Evolving AI, data platforms and automation technologies are creating opportunities to reimagine entire customer value chains that were originally put in place with the capabilities then available. Technology executives embarking on this journey of rethinking their customer value chains are transforming their businesses, and driving growth, efficiency, and durability in their business models.
This document is a framework for navigating the transformation landscape across the plan-to-cash cycle and provides learnings and best practices from the collective intelligence of leading CIOs/CDOs/CTOs in the Executive Technology Board (under Chatham House Rules). It’s worthwhile to explore this in four chunks – enterprise architecture and ecosystems; harvesting data across the value chain; AI use cases; and transforming customer operations.
Enterprise Architecture and Ecosystems
Agile and adaptive enterprise architectures
Market volatility and changing conditions are highlighting the importance for enterprise applications to be super agile and supremely adaptive. But there should be a balance between accommodating local market needs and regulatory requirements flexibly and standardizing processes and systems to drive global efficiency. The key to solving this is being intentional about enterprise architectures that balance front-end customization with back-end standardization – commonly represented by a 'thin core'.
The duality of deep local competitiveness and strong global standards.
The Apple iPhone application model serves as a good example of a way to solve for the duality of local/global - it is built on one single operating system but provides for a variety of apps. Best in class enterprise architectures are leveraging this “app store model” concept – recognizing that in the Apple ecosystem, the application flexibility is only possible because what is “under the hood” is extremely standardized. A similar “super-modular” approach to a global enterprise architecture includes a lean core made up of a minimum number of products with a long shelf life, along with best-in-class applications at the edge that address local market requirements.
Being intentional about designing for changing geopolitics.
Each corporation has to assess and weigh its own risks in the geopolitical arena - and work out its own responsibilities, for instance as it relates to sovereignty laws and associated implications on licensing, data residency, and control over IP. But what is just as important is the realization that geopolitical conditions are constantly changing and enterprise architectures must be designed for flexibility. A case in point is China, where many foreign corporations two years ago were focused on “segmenting and separating”, leading to significant duplication of costs – but have switched to a model with strong identity management, deep security monitoring and well-defined resolution plans for when things go wrong – but the overall architecture remains global.
Thin core with a compelling, competitive edge.
An inherently monolithic and cumbersome legacy base, highly risky and expensive upgrades and migrations, and growing allocation of the overall IT budget towards other areas like AI, Cybersecurity, etc are all driving the need for a leaner core – on the ERP (sometimes on the CRM) side. And successful enterprises are complementing this thinner core with lighter-weight external tools from the partner ecosystem.
The potential of autonomous processing.
A progressive evolution from a “thick core” to a “lean core” to an “autonomous core” is clearly in the making and requires careful consideration. Data proximity is key for agentic AI, and the next versions of enterprise architecture will feature more autonomous processing in and close to the ERPs. Similarly, much of the AI, BI and GenAI layers in the enterprise can be intentionally architected separate from and around a core enterprise data platform.
Domain and business requirements rule.
In the end though a “pure play” enterprise architecture approach can be limiting for the business and therefore a non-starter. The balance of in-market relationships and automation needs to be carefully weighed - for instance with small retailers, although there are benefits to automating orders, digital ordering can significantly reduce customer engagement - a non-desirable outcome. To address this, we have seen clever architectures use gamification in automation – as an example through loyalty points that preserve / build digital relationships. In another example of domain specific nuances to enterprise architectures, consumer goods companies have built out inexpensive ERPs so they can provide it to small-market customers and foster integrated ordering processes with automation and better data readiness. Finally prioritizing enterprise architectures with adoption in mind is critical – changes in your own sales forces work, for instance, store prioritization and route planning applications are always first on the list before tools that can get in the way of local market relationships.
Harvesting Data across the Value Chain
A Comprehensive approach to Data.
As strategic and as foundational products and platforms (system of record, intelligence and engagement) can be, it’s the Data layer that is seen as the true driver of success in Enterprise Architectures. For most corporations this journey starts with getting a good handle on data and its sources – across internal systems including the ERP, CRM, supply chain, logistics, customer care and other internal applications. And include external sources that provide, for instance, customer usage, search and purchase information, competitor information, social media, market data, and sell-out information from retailers.
Modular governance model.
