Three rules that apply from today to the end of the program.
1. Nothing goes out without approval. Everything you make from here is real work on a live product. Drafts, briefs, research and plans — produce them freely. But no publishing, no sending, no posting, no live page changes, and no contacting anyone, without asking me first. When in doubt, it needs approval.
2. Your Day 8 positioning is a hypothesis, not settled truth. If something you learn this week shows it was weak, revise it. Do not force six pieces of work to fit an assumption you no longer believe.
3. Every task carries the same five things — objective, evidence, deliverable, review, and a signal with a stop rule. They are on every task sheet from here.
Friday's map ended with a question it did not answer: all five channels deliver someone somewhere. Today is the somewhere. This is the first thing you will make that could genuinely ship — so it is also the first time the standard matters.
- Nearly every channel ends at a page.
- Message match — the promise that brought them has to be there on arrival.
- Lead with the problem. A reader who does not recognise themselves never reaches the features.
- A value proposition is an outcome, not a feature list.
- One page, one action.
- Objections belong on the page, not in your inbox.
- Write simpler than feels professional.
- Herald — and the line it will not cross.
- What this lesson did not cover, and how to go deeper.
1. Nearly every channel ends at a page.
An article, a post, a cold email, a community reply — most of them end with someone clicking through. What they land on decides whether that work mattered.
Not all of them. A community reply can solve the problem outright, a post can build familiarity with no click, an email can want a reply rather than a visit. But every channel needs a defined next action, and for us it is usually a page.
A page that cannot move the right visitor forward reduces the value of most of the effort that brought them there. Not all of it — they may still learn something, or remember you later. But the expensive part is lost, and it is lost silently.
This is why pages come before content and well before search. Optimising to send more people to a page that does not work is a way of scaling a problem.
Traffic is the part everyone measures. The page is the part that decides what the traffic was worth.
2. Message match — the promise that brought them has to be there on arrival.
Someone clicked for a reason. Something promised them an answer. That promise must be visible, in recognisable words, in the first screen.
When it is not, the visitor does a fast, unconscious check — is this the thing I clicked for? — decides no, and leaves. They will not scroll to find out they were right.
Breaks: the post said "stop rewriting every post for five platforms". The page opens with "The complete social media management platform."
Matches: the post said the same thing, and the page opens with "Write it once. It arrives right on every platform."
The practical consequence: you cannot write a page without knowing what sent people to it. One page serving five very different promises usually serves none of them.
3. Lead with the problem. A reader who does not recognise themselves never reaches the features.
The instinct is to open with what the product is. The reader has not yet agreed they need one.
So the first job of the page is recognition — the visitor seeing their own situation described accurately enough that they think this is about me. Only then does what you built become interesting.
This is where Day 7 pays off. You collected the words real people used about this problem. Those words belong here — not the internal vocabulary.
Internal: "Persistent brand context across sessions."
Theirs: "It knows my business."
The second is not dumbed down. It is the sentence a customer would actually say, and a reader recognises it instantly.
4. A value proposition is an outcome, not a feature list.
A feature is what it does. A value proposition is what changes for the person — ideally with something concrete attached.
| Feature | Per-platform drafting from saved brand context |
| Outcome | Write the post once; it arrives right on every platform — and it already knows how you talk |
And the test that kills most first drafts, which is the swap test again in a different coat:
If the sentence would be equally true of a competitor, it is not a differentiator. Rewrite it or drop it. "Powerful", "easy to use", "all-in-one" and "save time" all fail — every competitor claims them, so they carry no information.
5. One page, one action.
Every additional ask reduces the first one. Start free trial, book a demo, read the docs, join the newsletter, follow us — five asks means most people do none.
Decide the single thing you want the right visitor to do, make it obvious, and repeat it rather than adding alternatives.
Two practical points. Button words matter more than they should — "Start free" and "Submit" perform differently for the identical action. And say what happens next. "Start free — no card needed" removes a hesitation that "Get started" leaves sitting there.
6. Objections belong on the page, not in your inbox.
Every visitor has reasons not to act, and they are thinking them whether or not you address them. Answering them is not defensive — it is the difference between a page that persuades and a page that only describes.
You already know what they are. Day 7's three-star reviews are a list of objections written by the people who have them.
How long will this take to set up? · What happens to my existing accounts? · Can I get out of it? · Is it going to sound like a robot? · What does it cost when I grow?
An FAQ is where the ones that do not fit the flow go. It is not filler, and it is the section most likely to be read by someone close to deciding.
7. Write simpler than feels professional.
This is the part that surprises people.
A note on numbers, and it applies to me too. You will find industry analyses claiming simple copy converts several times better than complex copy — one widely repeated comparison puts it at roughly 13% against 2%.
I have not traced that to a primary study, and neither has most of the internet repeating it. So treat it as hypothesis, not evidence — exactly as Day 7 told you to. The direction is consistent wherever it gets measured. The precise multiple is not something either of us can defend.
