Salesforce lost over $200 bln in market capitalization after only 34% of its customers adopted the Agentforce autonomous AI platform. Analysts at KeyBanc Capital Markets and Bernstein simultaneously downgraded the company, calling the product "not ready" for mass adoption. The reason for the failure is not a lack of interest in the technology itself, but a critical problem with data quality and organizational readiness for autonomous AI in marketing and sales.
Why companies aren't ready for agentic AI in marketing
Salesforce launched Agentforce in 2024 as a platform for building autonomous AI agents designed to automate customer service, sales, and marketing tasks. CEO Marc Benioff called agentic artificial intelligence the next evolution of enterprise software. Reality told a different story: a KeyBanc study showed that only 23,000 of the company's 150,000 customers are using the technology.
Analysts identified two key reasons for slow adoption. The first is data readiness. Autonomous AI agents require clean, structured, and interconnected data to make decisions, but most corporations operate with fragmented CRM records, siloed systems, and contradictory customer information. Companies spend as much time preparing and organizing data as they do using the tool itself.
The second reason is product maturity. According to surveys of Salesforce partners and customers, most Agentforce implementations remain at the pilot stage rather than full-scale deployment. A CIO survey conducted by KeyBanc showed that more organizations plan to cut Salesforce spending next year than increase it.
"Customer data isn't organized for serious AI work, and Agentforce as a product simply doesn't measure up," KeyBanc analysts led by Jackson Ader concluded.
Salesforce's response and market reaction to agentic AI failure
Salesforce shares fell more than 50% from their December 2024 peak, with market capitalization shrinking by $200 bln. Investors doubt that Agentforce will become a new growth driver for the company. Benioff publicly called the KeyBanc report inaccurate, citing internal statistics showing that Agentforce is the fastest-growing product in Salesforce's history.
Not all analysts share KeyBanc's pessimism. Andreessen Horowitz reported that companies actively investing in artificial intelligence increased their median Salesforce spending by 3% over three months. Guggenheim and Monness, Crespi, Hardt upgraded stock ratings, pointing to growth potential. Salesforce is investing in solving the problems: the company added technology for automatically extracting customer data from external sources and acquired Informatica to improve data integration and management before deploying AI agents.
What this means for the Russian market and brands
Salesforce's experience shows that agentic AI technology exists, but most companies aren't ready to use it due to data chaos. For Russian brands, this is a lesson in priorities. Before implementing autonomous AI in marketing, you need to get your CRM in order, unify customer data sources, and establish processes for keeping them updated. Without this foundation, any AI will be working with incomplete or contradictory information.
The data quality problem is equally acute in influencer marketing. Brands often select bloggers based on scattered metrics from different platforms, without a unified system for assessing reach, engagement, and cost per contact. AI can help here—but only if data from previous integrations, CPM rates, conversions, and media plans are structured and available for analysis. The ETC team builds blogger selection and media buying on a unified database with campaign history, which allows forecasting KPIs for new placements more accurately than manual work with fragmented spreadsheets.
Frequently asked questions
What is agentic AI in marketing
Agentic artificial intelligence consists of autonomous software agents that independently make decisions and perform marketing tasks without human intervention: processing customer requests, segmenting audiences, launching ad campaigns. Unlike conventional AI, which provides recommendations, agentic AI acts on its own, but requires high-quality structured data to function correctly.
Why do companies adopt AI agents slowly
The main reason is data readiness: most companies have fragmented customer information scattered across different systems and containing contradictions. AI agents can't work effectively with such data. The second reason is the immaturity of the products themselves: many solutions are still in pilot stages and not ready for large-scale deployment.
How to prepare for implementing AI in marketing
First, get your data in order: consolidate customer information into a single CRM, standardize formats, set up automatic updates and quality checks. Then structure your marketing campaign history with performance metrics—this is the foundation for training AI. Only after this does it make sense to select and implement autonomous AI tools.
In brief
- Only 34% of Salesforce customers adopted the Agentforce agentic AI platform, causing the company to lose $200 bln in market capitalization.
- The main reason for slow adoption is poor data quality and fragmentation at corporate customers, not lack of interest in the technology.
- KeyBanc and Bernstein analysts simultaneously downgraded Salesforce, calling Agentforce "not ready" for mass adoption.
- Of 150,000 customers, only 23,000 are using the technology, with most deployments remaining at the pilot stage.
- For Russian brands, the Salesforce lesson is clear: first organize your data (CRM, campaign history, metrics), then deploy AI.
- In influencer marketing, AI is effective only with a structured database of past integrations, reach metrics, and placement costs.
Want to see where the market is heading before your competitors do? The ETC team builds a media strategy and media plan for your niche — with reach forecasts and KPIs fixed in the contract.
