{"id":11149,"date":"2026-06-29T07:34:24","date_gmt":"2026-06-29T13:34:24","guid":{"rendered":"https:\/\/attentionmedia.io\/?p=11149"},"modified":"2026-06-29T07:34:24","modified_gmt":"2026-06-29T13:34:24","slug":"agentic-ai-is-rewriting-martech-economics-and-infrastructure","status":"publish","type":"post","link":"https:\/\/attentionmedia.io\/?p=11149","title":{"rendered":"Agentic AI is rewriting martech economics and infrastructure"},"content":{"rendered":"
\"\"<\/div>\n

Marketers embraced AI when it was priced like an all-you-can-eat buffet. Providers\u2019 shift to token-based pricing comes just as agentic workflows are becoming part of everyday marketing \u2014 and agents use many, many tokens. Martech\u2019s infrastructure needs to change if it\u2019s going to keep costs down amid growing demand.<\/p>\n

The moment AI connects to your business systems, the chatbot becomes much more powerful. Instead of answering one question at a time, it can pull customer records from your CRM, analyze campaign performance, search the web, and generate a personalized report in a single workflow. That\u2019s made possible by tool calling, which lets AI access external systems via APIs and Model Context Protocol (MCP) connections.<\/p>\n

The result is a huge productivity boost for marketers. AI can chain together multiple tools without requiring users to jump between applications. The catch is that every tool call consumes tokens. AI agents, in particular, use an incredible number of them because they pass the entire task history, their internal reasoning, and any external tool data back through the model at every step of their problem-solving loop.<\/p>\n

The token cap reality<\/h2>\n

Let\u2019s look at a real-world example of how that works.<\/p>\n

A typical daily pipeline \u2014 search 200 results, summarize them, generate five headline variations \u2014 can easily run 4,000 to 5,000 tokens or more per run. Over a 30-day month, that can reach well over 100,000 tokens, well past the free-tier limits on OpenAI, Anthropic, and similar platforms, and even enough to blow through a $20 subscription well before the month is over.<\/p>\n

\"\"<\/figure>\n

(All token estimates in this article are based on standard tokenization metrics used across the industry \u2014 the same method providers use to calculate your bill. They are rough projections, not exact measurements from a live pipeline, and actual usage will vary based on model, prompt structure, and output length.)<\/em><\/p>\n

Why Claude Cowork and similar workflows hit the wall<\/h2>\n

Unfortunately, there\u2019s no correlation between the amount of tokens used and the quality of the result. As Scott Brinker and Frans Riemersma note in the State of Martech 2026 report<\/a>, \u201cmore input does not automatically mean better output\u201d \u2014 but you are still paying for every bit of it. <\/p>\n

Claude Cowork and other tool-heavy environments make the problem visible fast. Every file read, every search, every API call adds a billable token interaction. Users who start the month with a $20 subscription often find themselves throttled by week two.<\/p>\n

The consequence is choosing between throttling your workflow or paying astounding overage fees. Neither is sustainable for a marketing team that needs to run pipelines daily.<\/p>\n

The answer is owned context, not a single provider<\/h2>\n

Fortunately, there is a solution: Keep the raw data under your control. Store it in a shared team database like PostgreSQL or Qdrant, in a cloud data warehouse like Snowflake or BigQuery, or in a folder in shared cloud storage \u2014 and use lightweight, non-LLM filtering logic to pull out relevant pieces before anything touches the model.<\/p>\n

Setting that up might involve an LLM once, the same way you might use an AI assistant to write a formula or a script. But once it is in place, it runs automatically on every batch of new data \u2014 and it does not call an LLM at all. Simple keyword scoring or vector similarity search, both orders of magnitude cheaper than an LLM call, rank the data by relevance.<\/p>\n

When a social listening pipeline pulls 500 tweets about a brand, the filtering step quietly selects the 10 most relevant ones and sends only those to the model. The token bill typically drops by 60% or more. The insight quality stays the same.<\/p>\n

Beyond the one-off agent<\/h2>\n

There are a number of tools that can do this type of filtering. The open-source Hermes Agent<\/a>, Claude Cowork, Claude Code, and Perplexity Computer all connect an LLM to external tools, allowing it to call APIs, read files, and automate workflows that can require switching between half a dozen applications. However, Hermes runs on your infrastructure and is provider-agnostic. The others are tied to the models and infrastructure of Anthropic and Perplexity.<\/p>\n

Other notable tools in the broader agent ecosystem include:<\/p>\n