Depending on where you get your news, we're either entering a golden age of AI and automation or the robots are coming for our jobs.
Proponents on each side of this coin have been espousing their ideas for more or less a decade, but they all tend to ignore what's emerged as an objective truth: Most business process automation implementations are beset by obstacles including legacy infrastructure, a tendency to amass assets, and a monolithic approach to implementation.
Business leaders tend to think of legacy infrastructure as an investment that's already been made. Unfortunately, as new pay-per-use and outcome-based models emerge, legacy infrastructure is becoming a sunk cost that's difficult to abandon but of dubious value. Substantial investments in hardware and software are in a similar boat, and they're becoming increasingly unnecessary capital expenditures eating into budgets that could be put toward AI and automation systems.
A monolithic approach sabotages an organization's implementation efforts by stretching project timelines and delivery dates, potentially to the point where a completed automation project is no longer relevant to the business by the time it's delivered. A study commissioned by Kofax and conducted by Forrester found that 45% of enterprises attempted to avoid this by opting for ad hoc approaches to automation solutions involving siloed use cases and solutions from many vendors versus a holistic, integrated approach.
Successfully implementing automation systems might never be 'easy,' but plenty of solutions and platforms have emerged to make the process as frictionless as possible for organizations with a wide variety of needs (and budgets). As you begin to construct your company's implementation road map, make sure you focus on the potential of the following three methodologies.
1. Automation-as-a-service model
For organizations with existing (and ideally extensive) sources of data or input information, automation-as-a-service solutions can offer the power of automation at a fraction of the cost.
Automation-as-a-service comes in the form of a prebuilt solution, providing access to the latest automation tools and technologies with the flexibility, agility, and scalability of an as-a-service product. With the level of uncertainty that continues to pervade all kinds of industries, automation-as-a-service is a compelling way to reap the benefits of automation technology without incurring a potentially prohibitive upfront cost.
2. Role-based automation
Role-based automation is essentially the creation of a digital colleague - an automated, AI-powered aide who can perform processes 70% faster at just 50% of the cost.
These digital colleagues will not replace your current employees, but they can augment the capabilities of humans and perform the most mundane and repetitive tasks. Most workers (87%, according to Forrester) are fine with the idea of reskilling to work in conjunction with bots, and business leaders are excited by the prospect of a clear and easily measurable ROI.
3. Do-it-yourself platform
What if your subject matter experts could automate some of the processes that they are responsible for without ever needing to involve an engineering team? Do-it-yourself automation platforms promise these capabilities, with a library or marketplace of available bots designed around automating certain specific business processes that 'citizen developers' or ordinary employees can use without the need for any coding experience. These platforms can help organizations implement companywide automation at scale, without the need for an army of expensive developers and data scientists.
There's no one-size-fits-all strategy to overcome automation obstacles, and the right solution is largely dictated by the maturity of the organization. That means CIOs will need to use a mix of different levers to achieve the desired results. An enterprise-wide approach is a great start, followed by an assessment of existing assets and the possibility of leveraging legacy infrastructure.