First Look: Salesforce Marketing Agent With Agentforce

Salesforce’s flagship conference, Dreamforce, is taking place this week (September 17th-19th, 2024). Dubbed the “World Premiere of Agentforce”, there’s no surprise that this will take center stage this year. 

In basic terms, Agentforce allows organizations to build and manage autonomous agents for tasks across various business departments – in fact, there’s an “Agentforce in every app”. There’s the potential to build Campaign Agents, which is exactly what we’ll be taking a first look at right now. 

Agentforce for Marketing

Agents are your trusted sidekick to automate tasks and uncover insights that would otherwise be burdensome for marketers – analyzing metrics, generating campaign plans, and continually suggesting optimizations for your campaigns. 

• Start by identifying your marketing goals and feeding this to the agent. For example, I want to drive more up-sell opportunities for the sales team. This is the baseline that everything else will be developed from.
• The agent develops a campaign brief. 
• The agent will create a campaign using the goals you defined, the brief you approved, the data in Data Cloud, and your brand guidelines. This includes identifying and creating the target audience segments and also generating the first draft of our email.
• Work with the agent to refine the campaign at any stage, for example, select different content or recommend additional subject lines and body copy. 
• The agent then constantly monitors how campaigns are performing and takes action to optimize where necessary.

The example in the video shows the agent taking care of the follow-up messaging for customers that engage by creating a sequence in flow (Campaign Flow) with multiple paths that will send contacts down a path according to how much or how little they’ve interacted with the campaign. These paths tap into Einstein Send Time Optimization, which determines when emails should be sent based on past engagement data from your database. The idea behind it is that by optimizing the time the email is sent, your audience is most likely to open and click.

Salesforce’s overarching aim when they conceptualized Agentforce is that the AI agent and human should coexist. While agents are autonomous and extremely intelligent (thanks to the Atlas reasoning engine that underpins all agents), the human sets the baseline that the agent then works on. As you saw, the human has the opportunity to interrupt the agent, for example, to refine the email body copy. 

Final Thoughts

While some may think this is gimmicky, the initial reaction to Agentforce has been promising. I believe that agents will be a huge gain for marketers, producing more insightful results than previous co-pilot iterations. Personally, I would love to play around with this and see how imaginative it can be based on a simple starting command.