M5.L2 · Content Marketing
Your unfair advantage: research, experience, opinion
What you'll be able to doGenerate three differentiated content ideas — one per moat where possible — and evaluate each with the "only we can write this" test.
Learn
From inventory to ideas
Yesterday's inventory is potential energy. Ideas are the conversion of it. This lesson turns "we have a product database" into a working headline a national content farm cannot write and an AI assistant will happily cite — and it gives you the gate every idea in this Program must pass before it earns a place in your plan.
You leave with three ideas. Your 30-day plan in Lesson 5.4 is anchored on them, so make them ideas you can actually deliver.
Moat one: primary research, on a budget of zero
Original numbers are the strongest content there is, and you do not need a research department. You need to mine what you already hold: sales records, support tickets, job notes, reviews, returns — or run a ten-question survey with a free form tool, as you did in Module 2. Original numbers are citation magnets: they are what journalists link, what rivals grudgingly reference, and what AI assistants quote, because a number that exists nowhere else can only be got from you. That fact does heavy lifting again in Module 6.
Watch Storkflow. Lena's product database holds anonymised timing data on 2,100 client onboardings completed through the product in the past year — aggregated, with no firm or client identifiable, which is the sentence her GDPR review needs. Inside it, a story nobody in her industry has published:
The median onboarding runs 11 days — and 8 of those days are spent waiting for client documents. The slowest single item is proof of identity (median 6 days from request to receipt). The engagement letter — the step firms agonise over — takes a median of 1 day.
Working headline: "Where client intake actually stalls: data from 2,100 onboardings." The bottleneck is not where her buyers think it is, and she can prove it. That is information gain in its purest form.
Moat two: experience — a record, not an opinion
"We did X; here is what happened." Before-and-after. The post-mortem of a failure. Experience content is unfakeable because it is a record of events, not a viewpoint — and it is the first E in E-E-A-T, the trust framework Module 6 examines. A machine can imitate the format; it cannot have been there.
Storkflow's second idea comes from its support team, which has hand-migrated roughly 40 firms off email-and-Excel intake and kept notes. Working headline: "We watched 40 firms leave email and Excel: the five mistakes every practice makes in week one." Importing every legacy client on day one. Rebuilding the old chaos as checklists. Forgetting to tell clients the process changed. The list is specific because someone was in the room — and "someone was in the room" is precisely what cannot be generated.
Moat three: opinion — with receipts
A contrarian position is only content if evidence rides with it. Without receipts, an opinion is a tweet; with them, it is a position paper.
Storkflow's third idea: "The intake module in your practice suite is switched off for a reason." That is a shot at the incumbents — the pack's canonical alternative is "the half-built client-intake module inside their practice-management software… usually left switched off". The receipts: Lena's Module 2 desk research, where forum buyers write "chasing documents" and "client intake" while suite vendors write "onboarding automation"; competitor reviews showing the module unused; and her own stall data showing the bottleneck is chasing, which a bolted-on checklist doesn't touch. Opinion, evidence, verdict. Someone could argue back. Good — content people can argue with is content people read.
The gate: "only we can write this"
Every idea now faces one question: could a competent stranger with a chatbot and a search engine produce this piece? If yes, it fails. No appeal.
Watch a fourth Storkflow idea die at the gate. "The ultimate guide to client onboarding." A stranger writes it in an hour. A thousand versions already exist and a chatbot will produce the thousand-and-first free. Nothing in it requires Storkflow to exist — no number from the database, no scar from a migration, no verdict anyone would dispute. It is not a bad topic; it is a commodity, and Lesson 5.1 told you what commodities are worth.
Be strict here, because this gate is also the module project's floor: a plan built on ideas that fail it cannot pass, by rubric design.
An idea is not a format
Keep these separate. "Where client intake actually stalls" is an idea. Blog post, four-part LinkedIn series, chart graphic, 60-second video, newsletter issue — those are formats, and one idea can wear all of them. Decide the idea first. Lesson 5.3 decides the formats, when you choose your channel and repurposing map. Students who decide format first end up with "we should do a podcast" — a format in search of an idea, which is the wrong way round.
Every idea carries a funnel job
Tag each idea with the work it does. Attract — reaches strangers who've never heard of you (the stall data: quotable, searchable, surprising). Build trust — convinces people already watching (the migration mistakes: proof you've done this before). Convert — moves someone to act (the switched-off module: aimed at buyers currently tolerating their suite). A healthy trio covers at least two jobs; three ideas doing the same job is a plan with a limp.
And carries an honest price tag
Last discipline. Estimate the effort per idea: an evening, a weekend, or a fortnight of small sessions. The stall-data piece is a weekend — export, analyse, chart, write. A migration-mistakes post is an evening, because the notes exist. Be honest now, because Lesson 5.4 schedules these ideas into real calendar weeks, and a weekend piece pencilled in weekly is how content plans die in week three.
Now write your three.
Do
Exercise 5.2.1 — Three ideas that pass the test
Create three content ideas — one per moat if you can (research, experience, opinion). For each, name the specific asset from your 5.1 inventory it draws on, and state plainly why a stranger with AI couldn't write it. These three ideas will anchor your 30-day plan.
Write these down — in your plan document, or on the worksheet at the end of this lesson. Answer all six for each of the three ideas, with your 5.1 inventory open beside you.
| What to write, per idea | Guidance |
|---|---|
| Working headline | 5–15 words, written as a headline, not a topic — "Where client intake actually stalls: data from 2,100 onboardings" |
| Which moat | Primary research · experience · opinion with evidence. If all three ideas share one moat, that's worth a second look — though a business with only one real moat shouldn't fake a second |
| The source asset | Up to 25 words; must reference something from your 5.1 inventory |
| Why a stranger with AI can't write it | 20–50 words answering the "only we can write this" test. Generic claims ("we know our customers") don't pass |
| The funnel job | Attract · build trust · convert |
| The effort estimate | An evening · a weekend · a fortnight of small sessions. Honesty here prevents plan fantasy in Lesson 5.4 |
When all three are written, check the coverage: if they share one funnel job or one moat, the trio has a limp.
