Chapter 16What Is the Best Way to Come Up with Ideas for Agents?
Marc Benioff leads through narrative. He emphasized at the secret kickoff that all the salespeople and solution engineers – all the product people and marketers and everyone else – should be scouring the globe for customer stories, that is, case studies of companies using Agentforce.
Public examples are better but harder to secure. (Many companies don't share what they're doing, for competitive reasons.) Cash bonuses were announced on the spot for good stories. Compelling examples were the news of the day.
Stories started coming in almost immediately. Most early adopters looked at customer service. This was expected. Service is an area where many companies feel understaffed and overwhelmed, and agents seem to like to do it.
Stepping back a moment, we can see customer service has a number of features that make it amenable to the agentic treatment.i It requires the following:
- Good communication, which agents can do via voice and text
- Rapid data retrieval, often from unstructured sources, another agentic asset
- Repeatedly addressing the same topics and questions
The most obvious initial use for Agentforce was as a Q&A agent, answering customer questions. It's obvious because it engages the AI agent's most salient superpowers:
- A conversational setting
- Need to search unstructured data (e.g., knowledge base)
- 24/7 work ethic and infinite patience
One of the first agents to be introduced in this space was from Salesforce ...
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.
Read now
Unlock full access