and
- Reduced costs of operations.
The first bucket of hard value generated by having more efficient systems and processes is usually a combination of the value of making better and faster decisions and automating processes to make better use of assets. This could have short-term impacts on revenue. The second bucket of value generation can be quantified by calculating how much additional opportunity your analytics capabilities can potentially generate. An example would be to calculate the revenue boost received from increased cross-selling (selling multiple types of your products to the same consumer) over time. Cross-selling is a classic example of a hidden opportunity that can be discovered using historical data on customer behavior to develop predictive models. Finally, the third bucket involves using analytics to improve operational efficiency, where the improvement can be directly measured.
The soft value that analytics teams can claim from projects is usually related to lifts in customer satisfaction or meeting a safety or environmental initiative. These projects typically have a very high return on investment. It should be noted that most projects can have both hard and soft savings. Analytics leaders should try to make sure that both types of value are fully conveyed. Over time, this can help the whole organization improve its ability to identify new analytics opportunities.