Where this goes: section 6.2 — Anchor content ideas — of your Marketing Plan. The three titles are the candidates for your hero asset in Lesson 5.3 and must each appear in your 5.4 plan.
Check
Rubric
Mark your own work against these criteria, with your 5.1 inventory in front of you.
| Criterion | 8–10 | 5–7 | 1–4 |
|---|---|---|---|
| Source specificity | Each idea names the exact asset it draws on, and that asset is in the inventory | Assets named but loosely ("our data") | Ideas float free of the inventory |
| Genuine differentiation | A stranger with AI demonstrably couldn't produce any of the three | One idea would survive the test; two wouldn't | Three commodity topics with your logo on |
| Funnel coverage | The trio covers at least two jobs, deliberately | Two jobs by accident | Three ideas, one job |
| Effort plausibility | Price tags match the assets (existing notes = an evening; fresh analysis = a weekend) | Estimates present but optimistic | Everything is "an evening", including original research |
Pass: 5+ on every criterion. Distinction: 8+ on all four.
Quiz — four questions. Pick an answer to see whether you were right.
1. Which Herzog Physio idea passes the "only we can write this" test?
- a) "How does physiotherapy work?"
- b) "10 signs you need physiotherapy"
- c) "How long recovery really takes: what 2,000 patient records from our Graz clinic show — including the cost of waiting" ✔
- d) "Why choose a local physiotherapist?"
Why: (c) can only come from Herzog's own records — no chatbot and no competitor holds them. The others are already written ten thousand times, and the next ten thousand cost nothing.
2. "The intake module in your practice suite is switched off for a reason" is only defensible content if…
- a) It's published on LinkedIn, where opinions perform
- b) It ships with receipts — forum and review evidence, plus data showing where the real bottleneck is ✔
- c) The founder writes it personally
- d) It's longer than 2,000 words
Why: opinion without evidence is a tweet. The receipts — customer language from desk research and the stall data — are what turn a provocation into a position someone has to answer.
3. Lena's stall-data idea could become a blog post, a LinkedIn series, a chart or a video. What does this lesson say about that decision?
- a) Pick the format first, then find an idea to fit it
- b) Produce all formats simultaneously at launch
- c) Decide the idea first; formats are decided in Lesson 5.3, when the channel and repurposing map are chosen ✔
- d) Formats don't matter if the idea is strong
Why: an idea is channel-independent raw material; a format is how one channel wears it. Deciding format first produces "we should do a podcast" — a format in search of an idea.
4. Why does every idea carry an effort estimate?
- a) To calculate the content budget for accounting
- b) Because longer content always performs better
- c) Because Lesson 5.4 schedules these ideas into real weeks — a weekend piece pencilled in weekly is how plans die in week three ✔
- d) To decide which ideas to delete
Why: the estimate is a promise to your future calendar. Plans fail on effort honesty far more often than on idea quality.
Advance
Three ideas, each anchored to an asset only you hold, each with a job and an honest price tag. That's the entire creative burden of the next month decided in advance — what remains is choosing where they live and when they ship.
Next: M5.L3 — One channel, done properly. You have perhaps a few hours a week. The next lesson makes sure they all land in one place your customers actually are.
Mark your own work
| Good | Not yet | |
|---|---|---|
| Headlines, not topics | Each title could be published as written | "Something about our data" |
| Anchored to the inventory | Each idea names its source asset | Ideas that arrived by brainstorm |
| The gate applied honestly | A stranger with AI genuinely couldn't write any of them | You argued the stranger down on a technicality |
| Two funnel jobs covered | Attract, trust or convert — at least two present | Three ideas doing the same job |
| Price tags believable | Effort matches what the asset still needs | Original research marked "an evening" |
Worksheet
THE SCHOOL OF NET MARKETING
Lesson 5.2 — Your unfair advantage: research, experience, opinion
THE GATE (every idea must pass)
Could a competent stranger with a chatbot and a
search engine produce this piece? If yes — it fails.
IDEA 1
Working headline: __________________________________
Moat: ☐ primary research ☐ experience ☐ opinion+evidence
Inventory asset it draws on: _______________________
Why a stranger with AI can't write it:
____________________________________________________
Funnel job: ☐ attract ☐ build trust ☐ convert
Effort: ☐ an evening ☐ a weekend ☐ fortnight of sessions
IDEA 2
Working headline: __________________________________
Moat: ☐ primary research ☐ experience ☐ opinion+evidence
Inventory asset it draws on: _______________________
Why a stranger with AI can't write it:
____________________________________________________
Funnel job: ☐ attract ☐ build trust ☐ convert
Effort: ☐ an evening ☐ a weekend ☐ fortnight of sessions
IDEA 3
Working headline: __________________________________
Moat: ☐ primary research ☐ experience ☐ opinion+evidence
Inventory asset it draws on: _______________________
Why a stranger with AI can't write it:
____________________________________________________
Funnel job: ☐ attract ☐ build trust ☐ convert
Effort: ☐ an evening ☐ a weekend ☐ fortnight of sessions
SELF-CHECK
☐ Three headlines, publishable as written
☐ Each idea names its inventory asset
☐ Each survives the gate honestly
☐ At least two funnel jobs covered
☐ Effort estimates I'd bet a calendar on
Next: Lesson 5.3 — One channel, done properly.
theschoolofnetmarketing.com/learn/one-channel-done-properly