5 key stats on agentic AI

Traditional prompt-driven AI is giving way to agentic AI, which consists of autonomous agents capable of independently performing tasks, making real-time decisions, and learning from experience. Though it’s still in its early phases, agentic AI enables marketers to enhance operational efficiency, personalize customer interactions, and drive innovation.

Here are five stats marketers need to know about this next phase of AI.

Key stat: 33% of enterprise software applications will incorporate agentic AI by 2028, up from less than 1% in 2024, according to an October 2024 report from Gartner.

• This will enable 15% of day-to-day decisions to be made autonomously.
• “Few will mourn the lost minutes spent on mind-numbing tasks that bookend the work that employees are most proud of,” said Dan Van Dyke, vice president of generative AI (genAI) at EMARKETER. “[Now] is a great place to start exploring agentic automation.”

Key stat: There will be 116.8 million US generative AI (genAI) users in 2025, according to a June 2024 EMARKETER forecast.

• By 2026, this number will reach 127.2 million, representing 36.7% of the US population.
• GenAI models continue to improve daily, enabling users to draft documents, summarize data, generate code, and handle customer service inquiries. However, these models still need human prompts to initiate each action.

Key stat: 71% of sales employees spend their time on non-selling tasks, and 66% of customer service reps spend their time on non-customer-facing tasks, according to 2024 Salesforce data.

• Agentic AI automates repetitive and complex processes, saving resources and time.
• It also lets organizations take on additional work or handle demand surges without adding staff.

Key stat: 46% of marketing executives worldwide believe genAI will significantly enhance real-time decision-making capabilities, according to July 2024 data from Coleman Parkes Research and SAS.

• AI agents can quickly gather and analyze data, uncovering trends and patterns that traditional tools might miss, helping organizations make informed decisions.
• 30% of CMOs worldwide say insight generation and 27% cite predictive analytics as top use cases for genAI budget priorities, according to data from the Boston Consulting Group and Association of National Advertisers.

Key stat: The top marketing AI use cases include automating customer interactions, generating content, analyzing performance, automating data integration, and driving best offers in real time, according to May 2024 data from Salesforce.

• AI agents can create content from existing resources or data.
• They can create and automate campaigns, personalizing them based on customer data and other important info (like daytime or region).
• Agentic AI can also be used in journey orchestration, flow orchestration, and to recommend the next best actions or offers.

Want to learn more about agentic AI and why now is the time to embrace it? Download “EMARKETER’s Field Guide to Agentic AI.”

In the news:

Advertisers increase spend on Google, Pinterest amid TikTok shutdown, Meta’s content moderation makeover

Google, YouTube, and Pinterest’s shares of US paid media traffic increased between Q4 2024 and January 20, 2025, while TikTok’s and Meta’s shares declined, according to analytics provider MikMak, which analyzed real-time commerce data from 2,000 brand customers and adjusted it for seasonality. (MikMak is an EMARKETER KPI data partner.)

Our view: Uncertainty around TikTok and Meta are spurring further diversification in the ad market. While Meta and Google will remain the largest ad platforms—as well as the biggest financial beneficiaries of a TikTok ban—advertisers are looking for alternative places to spend their dollars.

This was originally featured in the EMARKETER Daily newsletter. For more marketing insights, statistics, and trends, subscribe here.