No more compromise between user experience, efficiency, and effectiveness

We looked previously at the flexible infrastructure of AI that’s accessible to companies today – and how this is accelerating a shift to hyperintelligent enterprises. We now consider the consequences of AI on that most common of enterprise projects – business transformation.

The traditional approach to business transformation is to set a future vision of how the enterprise will operate. A time horizon of three to five years is set to prepare for the transition to this new operating model. Wholesale changes occur over this planned transition period, and teams are set goals and milestones.

When successful, it’s a careful and considered journey, but common challenges tend to impair progress:

  • Internal policy changes to support the overall goals are often not applied or enforced consistently.
  • Momentum for change stalls as new business priorities or opportunities arise.
  • A poor user experience leads to resistance to change and low adoption of new ways of working.

Transformation that applies an AI-first approach can help overcome these barriers. It allows operating models to be transformed incrementally and with minimal disruption over a shorter period. More importantly, it does not demand compromises to be made between user experience, efficiency and effectiveness.

Business transformation with AI creates new routes to success

A typical traditional transformation plan has a five-year time horizon, characterized by teams working independently on separate project streams. At agreed points in time, the teams come together to report on progress and highlight any issues, delays or dependencies.

Where required, remediation plans are prepared and new milestones are scheduled. Everyone works on their separate streams until the next meeting. It is a linear or waterfall approach, driven by time, goals and budgets. This is often represented visually in the form of a “T-Map” (see example below).

Working as multiple teams focused on time based deliverables

In contrast, an AI-driven transformation plan drives incremental changes using cross-functional teams working in an agile manner. There is shared buy-in to agreed outcomes and success is measured by adoption and customer/user feedback, rather than the completion of milestones across separate workstreams. The “AI T-Map” below shows how this differs from the traditional approach.

Working as one team focused on business outcomes

Take the example of an organization that wants to transform its customer journey and offer the most user-friendly signup experience. First, it needs to apply design thinking to create the “golden path” for new customers by adopting an AI-first approach. This will define a new operating model and involve changes in policy, people, process, technology and data to transform the way that the service is delivered.

To be successful, multi-disciplinary teams must work together, sharing knowledge and collaborating on tasks. Incremental progress is made as they align different parts of the operating model, including the required adoption of policy changes, to deliver the improved experience. Rather than working linearly and reporting back at set times, they will be working in an agile way to deliver agreed outcomes as a team, testing changes and taking customer/user feedback as they proceed.

An AI-first approach delivers a better user experience, but also improved efficiency and effectiveness

Transformation programs have typically focused on efficiency. Applying “lean” methods to processes has helped to cut waste and reduce cost, but often at the expense of effectiveness (service levels) or customer/user experience.

However, businesses are increasingly measuring their success by the quality of their customer feedback, using tools such as net promoter score (NPS). This is driving an AI-first approach to transformation, ensuring that user experience and effectiveness (both of which result in incremental business value) are prioritized in the program goals – whilst continuing to deliver cost efficiencies.

In a survey carried out by Capgemini last year, the leading adopters in driving AI transformation identified the need for a methodology framework and multi-disciplinary teams as their two critical success factors.

We use the “Five Senses of Intelligent Automation” framework to facilitate a design-thinking approach, prioritizing technologies that enable businesses to build scalable operating models that improve how they:

  • Monitor what’s happening in their environment.
  • Communicate and interact with customers/users.
  • Store data and enrich their knowledge
  • Analyze that knowledge to form insights.
  • Act on the insights to improve operations and outcomes.

This is how we create and improve the “golden path.” Multi-disciplinary teams apply design thinking to establish the best experience from a user perspective in the five areas above, both from a standalone and a combined point of view. They then consider how new AI technologies can support those objectives. Instead of focusing on individual processes and viewing them as a linear discrete journey, they look at the whole connected experience.

When making this transition to a hyperintelligent enterprise, tools such as Celonis (a process-mining application) can automate the monitoring and analysis of transactions. Tracking and understanding deviations from the “golden path” enables further improvements to be made to the new operating model, including compliance with required policy changes.

AI creates a portfolio approach to transformation

Using AI-first design thinking to create a new business operating model requires the adoption of an agile portfolio approach to transformation.

This approach will yield benefits more quickly and with less business disruption than, for example, five-year ERP-led transformations or two-year offshoring programs operating in isolation.

Multiple projects are integrated and run simultaneously, bringing high-value business improvements that enhance the user experience and deliver incremental value, whilst optimizing existing investments.

AI-first transformation – the best of both worlds

The classical approach to transformation in the West is to drive through large-scale change in pursuit of specific results. Whereas in the East, transformation is more of a continuous process that embraces emerging opportunities.

AI-first transformation blends the best of both approaches. It allows organizations to make considered, incremental changes over time that align to their longer-term transformation goals.

Start now, start AI SMART
Early adopters of AI also identified the setting and measurement of clear objectives as key enablers for the success of their AI transformation programs.

We recommend an AI SMART approach to making sure that any measures align to the following criteria:

  • S – Strategic – ensure that the design principles for the transformation contribute to the strategic vision for the enterprise
  • M – Motivational – communicate positive messages to a broad cross-section of the organization to make sure that changes to ways of working are understood and fully supported
  • A – Adoption – rather than allow exceptions to the new operating model, fix the root causes for deviation and reward compliance
  • R – Relevant – priorities impacts on customer satisfaction and service levels and not just cost reduction
  • T – Technology – use tools like Celonis to automate monitoring and analysis.

Making the vision a reality
In this article, we have introduced the concept of AI-first transformation. This is inspired by our strong belief in the power of design thinking. It emphasizes the need to visualize the AI outcomes required to achieve successful business transformation in the new age of hyperintelligent enterprises.

This AI-first transformation approach is supported by two distinct methodologies. The first of these was explained in chapter 2 and is applied in the design phase to create a vision of the future operating model. An example of this would be the “golden path” of a transaction or interaction that we considered earlier.

In the next article of our hyperintelligent enterprise series, we will introduce the second methodology. This is used during the implementation phase to turn the design into a reality. It is called our ESOAR approach, and it comprises the following five steps:

  • E – Eliminate all unnecessary and sub-optimal transactions/interactions
  • S – Standardize the operating model for the golden path
  • O – Optimize existing investments
  • A – Automate to create new AI solutions
  • R – Robotize where appropriate

ESOAR is complementary to the “Five Senses of Intelligent Automation” design methodology, and the best results are achieved when they are operated together. When design and implementation teams work jointly throughout the transformation journey, incremental changes can be made with confidence, whilst delivering step-change improvements in service delivery.