Article

Outsourcing in a digital age: one size does not fit all

Emerging technologies like Robotic Process Automation, AI and advanced analytics have shaken up the outsourcing landscape, changing how companies approach routine tasks

“If you deprive yourself of outsourcing and your competitors do not, you’re putting yourself out of business,” Lee Kuan Yew, the first prime minister of Singapore, once said. Virtually every major enterprise has followed this advice for decades, outsourcing routine processes and IT operations to third parties in order to save money and focus on their core business objectives. But with wages increasing in traditional outsourcing hotspots like India and China, and mounting political pressure to keep jobs onshore, the golden era of outsourcing may be coming to an end.

Emerging technologies like Robotic Process Automation (RPA), artificial intelligence (AI) and advanced analytics are also changing the way firms approach outsourcing, while retaining many of its benefits. In fact, RPA has been described as outsourcing to a software robot rather than a foreign country.

If you deprive yourself of outsourcing and your competitors do not, you’re putting yourself out of business

Lee Kuan Yew, first prime minister of Singapore

A 2018 Harvey Nash/​KPMG survey found that 31 per cent of chief information officers worldwide are now investing in RPA “minimally, moderately or significantly.” Pat Geary, chief evangelist at Blue Prism – one of the pioneers of RPA – says it solves the “human middleware” issue that exists in enterprise environments.

“This is where human workers act as ‘human-​robots’, performing repetitive tasks to bridge system processes that aren’t integrated or that have no system-​based interoperability,” he explains.

This creates new advantages and opportunities for enterprises of all sizes, says Brian Terrell, managing partner at BTerrell Group, which specialises in providing accounting, finance and automation technology to mid-​market companies. “Unless a business process absolutely requires an authentic human actor, [RPA] offers new gains in speed, accuracy, and economy by interacting with a business critical application directly through that application’s user interface.”

For example, an RPA bot can quickly log into a system, open an application, enter data, and log out again. Copy and paste errors are typically reduced, and tasks are completed faster. As bots can be deployed at much less cost than a human worker and scaled relatively easily, they can also allow business leaders to re-​evaluate outsourcing and automation possibilities previously thought to generate an insufficient return on investment.

Of course, bots require training before they can be deployed. Larger companies typically invest in both human and technical resources to support large-​scale RPA and AI implementations, and engage service providers to bring these initiatives to scale. “Mid-​market and smaller companies may opt for a more measured and affordable approach by relying on service providers to manage the development, support, and maintenance of an entire implementation ‘as a service’,” says Mr Terrell, “such as the Robotics-​as-​a-​Service offerings currently being offered by some service providers.”

Mr Geary says that for larger organisations, the key to a successful implementation lies in gaining company-​wide support from key stakeholders early on in the process. “Getting the IT department involved at the beginning is key too, as they support RPA on a number of critical fronts, such as compliance with IT security, auditability, help with infrastructure configuration and scalability.”

Leslie Willcocks, professor of work, technology and globalisation at the London School of Economics’ Department of Management, says that while reshoring does happen as a result of RPA, it’s not a major movement – at least not yet. “Generally these disruptive tools produce much more business value the more they are scaled,” he adds. “Scale can also justify the large amounts of effort and money that needs to go into getting the data right and training the robots.”

Done right, RPA typically delivers savings of 25–50 per cent, according to the Institute for Robotic Process Automation & Artificial Intelligence. But if changes are made to an interface or any other aspect of a legacy app, it can cause problems for the bots. What’s more, standard RPA tools can only handle structured data – anything unstructured, such as emails, needs a more advanced form of AI-​infused RPA known as cognitive automation.

While machine learning and AI tools are generally more complex to implement than RPA, they can supercharge companies’ digital transformation efforts. The success of these advanced tools rests upon reliable data sets and systems and people that are capable of working seamlessly with such technologies. None of this is easy, but achieving a high level of “intelligent automation” across the business could unlock a major competitive advantage.

That’s also true of advanced analytics, which can help enterprises of all sizes predict future trends, events and customer behaviours, and improve their efficiency and agility. As firms grow, advanced analytics tools can also help them see which business processes need to be extended, and which are no longer fit for purpose.

All these tools can potentially change the nature and scope of outsourcing contracts. In fact, as counterintuitive as it might sound, many outsourcing providers are using RPA and cognitive automation tools to automate low-​level tasks and reduce staff overheads – something companies may increasingly insist upon when awarding new contracts.

Outsourcing will have to change its skill set to survive in the era of automation

“[Outsourcing] providers obviously have to move to more automation without cannibalising too fast or too badly their more traditional labour arbitrage business model,” says Mr Willcocks, “and they have to find ways of sharing the gains from utilising automation.”

In short, outsourcing will have to change its skill set to survive in the era of automation – just like the rest of us.