Three Years In — Time for an Honest Assessment
Generative AI entered the marketing mainstream in late 2022. Three years of real-world use later, the results are in. Not the conference keynote results. Not the vendor case study results. The results that marketers who use these tools daily will recognise as true.
Some of the promises came true. Some didn't. And a few use cases nobody was talking about in 2022 have turned out to be genuinely transformative. This post separates all three.
What Generative AI for Marketing Actually Delivers (The Real Part)
First drafts at speed
The original promise — generate content faster — is real. A skilled marketer using Claude with a well-configured skill file produces first drafts in 20% of the time it takes to write from scratch. The draft still needs editing. The editing is still skilled work. But the blank-page problem — the single most productivity-destroying element of content creation — is essentially solved.
Verdict: Real, and significant. Expect 3–5x speed improvement on first drafts for marketers who learn to brief AI well.
Creative variation at scale
Generative AI's ability to produce 10 variations of an ad headline, email subject line, or social post in under a minute is genuinely valuable. Not because AI produces better variants than humans — sometimes it does, often it doesn't — but because having 10 options to evaluate and test is structurally superior to having 2. Testing velocity has increased meaningfully for teams using generative AI for creative variation.
Verdict: Real. Testing teams are the biggest beneficiaries.
Content personalisation at scale
Dynamic personalised email content, localised ad variants, and audience-specific landing page copy are now achievable for teams without enterprise budgets. Generative AI makes the production of personalised variants affordable at scales that previously required large content teams or expensive automation platforms.
Verdict: Real, and underused. Most teams using generative AI aren't yet applying it to personalisation at scale.
Strategic thinking support
This one surprised most early sceptics. Advanced models like Claude Sonnet 4 handle strategic marketing tasks — messaging frameworks, campaign briefs, positioning analysis, competitive strategy — with enough quality to serve as a genuine thinking partner rather than just a writing tool.
Verdict: Real for marketers who brief AI well. Not a replacement for senior strategic judgment, but a powerful accelerator of it.
What Generative AI for Marketing Has Not Delivered (The Hype Part)
Autonomous campaign management
The 2022 prediction: AI will run your campaigns end-to-end, optimising creative, audience, bidding, and messaging without human input. The 2026 reality: AI handles optimisation within constrained parameters very well. It does not handle the full campaign intelligence loop without experienced human oversight. Campaigns run on pure AI autopilot without strategic guardrails consistently underperform campaigns with human judgment applied at key decision points.
Verdict: Hype. AI is a campaign co-pilot, not a pilot.
Replacement of creative judgment
Generative AI produces content. It does not have taste. It cannot tell you whether the campaign concept is differentiated, whether the headline will emotionally land, or whether the strategy will work in your specific market context. The content quality ceiling for AI-only output is recognisably lower than the ceiling for AI-assisted human output. The marketers treating generative AI as a replacement for creative judgment are producing visibly mediocre work.
Verdict: Hype. AI amplifies creative judgment — it does not substitute for it.
SEO content at scale without quality cost
The prediction: generate hundreds of blog posts per month with AI and win organic traffic at scale. The reality: Google's quality systems have improved faster than AI content quality. Thin AI-generated content at scale performs worse in 2026 than it did in 2023. Teams who invested in AI content farms are now dealing with traffic corrections and index bloat. Quality beats volume, with or without AI.
Verdict: Hype as a volume play. Real as a quality accelerator for well-edited AI-assisted content.
What Nobody Predicted That Is Now Transformative
Role-specific AI personas that hold context
The emergence of skill files and role-specific AI configurations — where a model is loaded with a professional identity, brand context, and behavioral rules — has created something marketers didn't anticipate: an AI that behaves as a consistent team member rather than a stateless chatbot. Teams using Claude with role-specific skill files report that the consistency and quality improvement is more impactful than raw generation speed.
AI as a thinking partner for strategy
The use of Claude as a sounding board — to challenge assumptions, identify weak arguments, stress-test campaign logic, and surface what's been overlooked — has emerged as one of the highest-value applications in professional marketing. Not generating answers, but improving the quality of the questions and the thinking behind them.
The Honest 2026 Verdict
Generative AI for marketing is real, impactful, and growing. It's not autonomous, it's not infallible, and it's not a replacement for skilled marketing professionals. It's a force multiplier — and the multiplier is significantly larger for the professionals who invest in learning how to brief it well.
The biggest gap between marketers getting 2x value from AI and marketers getting 10x value is almost always the quality of their prompts and configurations. A Claude skill file from KissMySkills bridges that gap immediately.