Realizing the Automation Potential: A Machine-First Approach to Achieve Cognitive Operations

Enterprises often focus their digital transformations initiatives on operations that directly touch customers. They build mobile apps for customers to purchase their products and services, or answer their inquiries and complaints through social media. While these efforts are important, we find that relatively few digital transformations zero in on operations such as supply chain, finance, and internal knowledge networks that enable a company to respond to a rapidly changing business environment.

While customers may not have a direct exposure to these operations, transforming is not only essential to improve the customer experience but also to drive cost efficiencies and improvements in profitability, quality and other key metrics.

Let me build on that further.  A new mobile app may make it easier for customers to choose and order a product, but if product doesn’t arrive at the customer’s door on time because the systems and processes are broken, that mobile app is for naught. If the customer credit checking system is faulty or the firm’s inventory system is not accurate (it turns out the product won’t be in stock for a month), the improved digital experience of the mobile app will go only so far.

It is essential for digital transformations to reach much deeper into a company’s operations. They can do so by:

  • Predicting the demand patterns by product type and by location to ensure that warehousing and distribution can accommodate orders. It is equally important for each customer order to trigger an appropriate internal production plan, including orders for raw materials.
  • Ensuring staffing plans for production and distribution are adequate to handle the orders. In turn, this will require systems that track staffing needs, worker availability and other factors.
  • Identifying peak and trough periods including seasonal variations to make sure supply can meet demand.

The good news is that the data and the technology to make that possible are available in abundance. What happens behind the scenes in a company can be automated and enhanced with powerful digital technologies such as artificial intelligence software, cloud computing, and analytics. We refer to this as a machine-first approach to digital transformation.

The Key Elements of a Machine-First Approach

The machine-first approach to creating intelligent operations has three main elements:

  • Using automation technologies to take over routine manual tasks and processes.
  • Extracting insights from data generated by those automated processes – insights that direct the next steps to take. This means giving machines the “first right of refusal” to taking on work. By this, we mean letting computers execute an action rather than people, if possible, and having those people perform more complex tasks that can’t be automated today.
  • Identifying process disruptions that are about to occur and taking action to prevent or correct them.

Two factors make it possible today to take a machine-first approach to digital transformation. The first is analytics and AI. The insights and automated tasks enabled by these technologies can make business processes more effective and efficient (faster, more accurate and less costly).

The second factor is an agile approach to improving business processes, driven by design thinking. In agile, a team conducts rapid trials and makes continual improvements. Design thinking places the objectives of customers and other stakeholders at the center of designing business processes.

Data at the Core of Cognitive Operations

Taking a machine-first approach to building cognitive operations requires executives to move on three fronts:

  • Moving from being process-centric to being data-centric. With data proliferating from a variety of sources, enterprises must translate it into information that makes work more effective and efficient. For example, a product development department should get real-time information on who is using the company’s products and services, and where and when they use them. If the data shows that a certain feature-set has no appeal to a customer segment, even though it may be technologically superior, product developers need to rethink the product.
  • Automating not just the operation but the delivery of insights. Intelligent automation involves extracting data-derived insights while eliminating effort, and sharing the insights where they are needed in an organization. Business operations for filling orders, responding to customers, ordering supplies, and so on must be conducted automatically without human intervention. Systems supporting a cognitive operation provide alerts that predict problems before they happen.
  • Providing real-time visibility for people to act. With automated systems delivering real-time alerts, cognitive operations give managers and workers insights to perform tasks that machines cannot. This requires a business operations center that displays well-designed data visualizations and useful dashboards that instruct people to take actions when necessary.

Today enterprise leaders can take advantage of machine learning, AI and advanced analytics capabilities to identify operational inefficiencies and determine how to fix them. In many cases, this does not require a huge overhaul of business operations. Companies often can fine-tune their business processes and – thanks to automation – see benefits in short order.

However, leaders need to constantly eliminate redundant and outmoded systems and processes, while investing in systems and processes that generate better results.

The Cognitive Operations Difference

Companies that have used a machine-first approach to designing cognitive operations illustrate the advantages of this approach.

Firms we have worked with have experienced the machine-first difference throughout their operations:

  • Improved cash flow: An automated cash forecasting system helped a British steel manufacturer analyze data from finance and other departments to reduce disputes and increased collections 28% in a market that is seeing decreased demand.
  • Significant uptime improvement: The machine-first approach at a global telecommunications company’s command center has reduced the downtime by 25%, increasing reliability of its 600-plus retail and business voice, data, video and security systems, which serve millions of customers around the clock.
  • Procurement savings: Using AI software to scan purchase requests across multiple plants and vendors has helped a global enterprise better understand the behavior that drives purchasing price variances across plants and to reduce annual expenses by $3.2 million.
  • Self-healing IT infrastructure: Deploying automated health checks for the PoS systems and IT infrastructure at a U.S. fashion retailer automatically resolved 95% of incidents detected. This was especially important during the peak shopping season when incidents increase significantly due to sheer volume of transactions processed. Proactive incident prevention improved mean-time-to-repair of incidents and provided 100% uptime for critical business applications.
  • Improving customer onboarding: Better integration of applications, and orchestration of the order-to-activate-service process with a case manager responsible for each order helped a leading telecom provider cut down order activation time to half while the Net Promoter Score rose to 20 from an abysmal minus-44. The company estimates the effort will cut costs by about $70 million over three years.
  • Improved employee experience with automated service fulfillment: Automation of more than 70 IT application- and infrastructure-related requests from end users at a global research company helped streamline the request process, leading to faster resolution times and improved end-user satisfaction.
  • 100% auditing of travel and expense reports: a major airline implemented new intelligent processes to automated much of its employee travel and expense management system, enabling the 20-member back-office team to audit 100% of expense reports instead of its past practices of auditing 20%.

These examples show the power of a machine-first approach to designing important but cost-intensive business operations and embedding them with automated intelligence. Companies that have done so have learned that any old method won’t get them there – only one that helps them reconceive what is possible in an increasingly digital world.


About the author(s)

Ashok Pai

Ashok Pai is Vice President & Global Head of Cognitive Business Operations at TCS. He helps organizations leverage digital technologies to reimagine their business models, products, processes, and services. He specializes in helping enterprises make major cost and operational improvements through digital transformation initiatives. Prior to this role, he headed TCS’ Business Process Services team and was responsible for more than 100 customers in North America, Asia Pacific, and other regions. He joined TCS more than 25 years ago, right after earning his Master’s degree from the Indian Institute of Technology (IIT) in Mumbai.

Continue your journey


  • Operating in an Insight-Driven World: Mastering the New Game of Analytics


    Every large company today is swimming in a rising tide of digital data. That data resides in their information systems and in the ecosystem that surrounds them – the data of suppliers, customers, distribution channel partners, and others with which…

  • Taking a Machine-First Approach to Cloud Migration


    Many companies have put cloud computing at the center of their digital transformations. Yet a number are struggling to shift their systems from on-premises technology to public clouds. Several challenges are slowing them down, but they can be overcome with…