AINative.Academy — Track 1 · AI Fluency

Your path from
curious to centaur.

A personalised curriculum built for how you actually work — not a generic module designed for no one in particular.

Role
Marketing Manager
Readiness
Curious
Track
Track 1 — Fluency
Levels
1 → 2 → 3
How this curriculum is shaped for you

You've tried AI a few times — it wasn't useless, it just didn't stick. That's not a you problem. It's a structure problem. This curriculum skips the confidence-building exercises designed for anxious skeptics and removes the scaffolding that would slow down an optimist. You start at the practical end of Level 1 and move fast to real work applications. Every exercise is built around marketing workflows: briefs, campaigns, research, copy, reporting. Nothing abstract, nothing hypothetical.

Level 1
Foundation
Meta Thinker
Estimated 3–4 weeks
Transformation → Prompts that sometimes work → A disciplined communicator who understands why outputs succeed or fail
01
1.1 · Structured Intent Specification
Brief Before You Build
You already write creative briefs. This module shows you that a good AI prompt is structurally identical to a good creative brief — goal, audience, constraints, format. Sandbox exercise: Take a brief you wrote last month. Rebuild it as an AI prompt. Run it. Compare the output to what an agency would have produced. Identify the gaps.
Sandbox
02
1.2 · Problem Decomposition
Campaigns Are Not One Prompt
A campaign launch involves a dozen distinct tasks — positioning, messaging hierarchy, copy variants, channel briefs, internal alignment docs. This module teaches you to decompose a complex marketing goal into the discrete AI-solvable sub-tasks it contains. Sandbox exercise: Take an upcoming campaign. Map every deliverable it requires. Identify which five could be AI-assisted this week.
Reflect
03
1.3 · Constraint Setting
What It Must Not Do
Generic marketing copy is the AI's default mode. This module teaches you that the most powerful line in any prompt is the constraint — don't sound like a SaaS company, never use the word "leverage", assume the reader is already a customer. Sandbox exercise: Write a product description prompt with zero constraints. Run it. Then add five carefully chosen constraints and run it again. Analyse the delta.
Evaluate
04
1.5 · Output Evaluation & Critical Reading
The Editor's Eye
The most dangerous AI output is one that sounds professional but is subtly wrong — wrong tone, false claims, missing nuance, generic positioning. This module builds the habit of reading AI output the way a good editor reads a first draft: with structured skepticism. Sandbox exercise: Run a competitor analysis prompt. Systematically evaluate the output against five quality criteria before using a single word of it.
Evaluate
05
1.4 · Assumption Surfacing
What Are You Taking for Granted?
When AI gets your audience wrong, it's usually because you didn't tell it who the audience was. This module trains the habit of making implicit context explicit before running any prompt — buyer stage, channel, brand voice, competitive positioning. Sandbox exercise: Before writing your next campaign email, list every assumption you'd normally leave unstated. Build all of them into the prompt. Compare the output quality.
Sandbox
06
1.6 · Iterative Refinement Thinking
First Draft Is Not Final Draft
The professionals who get the most from AI treat every output as a conversation, not a vending machine result. This module establishes the refinement loop as the default working pattern — what to ask for in the follow-up, how to give feedback that improves the next output. Sandbox exercise: Write a LinkedIn post in three turns. Document what you asked for in each turn and why the final version is better than the first.
Sandbox
07
1.7 · Perspective & Role Framing
Who Is in the Room?
A CMO reviewing copy has different concerns than a performance marketer or a first-time buyer. This module teaches you to invoke perspectives deliberately — not just "write marketing copy" but "review this as a skeptical CFO who needs to justify the budget." Sandbox exercise: Run the same campaign proposal prompt from three different stakeholder perspectives. Use the results to anticipate objections in your next internal presentation.
Build
08
1.9 · Reusable Asset Thinking
Build Once, Run Repeatedly
The difference between someone who gets occasional AI value and someone who gets systematic value is whether they save and reuse. This module teaches you to recognise when a prompt is worth templating. Sandbox exercise: Identify three prompts you've already found useful. Abstract them into templates with named placeholders — [CAMPAIGN_GOAL], [AUDIENCE_SEGMENT], [CHANNEL]. Save them as your first prompt library.
Build
Level 1 Mastery Gate

Before advancing, you should be able to: take any marketing task you're given and specify it as a structured, constrained, context-rich prompt without prompting yourself to do so — it should be the natural first step. You should also be able to read any AI output and tell someone why it's good, why it falls short, and what you'd ask for next. This gate is assessed in a sandbox evaluation exercise, not a quiz.

