Observed Signal · Jul 21, 2026 · Analysis · Source: https://martech.org/feed/ · Impact: 2/5 · Sentiment: Neutral
Stop Buying AI, Buy Capabilities
The article argues that the umbrella term “AI” is too broad for marketing and martech procurement and urges organizations to reframe AI initiatives as four distinct capabilities: generation, augmentation, insights, and orchestration. The author, Greg Kihlström, recommends retagging existing AI initiatives by capability to reduce duplicate spend, assign clear ownership, and clarify failure modes and costs. The piece cites Forrester (Jay Pattisall and Mike Proulx) to illustrate that AI adoption may follow the same normalization arc as electricity, and it contrasts ambient low-stakes AI uses with high-stakes domains where probabilistic failures have significant consequences.
Provides a practical framework for marketing and martech teams to classify AI investments into concrete capabilities, which aids procurement, governance, and cost control, but it is not a major platform announcement or regulation.
Track Forrester Signals & Market Shifts in Real-Time
Polaris7 autonomous intelligence agents track regulatory filings, primary sources, executive changes, and deal flow 24/7. Create your free Explorer workspace to monitor these entities.
Key Takeaways & Evidence Grounding
- The article proposes reframing 'AI' into four capabilities: generation, augmentation, insights, and orchestration.
- Author: Greg Kihlström, listed as Principal, The Agile Brand.
- The author cites Jay Pattisall and Mike Proulx at Forrester in support of the argument about AI’s normalization arc.
- The article recommends organizations re-tag current AI initiatives by capability to surface duplicate spend and clarify ownership.
- The webpage metadata shows a publication date of 2026-07-21.
Connected Companies & Entities
8 Entities mapped“Jay Pattisall and Mike Proulx at Forrester made the point [recently](https://www.forrester.com/blogs/ai-forces-a-redesign-of-how-marketing-a...”
“MarTech is owned by Semrush....”
“He has worked with some of the world’s top brands, including Adidas, Coca-Cola, FedEx, HP, Marriott, Nationwide, Victoria’s Secret, and Toyo...”
“He has worked with some of the world’s top brands, including Adidas, Coca-Cola, FedEx, HP, Marriott, Nationwide, Victoria’s Secret, and Toyo...”
“He has worked with some of the world’s top brands, including Adidas, Coca-Cola, FedEx, HP, Marriott, Nationwide, Victoria’s Secret, and Toyo...”
“He has worked with some of the world’s top brands, including Adidas, Coca-Cola, FedEx, HP, Marriott, Nationwide, Victoria’s Secret, and Toyo...”
“He has worked with some of the world’s top brands, including Adidas, Coca-Cola, FedEx, HP, Marriott, Nationwide, Victoria’s Secret, and Toyo...”
“He has worked with some of the world’s top brands, including Adidas, Coca-Cola, FedEx, HP, Marriott, Nationwide, Victoria’s Secret, and Toyo...”
Ontology Mapping & Concepts
Related Market Signals & Shifts
Recent verified developments and strategic activity across this market segment.
AI Reshapes Marketing Tech: Cost Cuts, Not Collapse
The article argues that AI is shifting the economics of the marketing technology stack by making coordination and workflow interfaces (the "surface" layer) far cheaper to reproduce, while leaving deeply integrated backbone systems that absorb operational liability (the "structural" layer) largely unchanged in cost. Generative and agentic AI enable fast internal prototypes for intake forms, lightweight approvals, asset browsers and dashboards, increasing substitution risk for vendors who sell coordination wrappers. The piece recommends a disciplined hybrid model: buy backbone systems that carry liability (rights enforcement, audit trails, activation integrations) and build thin, well-governed workflow surfaces where differentiation exists. It offers four tests (liability, integration complexity, internal capability, and differentiation/time horizon) to guide build-vs-buy decisions.
Marketing needs AI outcomes, not more AI pilots
A MarTech article (published 2026-06-10) argues marketing teams must shift from running many AI pilots to delivering measurable AI value tied to business outcomes. It recommends starting with high-value use cases (assessed for value and feasibility), preparing people and processes, measuring outcomes before scaling, and managing AI investments as a portfolio of three use-case types: defend (efficiency), extend (improve outcomes), and upend (new capabilities). The piece highlights often-underestimated implementation costs (data, governance, model monitoring, training, change management), emphasises building human+AI team intelligence, and suggests distinct metrics for each portfolio category to track operational, marketing/financial, and leading indicators of value.
Five Questions Before Buying an AI Tool
This MarTech contributor piece by Tonya Walker advises marketing leaders to treat AI purchases as operational commitments rather than standalone pilots. Citing Salesforce research that 75% of marketing teams have adopted AI, Walker outlines five evaluation questions covering data readiness (identity resolution, real-time sync), integration with existing martech workflows, decision ownership and human‑in‑the‑loop controls, what breaks when solutions scale, and the full operational cost (people, governance, integrations). The author warns that AI amplifies poor data and that rapid tool adoption without infrastructure creates "AI debt"—fragmented workflows, rising costs and eroded trust. Walker is a fractional CMO and founder of The Strategic Stack and Intelligence Desk; the piece frames AI adoption as a strategic decision requiring data, integration, governance and monitoring to realize value.
Track Real-Time Market Signals & Shifts
Set up custom watchlists to receive automated, evidence-grounded executive digests whenever material signals or shifts occur across your tracked landscape.