How AI Agents Will Reshape Your Growth Marketing Strategy

How AI agents will reshape your growth marketing strategy

AI is evolving at a breakneck pace, with Meta, Google, Microsoft, and Amazon alone spending over 50 Billion in Q2 2024 on AI. AI agents are at the forefront of this revolution, poised to redefine the way we interact with technology as well as how businesses engage with their customers. As a growth marketer, understanding the implications of this shift is no longer optional—it's essential for staying ahead of the curve.

Let’s dive into how AI agents are set to transform your customer data, unlock new possibilities for your marketing efforts, and reshape the customer journey itself.

What are AI agents?

Before diving into the impact, let’s clarify what we mean by "AI agent". It’s not just another tech buzzword. An AI agent goes beyond a simple chatbot that merely regurgitates pre-programmed responses. It’s a system that can:

• Take action
• Make decisions and adapt
• Operate across channels

How AI agents take action

An AI agent analyzes input, makes decisions, and carries out tasks based on its understanding until it can produce the desired output. The AI that powers this is typically a Large Language Model (LLM). For example, if you ask ChatGPT to calculate the average lines of code changed in a GitHub Pull Request, it will actually browse the web, extract the relevant data from GitHub by navigating to another tab, write code to calculate the average, and then provide you with the answer. This is a far cry from simple keyword-based responses or even "glorified autocomplete".

How AI agents make decisions and adapt

AI agents are not confined to a fixed script. They are inherently non-deterministic, meaning they don’t follow a fixed decision path. They can adapt their behavior based on new information and past interactions, evaluating whether they have enough data to complete a task. This learning ability makes them incredibly flexible and capable of handling complex, dynamic situations.

How AI agents operate across channels

AI agents don't just live in chat windows. They can function across various channels, like email, voice assistants, and even within software applications, tailoring their communication style to the specific context. You could for example have an AI agent be triggered based on an event in a customer journey.

To get an idea how far this can go, research projects like "Smallville" from the Generative Agents paper have AI agents power entire virtual worlds. A simulated village is populated by AI agents, each with individual personas, who plan events, interact, and go about their daily lives. Similarly, the Voyager project had an AI agent play Minecraft, write code to create new tools, evaluate those tools, and continue to evolve in its effort to complete its objectives within the game.

The shifting landscape of customer data

In general, we can think about two categories of AI agents:

• Company AI agents, built and shipped by companies to engage with their customers and essentially represent the company
• Personal AI agents, used by individual users and perform tasks on behalf of the end user, including interacting with businesses

Personal AI agents, those that represent individual users, are already impacting how we access information. Zero-click searches, where answers are provided directly to the user without needing to visit a website, are becoming increasingly common. Services like Perplexity, SearchGPT, and even Google's "AI Overview" are aggregating information from various sources and delivering concise answers, often bypassing the original website.

This has major implications for growth marketers. Gartner is predicting a 25% drop in search engine traffic by 2026 due to the advancements of Generative AI and AI Agents. With this change traditional attribution models are challenged, and we need to rethink how we measure the effectiveness of our content when a user might benefit from it without ever actually visiting our page.

Furthermore, as AI agents become more sophisticated, they will act as personalized interfaces, enabling users to interact with businesses directly through their agents. GitHub Copilot Extensions, for example, enable the AI agent to access your Stripe or DataStax account directly from within your code editor using natural language. This shift could lead to less direct traffic to websites and apps, impacting data collection and requiring new strategies for understanding customer behavior. Some companies like MultiOn are even making progress on AI agents browsing complex websites and recovering from failures in order to perform tasks on these websites like creating a reservation on OpenTable.

AI agents could also introduce new challenges, especially for new entrants to a market or those trying to drastically change an established product. Coding AI agents like GitHub Copilot or Cursor rely at least in part on the underlying training data to generate code and provide answers. As a result they will likely default to those solutions that have the most representation in the training data. That's great for the incumbent but makes entering a new space significantly harder.

The rise of resolution-focused customer journeys

One of the most exciting promises of AI agents is their potential to revolutionize the customer journey. Today, many customer journeys end up relying heavily on support teams to handle the unique and often unpredictable paths customers take. Traditional chatbots, despite having been used as the de facto solution for years, are often glorified FAQ repositories, frustrating users with their limited capabilities and generic responses.