Level 2
Daily Execution
Operator
Estimated 4–5 weeks
Transformation → AI is useful sometimes → AI is embedded in how you work every day, with measurable time savings you can name
09
2.1 · Workflow Identification & Mapping
Where Is Your Time Actually Going?
Most marketing managers have five to eight repeating task types that consume 70% of their time — writing copy, briefing agencies, synthesising research, building reports, drafting internal comms. This module maps your specific task landscape and ranks it by AI leverage potential. Sandbox exercise: Track your last five working days. Identify your three highest-time, highest-repetition tasks. Score each for AI leverage. These become the focus of the rest of Level 2.
Reflect
10
2.3 · Context Provision & Briefing
Give It the Room It Needs to Work
A marketing AI that doesn't know your brand voice, your ICP, your current positioning, and your competitive context will produce generic work. This module teaches you to build compact, reusable context blocks — a compressed brand brief — that you include at the start of any serious marketing task. Sandbox exercise: Write a 300-word brand context block. Run the same campaign copy prompt with and without it. Measure the quality gap.
Build
11
2.9 · Editing & Elevation of AI Drafts
From Serviceable to Signature
AI produces professional-grade drafts. Your job is to make them yours — adding the specific voice, the unexpected insight, the reference that makes the work memorable. This is not correcting mistakes. It is applying editorial judgment. Sandbox exercise: Take an AI-generated blog post intro. In 10 minutes, make it read like you wrote it from scratch. Document what you changed and why — that list is your editorial instinct made explicit.
Evaluate
12
2.7 · Research Synthesis & Source Judgment
The Research Co-Pilot
Market research, competitor analysis, trend synthesis — tasks that used to take half a day now take 40 minutes if you know how to direct the process and validate what comes back. This module teaches structured research prompting and the discipline of verifying before using. Sandbox exercise: Run a competitor positioning analysis for a real brand in your space. Build a verification checklist. Identify three claims in the output that need human verification before you'd use them in a presentation.
Evaluate
13
2.5 · Reusable Asset Construction
The Marketing Manager's Prompt Stack
By now you have enough real prompt experience to systematise it. This module guides you through building a personal prompt stack — a set of templates for your most common marketing tasks: campaign briefs, copy variants, research synopses, status updates, channel adaptations. Sandbox exercise: Build and test five templates from your real workflow. Each one should be runnable by a colleague with zero AI experience by filling in the named placeholders.
Build
14
2.6 · Tone & Voice Calibration
Making It Sound Like the Brand
Generic AI copy has a recognisable voice — fluent, professional, and completely indistinct. This module teaches you to encode your brand voice in terms that a model can operationalise — not "warm and friendly" (useless) but specific vocabulary choices, sentence length norms, things the brand never says, and reference examples. Sandbox exercise: Build a brand voice specification document. Run three different content types through it. Evaluate consistency. Refine the spec based on what breaks.
Build
15
2.4 · Output Validation Against Real Standards
Would You Actually Send This?
The final discipline of the Operator level: applying your professional standards to AI output before it goes anywhere. Not "is this grammatically correct" but "would I put my name on this, send it to the client, and be proud of it?" Sandbox exercise: Build a personal quality checklist for the three content types you produce most. Run three AI-generated pieces through it. The checklist becomes part of your prompt stack.
Evaluate
16
2.8 · Time & Leverage Measurement
Prove It to Yourself
This module closes Level 2 by making the value concrete and personal. You measure the actual time savings on four real tasks — before and after AI involvement — and document the delta. This is your evidence base for the next performance conversation and for unlocking Level 3. Sandbox exercise: Run four real work tasks with AI this week. Log time before and after. Write three sentences on each: what worked, what didn't, what you'd do differently next time.
Reflect
Level 2 Mastery Gate

Before advancing, you must be able to describe at least four real work tasks you now do differently because of AI, with a named time saving on each. You should have a working prompt stack of at least five templates you've tested and refined against real outputs. The mastery gate is a documented workflow audit — not a quiz, not a simulation.

