Many organizations aspire to automation levels that rival the likes of Amazon, where processes flow seamlessly between people, systems, and devices.
A recent Gartner report found businesses are evolving their use of artificial intelligence (AI) as part of their automation strategies, with one third of organizations surveyed applying AI across several business units.
If they’re honest with themselves, these organizations want automation to achieve some sort of goal — whether that’s improving the customer experience, increasing cost efficiencies, or making life easier for employees.
Automation can and should be applied to each of these goals, but it should be done in a fully orchestrated way across the entire organization.
Jakob Freund, co-founder and CEO of Camunda, an open-source workflow and decision automation platform, says today, many organizations automate locally — or within a single team, system, or device.
“That’s a mistake, since most processes are far more complex than that,” he says. “These organizations don’t need more automation for automation’s sake. They need process orchestration.”
Process orchestration coordinates the various moving parts (or endpoints) of a business process, and sometimes even ties multiple processes together across an organization.
“Process orchestration increases the overall level of automation by integrating disparate local automations, increasing visibility into processes, and making processes easier to change,” Freund points out.
Automation and AI: The Perfect Marriage
Muddu Sadhakar, CEO of Aisera, the developer of an AI-powered platform to automate customer service requests, says AI combined with automation is the “perfect marriage” that enterprises need going forward.
“One without the other will make enterprises weak and suboptimal,” he says.
From his perspective, the benefit of an automation strategy that runs through the organization can be summed up in three words: Cost, cost, cost.
“Automation helps create deflation in current inflation market conditions,” Sadhakar explains. “Other benefits are user productivity and enabling users with more time to think and be creative. Automation strategies that run through the organization also help to scale and grow companies efficiently without adding unnecessary resources.”
He points out that AI is still in early days — what he calls the first inning of a 10 inning game.
“We need leaders and C-level people to make enough bets and stay with AI projects, and those projects need money, resources and time allocated towards them,” he says. “These will take time to see results — patience is key. This is the main reason why AI has significant challenges.”
Wayne Butterfield, partner with global technology research and advisory firm ISG, agrees when used under the right circumstances, for the right reasons, automation and AI are not just “nice to have” but a necessity — especially when it comes to improving customer experiences.
“Hyper-targeted marketing campaigns, super accurate delivery windows, and personalized, proactive services all need AI to become a reality,” he says.
To be successful in transforming the way the business runs, the use of automation needs to be both strategic and cultural.
“It starts with the right strategy, of course, aligned with business outcomes, but for automation to deliver true business benefits, it needs to be embedded in your culture,” he says. “Your people really need to understand and believe they will be more efficient and have greater job satisfaction if they work alongside an automated co-worker.”
Butterfield adds that they also need to realize the business overall will benefit from providing customers with better, more personalized, and timely support through automation.
AI-Based Automation Needs Focused Leadership
Sadhakar laments the fact that currently no one is responsible for automation policy in the organization and enterprise and says CEOs and CFOs must put more focus and prioritization on automation policies.
“Going forward, automation should be the focus at each business group and department – it must be a mandatory part of business planning,” he explains.
Each C-level executive should provide a plan of how much automation each quarter/year they plan to implement to reduce the number of resources their division needs.
They should also come up with tangible KPIs that will impact cost reduction and generate savings for sustained growth.
Butterfield says this focus must come from the board as a priority item, enabled through technology and implemented by all.
“AI and automation are as much a capability as a technology – therefore, even if someone is taking responsibility, the organization will only be successful if everyone is aligned,” he says.
Freund says that while automation involves a broad set of stakeholders across both business and IT, it’s not always easy to get them all on the same page.
Depending on the organization, a technical leader like an enterprise architect might spearhead the automation process by kicking off a proof of concept (PoC), organizing a team to execute it, and presenting the results to business stakeholders.
“Once stakeholders are convinced of a project’s feasibility, the PoC team can lead process orchestration efforts across the organization,” he says. “They’ll plan the strategic order of projects and tie them together with end-to-end process orchestration technology to avoid automating in silos.”
Scaling Automation Essential to Competitive Survival
Freund notes there are a lot of challenges in scaling AI: First, AI is just one of many process endpoints. “Integrating it seamlessly with legacy systems, microservices architectures and SaaS applications can be a big challenge,” he says.
Second, there is a major shortage of AI and data science talent, which can prevent larger AI projects from getting off the ground and gaining the momentum they should.
And finally, AI can be resource-intensive and involve extensive amounts of computing power that not every organization can afford.
He notes that process orchestration can help scale AI within an automation context by tying this technology together with other process endpoints across the organization.
“Beyond that, many AutoML companies have taken steps to lower the barrier to entry for AI so these other obstacles aren’t as big,” Freund says.
The way Butterfield sees things, automation and AI are both more than a technology; they are a different way of doing things.
“You need trust and change management to extract real value from automation,” he explains. “Automation is much more difficult than a typical technology implementation, as it touches so many people in virtually every corner of the business.”
Scaling isn’t easy because it involves people, and it takes time to build trust and demonstrate the value needed to push the doors of widescale adoption open.
“It is not ‘if’ AI and Automation will scale, but ‘when’,” he adds. “Those that do not scale will simply become uncompetitive, both from a service and cost perspective, and be relegated to the dustbin of business history.”