1 step (efficient action) forward, and 2 steps (brand differentiation & creativity) back. Martech Futurist | June 26, 2026

Recent research and insights from HBR, Forrester, and MarketingProfs point at the same problem from different directions, and a product launch from Databricks adds a fourth angle. AI is everywhere in marketing now. Most organizations are aiming it at speed and cost savings, while the things that actually grow a business — distinctive brand, strong creative, customer experience, readiness for what's coming — sit flat or slip backward. Four developments are worth a CMO's attention this week.

1. Agencies adopted AI almost universally, and the creative isn't getting better.
Forrester's 2026 State of AI Inside US Marketing Agencies report puts usage at nine in ten agencies, with most of that effort aimed at internal productivity. Very little of it goes to making the creative itself stronger. What CMOs ask for tells the story. Cost efficiency pulls 71% of them. Revenue growth pulls 27%. So brands drift toward sameness at the exact moment AI is making everything else cheap and interchangeable. That's the part worth worrying about.

2. The CDP is turning into agent-driven infrastructure.
The customer data platform is starting to look like yesterday's architecture. Databricks just launched CustomerLake, built for always-on marketing that runs through AI agents and reacts in real time. If you haven't checked whether your martech stack can actually support agent-driven workflows, the ground is already shifting under you.

3. The AI's own personality is a performance variable.
This one almost nobody is tracking. HBR research finds that the way these systems talk to employees changes the quality of the work they produce and how much stress they carry. It even affects how predictable their output is. Most companies measure adoption and stop there — logins, query counts — without watching for friction or for how the tool reshapes day-to-day behavior. That gap is a governance problem, and it costs real productivity.

4. AI adoption stalls for organizational reasons, not technical ones.
When AI gets stuck in marketing operations, the technology usually isn't what's broken. MarketingProfs reports that only about one in three organizations have pushed AI past the experiment stage. The rest get held up by mundane things. There's no real strategy. Processes were never documented, data is messy, and tools have piled up faster than anyone can manage them. Layer weak change management on top, and the pilot never turns into practice.

The takeaway runs through all four stories. Adoption rates are the wrong scoreboard. What matters is whether AI is improving business results and lifting the quality of your creative, and whether your operation is actually built for agent-driven work. CMOs who own that question get to shape how this goes. Wait too long, and the CFO or CTO will frame it for you instead.

Featured Insights

The Cost of AI Productivity Is Less Creativity — Forrester Blog

Source: Forrester Blog (https://www.forrester.com/blogs/the-cost-of-ai-productivity-is-less-creativity/) | Author: Jay Pattisall | Published: June 24, 2026

Forrester's latest research on the state of AI inside US marketing agencies delivers a sobering finding: nearly 9 in 10 agencies now use generative or agentic AI, but the emphasis on efficiency over effectiveness is actively undermining brand differentiation and creative quality. Agencies prioritize enhancing internal staff productivity (81% for genAI) far above improving creative ideation quality (65%), and corporate marketers compound the problem by requesting cost efficiency from their agencies (71%) far more than marketing performance (49%) or revenue growth (27%). Forrester warns that this short-term efficiency focus is eroding companies' most distinctive asset — their brands — at the exact moment AI is commoditizing everything else. The call to action for CMOs is direct: reset AI expectations from efficiency to effectiveness, and demand that agency partners use AI to drive strategy, innovation, and growth — not just to cut costs.

The Canary in the CDP Mine: Databricks CustomerLake Is the Litmus Test for Agentic Marketing — Forrester Blog

Source: Forrester Blog (https://www.forrester.com/blogs/the-canary-in-the-cdp-mine-databricks-customerlake-is-the-litmus-test-for-agentic-marketing/) | Author: Joe Stanhope | Published: June 22-23, 2026

Forrester analyst Joe Stanhope examines Databricks' CustomerLake — a ground-up, AI-native customer data platform announced at the Data + AI Summit — and argues it represents far more than a new CDP entrant. CustomerLake is built around agentic AI workflows, always-on continuous customer engagement (replacing traditional campaign paradigms), and tight integration with enterprise data infrastructure. Stanhope frames it as a martech stress test: it will reveal whether enterprise marketers are truly ready for agent-first workflows, and whether the industry will shift from composable CDPs toward consolidated, intelligence-driven martech platforms. CMOs evaluating their data infrastructure should ask whether their organization is architecturally and strategically ready for always-on, agent-driven customer journeys — because CustomerLake signals that the market is moving there regardless.