Data drift is still a challenge and therefore many enterprises are putting in place a certification process for data accuracy. A bronze/silver/gold data layer template is often helpful in segmenting data based on the rigor and standardization necessary to the scope of its strategic impact. A centrally managed master data system and a rigorous master data organization – a gold layer of sorts - achieves seamless integration and facilitates deep, data-driven decision-making. A bronze layer on the other hand facilitates local flexibility in data models and data layers that are most relevant for local decision making. This clarity also ensures ownership close to where the data is created – and used.
Data Quality is the last unsolved problem in the Enterprise.
Across corporations, data quality remains the key concern. Harmonizing disparate formats and structures into one integrated data platform, implementing a solid master data management program, and validating and cleansing data regularly improves accuracy, consistency and usability. Spend numbers differ across companies, but up to a third of total cost of data can be from people costs – across data management, data engineering, and some data science, diluting the spend on data infrastructure, and there are early signs and some initial positive results from using GenAI-powered data cleansing approaches. It’s important to note that companies that have a data driven culture have a cleaner data model – since the awareness and usage of data promotes a fundamental ownership of data quality.
Standard corporate reports and empowered data citizens.
Effective data governance strikes a crucial balance between empowering data citizens to easily access and explore relevant data - and establishing standardized reporting frameworks that ensure management have consistent, reliable, and actionable insights to inform strategic decision-making.
Creating the ability for multiple parts of the organization to pull data is a great opportunity for “data citizens” to be more comfortable working with data and using it more effectively in their business functions. As an example, getting real-time visibility on red (dilutive) promotions around the world provides teams with the needed agility and decision making in trade promotion process. GenAI can also be an effective way to engage data citizens in interacting with the data by allowing business users to ask data questions in simple language through questions such as “if I change my marketing spend, what impact will it have”?
But it is important to differentiate between management reports and business insights – the former should be set and inunalterable with a governed data hierarchy in place, and certified data for analysis. A notable challenge with management reporting is report proliferation, and a “Netflix-style” portal with peer reviews, and star ratings can be an effective way to reduce the number of reports to the most useful ones, as well as change the model from push to pull – a member reported reducing finance and supply chain reports from 1900 to 27 reports!
Process Integration and Change Management are key.
Data insights start to bring value when they are integrated into a real business process – e.g. content production, sales planning etc – and across the process, end to end. As an example, across the product life cycle – from ideation to launch to production. During the product ideation process, we see successful examples of GenAI created “synthetic human focus groups”– a sort of “digital twin of existing customers” to generate ideas. At launch, customer feedback is incorporated to refine product design with “voice of the customer” initiatives that identify quality issues that can be addressed earlier in the product lifecycle and help build more customer-centric designs. For example, social media posts from end consumers showing how to open a package provide valuable information to the product teams on how to redesign future packages. After launch, data teams are also creating digital twins of the market for effective scenario planning, or using ML models to review store potential and asset optimization. And data is used to build customer loyalty programs more effectively and in a most cost-effective way.
At the end of the day, effective change management is crucial for driving digital and data transformation, particularly in customer operations. As companies implement new technologies to transform customer experiences, change management addresses cultural, process, and technological shifts – and by proactively planning adoption, companies ensure that digital and data transformation initiatives, such as AI-powered customer service or automated workflows, achieve their intended goals.
Top Gen AI Use cases
From Pilot to Production.
As organizations unlock the potential of harvested data for business insights, they also lay the foundation for the next wave of innovation: leveraging data to fuel Generative AI applications that drive transformative change in enterprise operations, customer experiences, and competitive advantage. And over the last six months we have moved from using Generative AI in proof of concept initiatives into to fully deployed use cases at scale in some areas. It’s probably helpful to look at these in three segments - Individual productivity, Team Productivity and Topline Growth.
AI enhances personal productivity in many areas such as coding efficiency (reducing development time by 20-30%), knowledge management (for instance interrogating field service manuals and generating best resolution plans). Team productivity increases with AI-driven content creation (30% reduction in time to market), or brand auditing and compliance automation (25% accuracy improvement), or risk management (proprietary LLMs for enterprise risk services now assess risk dynamically in the flow of work). Finally, early examples of AI fueling topline growth can be seen in new product and packaging designs, developing innovative materials through generative chemistry, or analyzing core compositions for drilling decisions in the energy sector.
Protecting the Core.