What is not in dispute: on a page, complexity costs you readers. Short sentences, concrete nouns, active voice, very few long words.
Cut every intensifier that survives only because it sounds confident.
Complex writing on a landing page is almost never precision. It is usually a writer protecting themselves from being wrong — and the reader, who owes you nothing, simply leaves.
8. Herald — and the line it will not cross.
Herald is the agent that owns conversion copy: the pitch, the homepage, the pricing page, the FAQ, the launch announcement.
Its own first instruction is the one worth taking:
Research tells you what is true. You decide how to say it.
It is told never to start from a blank page and its own instincts — it reads the product notes, the ICP, the objections, the positioning and the competitors first. Which is why this lesson sits after Day 7 and Day 8 rather than before them.
The rule that matters most is about claims. Herald has a constraints note recording what marketing may not say: metrics we cannot support, testimonials we do not have, capabilities not yet shipped. And the instruction is blunt — a line that violates it is not a draft to be edited later, it is a line that cannot exist.
When a claim is not backed by evidence, it has three honest options: find the evidence, weaken the claim to what is true, or cut it. The fourth option — write it anyway and hope — is the one closed off.
That constraint exists because of a specific failure. Ask any assistant for a landing page and you will get "trusted by thousands of businesses", because that is what landing pages say. It does not know whether it is true and it will not ask you.
Two boundaries. Herald writes; it does not publish, post or send — a person ships a page. And it does not research: if it needs to know what customers say or what a competitor charges, that is Scout's job, not a guess marked "to verify".
And it does not decide whether its own work is ready. Its instructions say so directly: you are the author, so you are the worst judge of whether your own copy is clear. Everything public goes to Editor first.
That is tomorrow.
9. What this lesson did not cover, and how to go deeper.
Covered: why the page is the hinge, message match, leading with the problem, outcomes over features, one action, objections, and writing simply.
Not covered:
- Design and layout — visual hierarchy, what the eye does first. Related discipline, different skill.
- A/B testing — how you would actually know which version won, and why most small sites never get enough traffic to find out.
- Page speed and mobile — a slow page loses people before a word is read.
- Forms and friction — every field costs you completions.
- Pricing page psychology — its own subject, and a harder one.
- Long-form sales pages — a different format with different rules.
Where to go. The fastest way to improve at this is to read the homepages of five products you have personally bought, and work out what each one's first screen was doing. You will learn more in half an hour than from any listicle of tips.
Today’s task — rewrite the LotsSocial hero (75 min)
Objective: a first screen that a real visitor from one specific channel would recognise and act on.
Evidence: your Day 7 research and Day 8 positioning.
Deliverable: two versions of hero + one section, a claim inventory, and your verdict.
Review: none today — that is tomorrow's lesson, and this is what you will hand it.
Signal: would the visitor you described take the action you asked for?
Do the required deliverable first. Anything beyond it is optional — a finished minimum beats an unfinished maximum.
Reference: the real LotsSocial notes are at /lotssocial — look things up rather than reading it through. Note 8 is the one to read once, before you write anything: it lists what we may not claim.
Herald is at lotsagent.com/agents/herald. Editor, when you need it, is at lotsagent.com/agents/editor.
1. Answer four questions before writing a word (10 min)
- Who is arriving? One person, from your Day 8 ICP. Not "small businesses".
- What did they expect to see? What did they click, and what did it promise?
- What do we want them to do next? One action.
- What might stop them? The strongest objection, not the easiest one.
- What must they believe before they act? Conversion is not only wording — it is whether the reader has enough confidence to move.
If you cannot answer the first two, you are not ready to write. Go back to Day 7.
2. Write it yourself (25 min)
The hero — headline, subhead, call to action — and one other section you think matters most. Not the whole page.
Use the words you collected from real people. Run the swap test on your headline before you move on.
3. Now run the same brief through Herald (15 min)
Same four answers, same positioning, same evidence. Give it what you gave yourself.
Then compare honestly. Which is better, and why? Name the specific line where one beats the other. "Herald's was more polished" is not an answer.
4. Build the claim inventory (20 min)
Go through both versions and highlight every factual or product claim. Mark each one:
- Verified — you have seen the evidence and can link to it.
- Supported by product knowledge — the site or docs say so.
- Assumption — you believe it but have not checked.
- Unsupported — you cannot back this at all.
Be strict, especially with your own version. Anything in the last two categories is either evidenced, weakened to what is true, or cut. There is no fourth option.
5. One line on the signal (5 min)
If this page went live, what single thing would tell you it was working? Not visits.
Then post one line
In #daily-log: one claim you had to cut or weaken, and which category it fell into.
The cut ones are more interesting than the kept ones. Everyone writes at least one sentence they cannot back — noticing it is the skill.
Tomorrow — Day 11: The review gate. You have made something that could ship. Tomorrow is how you decide whether it should — with Editor, and with a checklist you will use for the rest of this program.