Machine learning in a business setting has by and large proven to be discouraging. Here’s what can be done better
According to research from McKinsey, about 56% of businesses had resorted to AI adoption by the end of 2021, up about 6% from 2020. Yet, inherent challenges remain. Companies continue to struggle to find exact use-cases for AI to improve the way their businesses operate. The use of machine learning, for example, in solving obsolescent business practices is particularly challenging. To this end, research from the Boston Consulting Group found:
“The business environment is inconsistent and full of uncertainty. It changes frequently and lacks the relatively prescribed borders of a machine-learning competition. Moreover, business issues are diverse, ranging from predicting rare events, such as equipment breakdowns, to making ultra-granular decisions about personalized products for individual customers.”
A roadmap to better adoption
Unsurprisingly, machine learning in a business setting has by and large proven to be discouraging. As such, a 2020 study from BCG Gamma and MIT Sloan Management Review found only about 10% of companies to have significantly benefited from the benefits of AI. Hence, to “address the divide between the promise of machine learning and the disappointing results in real-time decision making and other business applications,” certain aspects have to be kept in mind, according to BCG:
- Targeting business value: At most companies, machine learning and AI – the blanket term for algorithms running on data using ML as a segment – are seen as technical problems. However, for AI to have real business value, both the potential and its limitations need to be well understood.
According to experts, the implementation of AI, i.e. industry- and sector-specific business applications, is what will drive competitive gains and digital transformation for companies. By contrast, “horizontal, cross-industry AI solutions, such as adding off-the-shelf image recognition to medical equipment or cameras, will increasingly be less valuable,” according to BCG.
“For this reason, companies should assess the tangible business value of AI through the lens of a 10-20-70 formula. That is, 10% of the effort will lie in building an adequate machine-learning model—an algorithmic set of rules or instructions to help the system learn on its own; 20% will involve high-quality data and technology implementation and innovation, and 70% will focus on developing new business processes or transforming the way business functions operate,” writes BCG.
- Use of external data: Several organizations mistakenly assume the data generated internally is sufficient to drive machine learning models. Collecting relevant exhaustive and accurate data instead, although rather arduous, provides clear competitive advantages based on dynamic changes in environments.
For example, in a project with a major airline to build a forecasting tool to predict consumer demand across different routes of travel pickups post-pandemic, BCG found it more prudent to check external information, such as travel search patterns across demographics, transaction data across airlines, travel restrictions, consumer economic activity, etc, rather than relying solely on the carrier’s reams of historical data.
“With this data, machine-learning forecasts of the number of passengers per route were 20% more accurate than those provided by older predictive systems, resulting in improved flight and crew planning and some $60 million in cost savings in 2020.
The airline’s AI system drove a decision support tool that helped optimize routes according to real-time demand by airport, route, destination, day, and time and helped airline personnel react to scheduling changes and match flight and crew planning to shifting demand,” according to BCG.
- Breaking down complexities: Attempting to build smaller, more concentrated sub-models targeted at the business logic of relatively circumscribed problems may often turn out to be more useful than trying to assemble a vast impenetrable system.
Consider, for example, an AI-based B2B recommender system. In such a setting – where irrelevant suggestions risk an organization losing out on a potential sale – optimizing the recommender system is much more crucial than when compared to systems used by say, Amazon or Netflix, where a few incorrect/irrelevant suggestions wouldn’t risk the losing of customers.
Following the given constraints, consider the following three-step algorithm from BCG: “the first (step) identifies the scope of the project—for instance, whether the initial order indicates that the client requires maintenance or a whole new installation. A second algorithm identifies which product categories are needed to complete the project. And a third algorithm identifies which SKUs best meet the project’s specific requirements.
Importantly, this hierarchy of recommendations naturally supports the sales pitch: first the specific project, then other building activities related to the project, and finally the individual products needed to execute those additional parts of the job. This machine-learning approach increased the distributor’s sales by about 2%.”
Know more about the syllabus and placement record of our Top Ranked Data Science Course in Kolkata, Data Science course in Bangalore, Data Science course in Hyderabad, and Data Science course in Chennai.