Generative AI is bringing new solutions to old problems at an unprecedented rate – and many quick fixes today appear to be differentiative based on early time-to-market advantages. Across corporations that are furthest along on the Generative AI journey, one of the largest learnings has been the strategic importance of knowing what is core today, and acknowledging that this core changes over time. To put this in perspective with an example, most leaders today would agree that data centers are below the commodity line while LLM teams are differentiative and therefore above the line – but tomorrow this line may change – and the key is remaining focused on understanding and protecting the core, especially the core talent, and not getting locked out of your enterprise data.
At the end of the day, the success of AI initiatives hinges on effective governance and organizational structure. As AI models become increasingly sophisticated, questions arise about ownership, scaling ML expertise, and ensuring ethical and responsible AI development.
A Framework for Transforming Customer Operations
In the journey to transform customer operations, technology leaders play a key role in improving efficiency, agility, and innovation. This framework is designed to guide technology leaders through the key principles and actionable steps required to achieve operational excellence, using insights drawn from a successful global transformation at a leading consumer goods company.
An Executive Technology Board member company, like many other global enterprises, faced challenges with inefficient processes, outdated technology, and siloed operations, leading to issues like inaccurate forecasting, high overdue debt, and low operational efficiency. The transformation focused on optimizing processes, automating through technology, and building a global network of hubs to drive efficiency and value creation.
Key Components of Customer Operations Transformation
The key elements of the transformation initiative are 1) Process Redesign and Standardization (redesigning hundreds of country-specific processes to a dozen global end-to-end processes, focusing on cross-functional optimization), 2) Establishing a small number of global hubs with a mix of company and partner employees, centralizing operations and fostering talent development and 3) Technology Upgrade (from hundreds of bespoke systems to a dozen best of breed digital platforms (such as Kinaxis, Pega, and SAP to enhance planning, order management, and data management) as well as a few key analytics solutions built internally).
From this experience, we can draw Key Design Principles for driving a successful customer operations transformation.
Shifting the Center of Gravity: Shifting the control and ownership of processes to shared service centers empowers them to drive change and efficiency. In the decision to “transform first or transfer first”, transferring the power to the hubs makes them accountable for the transformation. Hubs are the digesters of complexity and the engines of harmonization and automation.
Thorough End-to-End Process Optimization: It is recommended to focus on optimizing entire processes, rather than individual functions, to address cross-functional challenges. Through the process, the focus is on Transfer, Eliminate and Transform – as processes are transferred to the shared service centers, redundant processes are eliminated, and existing processes are transformed for greater efficiency.
Systematically Breaking Down Silos: The transformation program can break down silos by organizing processes into four key stages: Plan, Execute, Deliver, and Collect. This ensures seamless integration across departments.
Business Outcome Focus: Prioritizing business outcomes over specific tools ensures long-term sustainability and adaptability. Examples of business outcomes include improved forecast accuracy, reduced overdue debt, optimized promotions and elevated customer service.
Synchronicity of Digital and Operations: Creating joint teams of technology and business experts that work together on the transformation is a key success factor. The transformation program leverages digital technologies to automate processes, improve data management, and enhance operational efficiency – but best results are achieved when technology is serving a business goal.
Design of a Global Standardized Blueprint: A standardized global blueprint ensures consistency across all markets. Standardization, while initially challenging, unlocks significant value in terms of efficiency, cost reduction, and customer satisfaction. Not surprisingly, labor arbitrage is not as significant a value driver as standardization.
Leverage Global/Local Deployment Partnership: Collaborating with global and local partners ensures successful implementation across markets. Several hundred go-lives can occur annually, with each market accountable for delivering the business case.
Robust Governance and Phased Market-by Market Transition: This ensures alignment and accountability across the global and local levels. A three-stage runway model - planning, deployment, and run - ensures a structured and iterative implementation process. A phased approach market by market allows for testing and fixing at each step.
Importance of Casting: The success of such a transformation project hinges on selecting the right people with the right business and technology skills and experience to lead and execute the transformation.
Getting Key Metrics Right Upfront: Transforming Customer Operations can have a significant impact on performance, and it is therefore critical to define which metrics the transformation should measure. They can range from cash flow (reduction in overdue payments) to efficiency (reduction in lead times for delivery, claims settling, and cash collection), forecast accuracy (to reduce the need for excess inventory), promotional management and customer service.
Continuous Evolution and Future Focus
As transformation continues in the future, new areas of focus will likely be AI Deployment to further automate processes and enhance customer interactions, Agentic AI to handle basic customer inquiries and free up human resources and Continued Standardization, further consolidating systems and processes to drive even greater efficiency.
Executive Technology Board (c) | North America & Europe