Does Your AI Have a Personality Problem? — Harvard Business Review

Source: Harvard Business Review (https://hbr.org/2026/06/does-your-ai-have-a-personality-problem) | Authors: Aleksandra Przegalinska et al. | Published: June 24, 2026

A controlled psychophysiological study of 58 participants reveals that AI interaction style — its persona — has measurable, significant effects on employee stress, work quality, and output predictability, yet these effects are almost entirely invisible to standard satisfaction surveys. Participants working with a hostile AI persona showed 72% higher peak stress (measured via skin conductance), produced lower-quality work rated a full point lower on a 7-point scale, and spent significantly more effort managing the AI rather than using it. Critically, their self-reported satisfaction scores showed almost no difference from those working with a supportive AI. The implication for CMOs deploying AI tools across marketing teams: adoption metrics and satisfaction surveys are insufficient governance instruments. Organizations must treat AI persona as a governed design variable, measure friction (not just adoption), and interpret employee workarounds as system design failures — not behavioral problems.

Why AI Underperforms in Marketing Operations — And 5 Foundations for Success — MarketingProfs

Source: MarketingProfs (https://www.marketingprofs.com/articles/2026/55144/ai-adoption-marketing-operations-strategy) | Author: Steffen Drucks | Published: June 2026

Despite near-universal AI tool adoption, McKinsey data cited in this article confirms only roughly 1 in 3 organizations have moved AI beyond experimentation in marketing operations. The article identifies five structural gaps that explain most AI failures: (1) absence of a defined AI strategy tied to specific business outcomes; (2) undocumented, non-standardized workflows that make AI integration inefficient at scale; (3) poor data quality and lack of structured, high-precision inputs; (4) tool sprawl creating cost inefficiency and governance gaps across the martech stack; and (5) change management failures that leave teams resistant or unprepared. The prescription is an orchestration-first approach: define what AI must achieve, map the process layer, clean the data, rationalize the stack, and invest in people transformation. For marketing operations leaders, this is a practical diagnostic framework for why AI pilots stall and what to fix first.

——————————

Put the four stories side by side and a pattern shows up. The companies getting real value out of AI have done three unglamorous things first. They've decided what the technology is supposed to deliver in business terms, beyond saving time. They've fixed the plumbing underneath it — cleaner data, documented processes, fewer overlapping tools — so the AI has something solid to run on. And they pay attention to how the AI behaves once it's deployed, not just to whether adoption numbers are climbing.

That split is starting to separate the winners from everyone else. One group still treats AI as a way to spend less. The other treats it as a way to grow and stand apart. Gartner found that 84% of companies are caught in what it calls a brand doom loop: they underinvest in brand measurement, lose confidence in what the brand contributes, and attract less funding for it — which makes the next round of underinvestment easy to justify. That decay is what the AI misalignment Forrester and MarketingProfs describe eventually produces, one quarter downstream. The CMOs who can tie AI spending to brand health and to revenue, and show it in the customer experience rather than only on the cost line, are the ones who'll keep the C-suite's confidence and a real say in strategy.

And the clock is running faster than it looks, because the agent-driven shift is already underway. CustomerLake, Forrester's read on the agencies, and the HBR work on AI personality are three views of the same turn. Marketing is moving toward always-on systems that run through AI agents and sit inside the customer journey itself. Companies that skipped the foundational work — the data, the processes, the governance, the strategy — won't be able to join that shift when it actually counts. By the time they notice, the buyers and the agents will already be talking to each other.