AI agents offer a different approach—one focused on resolution. Instead of simply providing hints or directing customers to help articles, they can proactively resolve issues using the following tactics.

Accessing and analyzing relevant data

Imagine a customer messaging an airline's AI agent about lost luggage. Instead of greeting the customer with a generic response, the AI agent checks the customer profile, retrieves the baggage IDs and flight information, and then uses that data to check the status of the bags. It might even proactively offer to schedule a delivery to their hotel based on their travel itinerary. By giving AI Agents secure access to relevant customer data, the customer experience can be completely streamlined.

Reasoning and problem-solving

Even if a customer provides incomplete information, a sophisticated AI agent can use reasoning abilities to fill in the gaps. In the lost luggage example, the agent might be able to deduce the customer's location based on their flight information and a partial address.

Offering personalized solutions

Armed with the necessary data and understanding of the customer's situation, the AI agent can offer tailored solutions. It might suggest alternative travel arrangements, offer compensation for the inconvenience (in alignment with set policies), or provide real-time updates on the location of the lost luggage.

This approach shifts the focus from deflecting inquiries to proactively resolving them, transforming potentially negative experiences into positive ones. By empowering AI agents to take action, we can provide customers with a VIP experience, tailored to their individual needs, even at scale.

Over at Twilio Alpha we are exploring this future of more resolution-focused customer journeys by building Twilio AI Assistants — a platform for building and hosting customer-aware and omni-channel AI Agents. We invite you to check out AI Assistants and start building towards this future alongside us!

New opportunities for growth marketers beyond resolution-focused journeys

The evolving data landscape presents challenges, but AI agents also unlock a wealth of new possibilities for growth marketers beyond just providing a resolution-focused customer journey.

Deeper customer understanding

AI agents can process vast amounts of unstructured data—social media conversations, support tickets, user research recordings—uncovering valuable insights about your customers, their pain points, and their motivations. This empowers you to create more targeted and effective marketing campaigns. Think of an AI agent that monitors your online community for sentiment, identifies customers at risk of churn, and triggers personalized outreach campaigns to address their concerns.

Automation of repetitive tasks

AI agents can handle tasks like data enrichment, report generation, content research, and even A/B testing automation, freeing up your time to focus on strategic initiatives. At Twilio for example we built "RFP Genie", an AI system that answers RFPs from potential clients by analyzing the requirements, gathering relevant information from your internal databases, and generating relevant responses. This level of automation can significantly improve efficiency and scale your team's impact.

Hyper-personalization

AI agents can remember user preferences, past interactions, and even subtle cues from conversations to deliver hyper-personalized experiences. For example, a fashion retailer's AI agent might remember a customer’s favorite color, shoe size, and past purchases from conversations with it and retain it in your Customer Data Platform (CDP) such as Twilio Segment. Equipped with that data, you can then later use it to recommend relevant products, curate personalized shopping experiences or improve your marketing campaigns.

Embracing the future of growth marketing

The rapid advancement of AI presents both challenges and immense opportunities. While enterprise-ready AI agent solutions are still evolving, companies are already experimenting and seeing impressive results.

Now is the time for growth marketers to:

• Stay up-to-date. Engage with the AI community, attend meetups, and follow industry leaders to keep your finger on the pulse of this rapidly changing field.
• Get your data in order. High-quality data is the fuel that powers effective AI agents. Invest in data cleaning and organization to maximize the value of AI insights.
• Experiment and iterate. Start with low-risk experiments, using AI tools to automate repetitive tasks, gain a better understanding of your customer data, or even try building your first AI agent using a framework like Twilio's AI Assistant platform.

AI agents are poised to change growth marketing in profound ways. By embracing this technology, understanding its implications, and continuously experimenting, we can create more personalized, effective, and ultimately, more human-like customer experiences—even at scale.

Dominik leads Product for the Emerging Tech & Innovation organization at Twilio. His team builds next gen prototypes and iterates quickly to help craft the long term product vision and explore the impact of autonomous agents & AGI on customer engagement. Deeply passionate about the Developer Experience, he’s a JavaScript enthusiast who’s integrated it into everything from CLIs to coffee machines. Catch his tweets @dkundel and his side ventures in cocktails, food and photography.