There was a time when businesses deployed artificial intelligence (AI) to automate existing, rules-based processes.
That was then. Today, some leading companies are going steps further, using advanced AI to implement outcome-based approaches.
As companies make the transition, a critical consideration for business leaders: re-thinking how they structure their service-level agreements (SLAs).
Why? For one thing, when a client defines business outcomes, service providers should be empowered to determine how best to achieve those outcomes.
Help desks, for example. In the past, help desk SLAs were typically structured with the stipulation that a certain percentage of customer inquiries or complaints were to be resolved in a specific amount of time, and the process might be based on rigid rules, like first-in, first-out (FIFO), to help prioritize responses.
Nowadays, sophisticated natural-language processing algorithms can be used to help identify issues faster and enable better prioritization based on a variety of criteria, including the urgency of a customer’s need. Moreover, such AI systems might be adept at handling perhaps 80 percent of emails with routine requests, leaving the remaining 20 percent to be handled by humans, who can then focus on the more complex cases, including those involving dissatisfied customers.
In essence, the use of advanced AI enables deeper human-machine partnerships that amplify each side specific strengths—the latter’s in processing huge amounts of data and handling repetitive tedious tasks, and the former’s in exercising judgement and interpersonal skills. Now, SLAs can be structured with a focus on a desired outcome, such as increased customer satisfaction.
Another example: the procurement process, which requires the analysis of data from multiple public and proprietary sources to help firms make better decisions regarding which suppliers to do business with.
This process was traditionally executed manually, and SLAs might have required the production of a certain number of intelligence reports in a specific time frame.
Now, AI systems can handle the initial processing and analyzing of all that data, which allows human procurement experts to cover three times the material than previously possible. This, in turn, enables those experts to better uncover the hidden risks of using certain suppliers, and the result is more optimal sourcing decisions.
Accordingly, SLAs can then be structured based on outcomes—for example, using incentives tied to the impact from enhanced insights obtained through the procurement market intelligence.
There’s something else to consider: AI systems, like the procurement assistant described above, can lead to higher job satisfaction for employees – in other words, a more engaged workforce—because people are freed from the more mundane and repetitive aspects of their daily work.
As such, for SLAs that cover AI systems, a business might consider incorporating metrics to account for “softer” factors, like employee engagement.
In sum, as companies expand use of AI for more than just mechanistic automation, they must consider moving beyond standard SLAs that typically focused on improving the efficiency and lowering the costs of existing rules-based processes.
This new reality offers considerable upside. Sophisticated AI systems are now enabling firms to make the transition to outcome-based approaches, and SLAs need to evolve accordingly to reflect that fundamental change.