Market Thesis
What becomes valuable after generation gets cheap.
AI made generation cheap. It did not make brand output good. Value is moving to whoever owns the process — the system that turns brand knowledge, research, approvals and performance data into work a brand can actually ship.
The Shifts
Three forces.
One converging category.
Adoption is broad. Workflow scaling still lags.
44%
report AI scaling across the enterprise, up from 38% a year ago1
Nearly nine in ten organizations use AI in at least one business function.1
Large enterprises scaling AI agents rose from 27% to 40% in a year; smaller organizations stayed flat at 22%.1
61% of teams still juggle two or more CMSs to manage their brands.2
Interpretation: adoption is real. Scaled workflow is not. That gap is where new infrastructure gets built.
Buyers ask a model before they buy.
58%
of consumers use generative AI as a go-to source for product recommendations3
39% use AI for product discovery, above 50% among Gen Z. 63% of Gen Z shoppers want product recommendations from AI agents.4
Agentic search grew 200% year over year between August 2025 and May 2026.5
AI-driven retail traffic rose 693% year over year across the 2025 holiday season, and by March 2026 those visitors converted 42% better than non-AI traffic.6,7
Interpretation: the recommendation surface itself is moving. A brand illegible to models loses distribution before a human ever sees it.
The budget exists, and it is moving.
$2.9M
average annual creator-marketing spend reported by brand respondents8
Budgets grew 171% year over year; 71% of organizations increased their investment.8
93% of CMOs using generative AI report ROI, alongside 83% of marketing teams.9
CMOs now allocate 15.3% of marketing budgets to AI, yet 70% say their processes are not mature enough to scale it.10
Interpretation: nobody needs to be sold on the category. The question is which system absorbs the spend.
The Breaking Point
Why the old model breaks.
Content velocity outpaces production capacity.
More assets, more channels, shorter cycles. Manual workflows cannot keep pace.
AI tools generate assets but do not run the process.
Prompt interfaces produce output. They do not know the brand, check compliance, or learn from performance.
Analytics and execution are disconnected.
Performance lives in one system, production in another. The feedback between them is manual or nonexistent.
Every person applies AI their own way.
Each team member, consultant and freelancer brings their own prompts and their own judgment. The output drifts brand by brand, person by person.
Where Value Shifts
What gets commoditized.
What compounds.
Generic copy and image generation
Prompt-based experimentation
Undifferentiated chat interfaces
Raw first-draft output
Horizontal “AI content” tools
Structured brand knowledge
Cross-brand performance patterns
Governed workflow systems
Performance-linked feedback loops
Embedded approvals and trust layers
Multi-channel expansion from one system
Cheap generation does not remove the need for infrastructure. It increases it. General-purpose models generate assets. They do not own brand context, workflow control, approvals, or performance-linked iteration. That is where durable value sits.
The Argument
The winners will own the process, not the model.
Model quality is converging, and it is rented. Anyone can call the same frontier models you can. What is not rented is the process around them.
The tools a brand team already uses don’t talk to each other. Research sits in one tool. Writing sits in another. Approvals happen in email. Performance lands in a dashboard nothing else reads. Every step works; the system between them doesn’t exist.
That gap is the entire play. One living record per brand, read by every product that touches that brand’s output — a verticalized process, not a better prompt. The harness is the product.
Generalized models produce generic output. A verticalized process produces branded output that improves with every cycle.
Each of these owns a step. None owns the harness. The category is crowded at the step level and open at the system level — because holding brand context, workflow control, approvals and feedback at the same time requires a company structurally committed to all four, and most tools are committed to one.
Our View
Where this market is going.
Content generation commoditizes faster than content workflow.
More product discovery moves into AI-mediated environments.
Content budgets keep shifting from services into software.
The winning systems combine brand knowledge, workflow control, trust, and feedback loops.
The tools brands use will be reachable by people and by agents, from the same record.
Jinn is building the layer where that value accrues.
The real budget opportunity is not replacing creators one for one. It is converting a variable, service-heavy cost structure into a repeatable software process.
“The biggest risk in this market is not moving too early. It is moving too slowly while the process layer gets claimed.”
All market data referenced on this page.