Level 3
Human + AI Collaboration
Centaur
Estimated 4–5 weeks
Transformation → Using AI for isolated tasks → Designing end-to-end marketing workflows where you've consciously decided what AI does and what you do
17
3.1 · Workflow Architecture & Sequencing
The Campaign Machine
A campaign is not a collection of isolated tasks — it's a sequence with dependencies. This module teaches you to design the full flow: brief → research → positioning → messaging → copy variants → channel adaptation → approval → reporting. Each step is assigned: AI-led, human-led, or collaborative. Sandbox exercise: Map your most recent campaign as a sequenced workflow. Annotate every step with its AI involvement level. Identify the two bottlenecks that AI could remove without requiring human oversight.
Build
18
3.2 · Human-in-the-Loop Design
Where You Must Stay in the Room
Not all marketing decisions should be delegated to AI. Brand-defining copy, sensitive messaging, regulated claims, client-facing communications — these need a human in the loop. This module teaches you to design those checkpoints explicitly. Sandbox exercise: Take a workflow you've built. Mark every point where a human must review before proceeding. Write the review criteria for each checkpoint. These become your quality gates.
Reflect
19
3.3 · Quality Control System Design
Building the Review Layer In
A quality checklist you apply ad hoc is better than nothing. A quality review layer built into the workflow itself — where AI checks its own output against your criteria before you even see it — is dramatically better. This module teaches you to design self-checking workflows. Sandbox exercise: Build a two-step workflow: first prompt generates the copy; second prompt evaluates it against your brand criteria and flags issues. Run it on three pieces of copy. Compare this to your current review process.
Build
20
3.5 · Synthesis of Conflicting Information
When the Research Disagrees With Itself
Market research, audience surveys, channel performance data, and brand tracking rarely agree. This module teaches you to use AI as a synthesis partner — surfacing the conflicts, mapping the different interpretations, and providing the analytical scaffolding for a human judgment call. Sandbox exercise: Feed three conflicting data sources into a structured synthesis prompt. Ask the AI to map the conflicts and suggest three possible interpretations. Use the output as the basis for a positioning decision.
Evaluate
21
3.7 · Scenario Simulation & Stress Testing
Test the Campaign Before It Goes Live
A campaign that hasn't been stress-tested is a campaign with known unknowns. This module teaches you to use AI to simulate audience reactions, identify messaging risks, and surface objections before launch. Sandbox exercise: Run a pre-mortem on an upcoming campaign. Ask the AI to simulate the response of three different audience segments — including one hostile segment. Use the output to refine your messaging before launch.
Sandbox
22
3.6 · Feedback Loop Construction
The Campaign Learns From Itself
Most marketing workflows end at execution. Centaur-level workflows close the loop — performance data feeds back into the next iteration of the brief, the messaging, the targeting. This module teaches you to design that feedback cycle. Sandbox exercise: Design a post-campaign review workflow where performance data becomes the briefing input for the next campaign. Run it on a real completed campaign. Identify three things you would have done differently.
Build
23
3.9 · Risk Identification in AI-Assisted Decisions
What Could Go Wrong, and Who's Responsible?
AI-assisted marketing has specific failure modes: hallucinated statistics, tone-deaf copy, legally ambiguous claims, culturally insensitive references. This module teaches you to audit AI outputs for these risks systematically — not just once, but as a built-in step. Sandbox exercise: Build a risk audit checklist specific to your category and channels. Run it on five recent AI-generated pieces. Identify the risk category that appears most often. Build a constraint into your standard prompts to address it.
Evaluate
24
3.10 · Collaborative Intelligence Design
The Centaur Manifesto
The final module of Track 1 asks you to look at your entire marketing role and redesign it — explicitly — as a human-AI collaboration. What do you do that AI cannot? Where does your judgment, your relationships, your taste, and your professional instinct create value that automation cannot replicate? Sandbox exercise: Write a one-page "operating model" for your role as a centaur. List what AI handles, what you handle, and where you collaborate. This becomes your professional statement of AI fluency.
Reflect
Level 3 Mastery Gate — Track 1 Completion

You have completed Track 1 when you can describe one specific marketing workflow that used to take 3–4 hours that now takes under 45 minutes — and explain exactly how you designed the AI involvement, where you stay in the loop, and how the quality is maintained. That is the product working. That is what Track 1 graduates can do.

What you'll leave Track 1 with

Concrete, demonstrable capabilities — not a certificate, not a completion badge.

✦
A working prompt stack
10+ tested, reusable templates for your most common marketing tasks — briefing, copy, research, reporting, internal comms.
✦
A brand voice specification
A document that makes your brand's voice operable — usable by you, your team, and AI to produce consistently on-brand work.
✦
At least one redesigned workflow
A campaign or recurring task that is now materially faster because you've deliberately designed the human-AI division of labour.
✦
A centaur operating model
A clear personal statement of what you do, what AI does, and where you collaborate — the foundation for everything in Track 2.
The Real Metric

"A campaign brief that used to take me half a day — research, positioning, messaging framework, copy direction — now takes 90 minutes. I still own every decision. I just don't do the manual scaffolding anymore."

AINative.Academy — Personalised Curriculum · Marketing Manager · Curious Track 1 · April 2026