AI and the customer service experience - Salesforce GM of C360 Applications Bill Patterson on "the most profound opportunity"

Salesforce later today will unveil a slew of product and strategy announcements at a special AI Day event in New York, with CEO Marc Benioff outlining the roadmap for the company that, as we saw last week at Connections, now regards itself as an AI firm.

More on that as the event rolls out, but there was a useful and articulate summation of Salesforce’s thinking from Bill Patterson, General Manager of C360 Applications, who took part in last week's Bank of America’s 2023 Global Technology Conference virtual event, during which he declared generative AI to be “the most profound opportunity of our lifetime” around enterprise software.

Salesforce’s starting point in the AI revolution is trust, he argued:

We've seen a lot of stumbling already in the early days of generative AI about people using public models or publicly-available consumer services in the wrong and inappropriate way, sharing information with these models that now becomes part of the model for the future, even if it's sensitive information, if you will. So first off and foremost, it is about a platform of trust for us around how we think about AI as a foundational service across our applications.

Patterson also pointed to the reality of enterprises today as being a large and diverse landscape of technology. He cited customer service as an example:

Oftentimes, the service organization has 20, 30, 40 different systems in use that manifest that customer experience or that operations like a contact center. It is not sustainable for an organization to have 30, 40 different AI strategies to try and kind of manifest a consistent experience for their customers.

Those two factors have shaped Salesforce’s AI thinking, he said:

Our approach, starting with the trusted foundation, said, ‘OK, first and foremost, we want to make sure that the data stays resident and secure and safe’. Second, we really want to help organizations leap into this AI era by not trying to have all of these different diverse of landscape of models that are just flying around inside of the organization. I want to make that more cogent, more coherent, more simple to utilize across every center of interaction the company does.

Customer impact

There is a further consideration, he went on:

The last horizon of AI really becomes, how do you think about the application of AI, specifically in how you serve, how you sell, how you market, how you digitally transact with organizations? That's really kind of the last horizon we're thinking about. How do we really think about fundamental new ways to break through the marketing domain or the service domain or the selling domain with those areas?

The focus on applying AI for the purpose of transforming customer experiences and the impact that this will have organizationally and operationally is something that is high-up on corporate agendas, he suggested, recalling a recent meeting with 130 service leaders in Chicago:

Customer service is one of those domains and disciplines that I think a lot of people are forecasting will have a large amount of labor disruption associated with the use of generative AI and generative AI practices. So I was there in a meeting, talking to customer service leaders, and the question came up, ‘OK, so does that mean that we're going to have less people working in the service center?’.

The reality is that for software companies - and for software technologies like what Salesforce offers - it means the software actually gets more valuable because we can make the users of that domain more productive, more sort of intelligent around how they drive interactions, and actually drive more scale that's there. So I think, first and foremost, while there may be less labor employed, the value of the software actually gets more a premium value associated with that.

The second horizon of that is, I think that we still cannot keep up with the demands of what consumers are generating in terms of brand expectations for sales, marketing, service interactions. So, AI and an always-on experience actually gives new ways to drive growth for companies like Salesforce, because what we're going to do is build technology that's in an always running, always on manner, even when your employees are not at work. So I think that gives another kind of horizon.

And then last, when you think about sort of kind of value creation, what companies are actually looking today to companies like Salesforce for is help to drive growth or savings. The opportunity to use the software to participate in the growth and savings of how the business revolutionizes itself, I think just becomes another growth lever for a company like Salesforce.

As to how Salesforce responds to Large Language Models [LLMs] taking on more of sales and service functions, Patterson pitched some differentiators, beginning with data itself:

Today, you can go out to a Large Language Model like ChatGPT and get a really great compelling sort of narrative written for you or a trip plan for you online. But what you don't get is a personalized response. You don't get something tailored to you as an individual or to your needs or to your kind of unique circumstances. So first and foremost, the data that sits inside of the CRM system, not only about customer profile, customer history, but also around your policies as a business, your operational policies, the way in which you communicate in your brand and your sentiment, or what you allow to happen in a regulatory environment, that sits inside the CRM core today.

So the ability to combine the CRM data structure and do it in a safe and secure way along with this generative technology, ultimately creates this incredible flywheel of innovation, but also a longevity of value for a platform like Salesforce to keep serving companies into the future with a more AI-first manner of how they interact with their customers.

My take

A taster menu for the full AI meal coming later today.