Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.

Many firms have adopted automation tools and technologies with great fervor and have seen success in selective projects. Yet, according to a study conducted by Ernst & Young, 30% to 50% of automation projects still fail. Many firms also struggle to scale automation at the enterprise level. 

I have had the privilege of setting up enterprise-level automation and process centers of excellence (COEs) for firms like Bank of America, United Healthcare and most recently at LexisNexis, where I learned hard lessons about what factors inhibit the success of automation and resulting scalability. From these lessons, a clear Enterprise Intelligent Automation Roadmap has emerged, a step-by-step model which can help most organizations achieve holistic and sustainable success. 

Enterprise challenges for automation success

A broad array of automation tools has evolved over the past few years, complicating automation’s adoption and success rate within enterprises. Every tool was built to address a particular class of automation needs, which are broadly defined by the type of data handled, nature of the process (variability), and predictability of the outcomes (whether it’s highly predictable work like simple data entry or more complex research or editorial work where the outcomes vary by who is performing the task). The following graph shows the broad nature of tools and how they align to work patterns.

Image by author

Often particular tools, like robotic process automation (RPA) or business process management (BPM), are chosen and pushed as the holy grail for automation with no upfront assessment of their suitability or capabilities. This is due partly to a lack of knowledge within implementation teams and partly to a push from top executives to justify large investments committed to a particular tool.

VB Event

The AI Impact Tour

Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you!


Learn More

For example, a senior executive might have signed a multi-million-dollar, multi-year license deal with a particular automation vendor. They want to justify the investment by pushing the tool across as many projects as possible. This is when we start to see failures — and as a result, disillusionment with automation at large.  

Another crucial factor that has impacted automation is the culture and discipline of process-driven operations. While the evolution of technologies has broadened automation’s scope and application, the underlying ability to discover, analyze and optimize business processes has not seen much progress until the last five to seven years. The ability to discover current state processes and analyze various KPIs, bottlenecks and patterns is key to successful automation, while the failure to scale can be broadly assigned to the seven critical factors explained below.

Image by author

Value: Lack of a metrics-driven approach to problem discovery and building a solid business case can impede success. The absence of demonstrable impacts after an automation project also contributes to the loss of continued sponsorship and funding.

C-suite sponsorship: When a select few junior leaders implement automation projects without C-suite commitment, we will see departmental success with automation but never enterprise-level adoption. 

Alignment to strategy: When automation is not aligned to corporate strategies that impact profitability, customer experience or revenues, automation loses steam as it will fail to get the required investments of time and money.

Technology and architecture: Lack of holistic technology architecture that combines data, integrations, user experience, and automation-pattern-based standards results in standalone automations that often incur tech debt.

Change management: Automation brings changes across people, processes, operations, systems, data, user interfaces, organization models, and roles, and when undermined, it ends with failed user adoption or backlash from IT and business stakeholders.

Process excellence: Lack of a disciplined approach to current state discovery, analysis, standardization, measurement of KPIs, and designing target state processes is a very common reason for the failure of automations.

Talent: Most firms don’t identify and invest in acquiring or developing the right level of skills across process, technology, change, and program management, which results in the inability to deliver repeatable success in automation.

Pillars of enterprise automation

The seven critical factors that lead to automation failures need to be addressed holistically and not on a gradual basis. Automation success boils down to five core principles:

  1. Process discipline: A disciplined approach to process analysis, design, documentation and architecture that is enforced at the enterprise level
  2. Standardization of automation patterns and tools: IT-led selection, standardization, scalability, and support of automation tools and architectural alignment
  3. Repeatable project and program management: Definition and implementation of automation life cycle phases and related methodologies, delivery models and RACI to align business and technology teams across ideation, qualification, experimentation, delivery and validation
  4. Enablement and training: Establishing an “automation community of practice” that will provide enterprise-level visibility to all programs, knowledge management, and expertise and promote automation across all groups
  5. Governance model: Definition, monitoring and reporting of technology adoption, business cases, financial outcomes, architecture standards and change management

Enterprises will usually go through a phased evolution in their intelligent automation journey, and their adoption of the principles will improve as they continue to demonstrate incremental value with each project. The Enterprise Intelligent Automation Roadmap brings all the principles together in a phased manner, aligned to the three phases of maturity that every enterprise typically goes through. The next section explains the key phases of this roadmap with suggested outcomes.

Enterprise intelligent automation roadmap 

Phase 1: Define

The “Define” phase is the formative phase where:

  • Automation vision and strategy are defined with buy-in from enterprise business and technology leadership.
  • Low-hanging problems with clearly defined outcomes are identified.
  • A combination of automation tools are identified, evaluated and experimented with, and approvals are obtained from the various technology, information security, data privacy and architecture teams.
  • Demonstrable proof points are established with quantifiable value delivered.
  • Business and technology sponsors are identified.

When successful, there is a clear alignment across all executive stakeholders about the role, value and applicability of automation.

Phase 2: Standardize

The “Standardize” phase focuses on creating repeatable models for:

  • People: clear definition of roles and responsibilities, skills, engagement models, training and certifications; team established with industry experts
  • Process: project methodologies, templates, governance, RACI, business case and ROI calculation
  • Technology: tools standardized by automation patterns, scalable infrastructure, implementation models and architecture standards
  • Organization: COE established as a shared service, funding (capital and operational), and key performance objectives
  • Change management: communication and employee alignment for change, feedback mechanisms, operational change planning and implementation resulting from automation, and proactive management of anxiety and fears arising out of automation

When successful, automation becomes a key component of enterprise business transformation programs and every automation project is delivered with predictable outcomes.

Phase 3: Scale

The “Scale” phase is the continuous optimization and expansion phase, where the focus is on achieving efficiencies of scale and standardization, resulting in:

  • Continued improvement in “time to operationalization,” “total cost of ownership” and “time to value” for automation projects
  • Availability of reusable assets and service libraries that will result in acceleration of delivery
  • Evolution from discrete departmental automation projects to enterprise-level automation platforms with measurable impacts across revenue, profitability and customer experience
  • Disciplined approach to aligning automations with value chains, capabilities and processes, resulting in the elimination of redundant automations across enterprise groups
  • Strategic partnerships established with automation vendors
  • Continuous reporting of all automation projects at the enterprise level with measurable cost, value and process alignment and reuse of assets/libraries
  • COE model evolved to target state operating model 

When successful, this phase marks the highest level of sustainable success for an enterprise where demonstration of innovation, delivery and value become ingrained into operating models.

Automation continues to be a key part of digital transformation for most global firms. Firms that approach automation with a holistic view that brings together technology, people, process, organization, and change management are the ones that will derive sustainable and scalable value. The enterprise automation roadmap is a journey and not a point-in-time exercise and will differ for each organization. An agile approach is recommended, where every step is calibrated with the progress of each phase. 

Vinay Mummigatti is EVP of Strategy and Customer Success at Skan AI.


Welcome to the VentureBeat community!

DataDecisionMakers is where experts, including the technical people doing data work, can share data-related insights and innovation.

If you want to read about cutting-edge ideas and up-to-date information, best practices, and the future of data and data tech, join us at DataDecisionMakers.

You might even consider contributing an article of your own!

Read More From DataDecisionMakers