Stop Using AI Only for Output: Claim the Judgment Layer

TL;DR: Most operators use AI for drafts and summaries — execution-layer work that’s already being automated. A 2025 peer-reviewed study of 205 real ChatGPT use cases found four higher-value modes — Deciding, Ideating, Critiquing, and Talking — that almost no team has built deliberate workflows around. That gap is where strategic leverage actually lives.

The Mode Problem No One Named Until Now

The tools are paid for. The licenses are running. Your team is generating content faster than before. And yet the return on AI investment inside most marketing operations is nowhere near what the spend implies it should be. That’s not a tool problem. It’s a mode problem.

Tim Gorichanaz at Drexel University analyzed 205 real-world ChatGPT use cases for a peer-reviewed study presented at the 2025 ASIS&T Annual Meeting. He identified six distinct modes in which people actually use AI: Writing, Deciding, Identifying, Ideating, Talking, and Critiquing. The data skews Anglophone and draws from Reddit, which limits perfect generalization — but the taxonomy maps onto how most performance marketing teams are working with uncomfortable accuracy.

Two modes dominate. Four are being left on the table. And the four being ignored are the ones that determine whether AI makes your operation more strategically valuable or just faster at producing the same volume of mediocre output.

If your team is spending $10K or more monthly on performance channels, the ceiling on what AI gives you is set by which modes you’re actually using — not by which tools you’ve licensed.

The Two Default Modes That Don’t Unlock Leverage

Writing accounted for 47% of observed use cases in Gorichanaz’s data — drafting, editing, summarizing, translating. McKinsey’s 2025 State of AI survey confirms this at enterprise scale: 63% of organizations using generative AI apply it primarily to produce text. Identifying — explaining a topic, answering a factual question, synthesizing a document — made up another 10%.

These two modes are where almost every team has built its entire AI workflow. Research a topic, get a synthesis. Need a first draft, generate one. That’s real productivity. It is also not the return the investment is capable of producing.

If your AI practice begins and ends at Writing and Identifying, you’re using a sophisticated tool to do work that was already being automated — just faster and at higher volume. McKinsey’s data makes the organizational cost of this explicit: 88% of companies use AI, but only 6% qualify as high performers generating meaningful enterprise-wide impact. High performers are 3.6 times more likely to have fundamentally reworked their workflows rather than simply deployed tools into existing ones.

Faster output from an unreconstructed workflow is not the same thing as better decisions from a restructured one. That distinction matters in every vertical where customer acquisition costs are high and decision quality determines margin.

Deciding: The Mode Your Senior People Are Still Doing Alone

Deciding-mode questions are everywhere in a normal operator week. Which acquisition channels have enough signal to scale right now? Is a brand’s lead quality problem a targeting problem or a landing page problem? How do you allocate budget across SEO and paid when both need attention and the budget doesn’t stretch to cover both fully?

Most senior practitioners solve these questions with experience and intuition. That’s not a failure — both are genuinely valuable. But AI used deliberately in Deciding mode adds something experience alone cannot provide: a structured pressure-test of the assumptions underneath the decision, applied before the decision hardens into a briefing or a budget line.

This requires giving the AI the actual context — competitive landscape, current performance data, historical benchmarks, strategic constraints — and then treating what comes back as a real input rather than a summary to skim. It requires a workflow that doesn’t yet exist in most teams, not because anyone blocked it, but because the pressure to show immediate output has consistently crowded out the space to build it.

For operators running paid media programs at scale, Deciding-mode sessions before each budget cycle — not after the data comes in — is where AI starts earning its keep at the strategic layer rather than the production layer.

Ideating and Critiquing: Where Competitive Gaps Get Found First

Ideating — at only 9% of observed use cases — has a specific application most performance marketing operators are not using and should be: mapping the positioning and content gaps the brand hasn’t recognized yet.

What angles of topical authority has the brand failed to establish that AI retrieval systems are currently filling from competitor sources? What community signals — forum discussions, aggregated reviews, third-party commentary — are shaping how large language models represent the brand in response to category queries? What framings of the brand exist in model training data that the brand’s own content has never addressed or countered?

These are Ideating-mode questions. Not “give me five content ideas” prompts with list outputs. An actual iterative exploration with deliberate constraints and real willingness to follow the output somewhere the team hasn’t already been. The first takes two minutes. The second takes twenty, requires a different posture toward the tool, and produces something that can’t be replicated by anyone who didn’t do it. For operators investing in precision targeting or iGaming acquisition, that asymmetry is where competitive advantage compounds.

Critiquing, at only 6% of observed use cases, is the mode with the most direct application to daily practice and the most organizational resistance. It requires using AI to find problems in work the team has already invested in. Running your own campaigns or content strategy through a structured AI Critiquing session — before the next planning cycle, not after performance drops — is exactly the kind of proactive work that separates teams operating at the judgment layer from teams operating at the production layer.

For teams managing legal marketing campaigns or forex lead generation, where compliance constraints limit what you can say and where you say it, Critiquing mode catches weak claims, sourcing gaps, and assumed premises before a client audit or a platform disapproval catches them first.

The reason Critiquing is underused isn’t a governance problem. Organizations have trained themselves to use AI to produce output, not to interrogate it. Reversing that habit is a choice, and it’s one of the more consequential choices available to a senior operator right now.

What This Means for High-CAC Vertical Operators

In Forex, iGaming, crypto, and legal — verticals where a single qualified lead can be worth hundreds or thousands of dollars — the cost of operating at the execution layer exclusively is not abstract. Every budget decision made on intuition alone rather than on AI-augmented pressure-testing is a decision where assumptions are carried forward unchecked. Every content strategy that gets reviewed internally without a Critiquing-mode session is a strategy that ships with the same blind spots it started with.

The Talking mode in Gorichanaz’s taxonomy covers AI as a conversation partner. For operators, the most valuable version is rehearsal. The client call where you have to explain why CAC spiked 40% in Q2. The internal briefing where you have to make the case for a new acquisition channel to a leadership team that wants a single ROI number covering everything. Role-playing the pushback, asking the AI to argue the other side, running through the version of the conversation that goes wrong — this doesn’t produce an artifact and won’t show up in a utilization report. EY’s 2025 Work Reimagined Survey of 15,000 employees found that 88% use AI at work, but only 5% use it in ways that fundamentally transform how they operate. Talking mode is one of the most direct paths to being in that 5%.

If you’re running a crypto acquisition program or scaling CDL driver recruitment and your AI workflow stops at content generation, you’re leaving the highest-value modes completely unclaimed. A structured marketing audit that includes how your team is actually using AI — not just what tools are licensed — is the fastest way to identify where the judgment layer gaps are costing you.

The tools are not the constraint. The mode is. And the mode is a choice.

Originally reported by Search Engine Journal, May 2026.