Positioning and conversion
A visitor decides in seconds whether a product is for them, so the page has to say who it is for and what it does before anything else, and clear the trust bar before it asks for money. After that, conversion is fixed one change at a time, measured on the pages that bring signups.
- Written by
- Plan steps
- 29
- Work
- about 47 h
- Checked claims
- 5
The concept, explained in full: Positioning.
On this page4 sections
29 steps from the launch plan
Do it
The steps from the launch plan that belong to this topic, in plan order. A tick here is a tick in your launch plan.
Make the offer legible in five seconds
- Write the one-sentence positioning statement45m
“[Product] helps [specific who] [do what] without [the old painful way].”
- Name the three things you replace30m
A competitor, a spreadsheet, or doing nothing. You need these named for every alternative page later.
- Write the ICP — including who it is NOT for45m
The exclusion half is what makes the copy sharp and the outreach list short.
- Make the H1 state the outcome, not the category30m
“Get paid on time” beats “Invoicing platform”. Categories are for directories.
- Make the subhead name the audience and the mechanism20m
Who it is for, and how it works. Two clauses, no adjectives.
- Cut to one primary CTA, repeated20m
Two competing CTAs halve both. Secondary actions go below the fold.
- Add 5–8 real product screenshots at 1920×10801h
Real UI with real data. Directories and review sites need these later anyway.
- Record a 60–90 second demo video2h
Product Hunt listings with a video tend to get more upvotes. Sources that give a multiplier disagree, from about 2× to 4×, so treat the direction as sound and the number as unknown. Reused a dozen times later.
- Publish plans with real numbers1h
Hiding price filters out exactly the self-serve buyers you can actually win at this stage.
- Find five products already charging for this problem1h
Competition is validation, not a warning: an empty category is far more often a dead one than an unclaimed one.
- Write down how each of them actually gets customers2h
Their social accounts, the posts other people made about them, their ad library entry. The channel mix is public, and it is your channel-tournament candidate list.
- Choose the version of the product you can show, not explain45m
A thing that demonstrates in seven seconds and a thing that needs three minutes are different businesses. Decide which one you are building while it is still free to change.
Clear the trust bar
- Add a consent banner that defaults to declining45m
Only if you take EU or UK traffic: both require an active opt-in before any non-essential cookie is set, and a pre-ticked box is not consent. Default-on tracking is the fine, not the banner.
- Write an About page with a real name and face45m
Anonymous SaaS converts worse and gets rejected by curated directories.
- Publish a support address a human reads15m
While you chase the first ten customers, that human is you, and the replies are the research.
- Make every promise and badge on the site true before you send traffic to it30m
A form that says “every Monday” with no Monday email, a “listed on” badge whose listing 404s, a count that stopped being right months ago: each one makes the rest of the page read as invented. Send the first issue before you promise a schedule, and show a badge only for a listing you have opened this month.
Convert the traffic you already have
- Test the homepage, one change at a time5h
Headline first — it moves more than anything else on the page.
- Test the pricing page5h
Plan naming, ordering, the annual toggle, and what the free tier is allowed to do.
- Add one exit-intent offer2h
The lead magnet from your email course, not a discount. Discounts train people to wait.
- Test a free tier against an extended trial10h
This decision shapes your CAC, your churn and your support load for years. In RevenueCat’s 2026 app data a hard paywall converts about five times freemium at day 35 (10.7% against 2.1%; its benchmarks post calls that trial-to-paid and its paywall post calls it download-to-paid), with the same one-year retention for yearly subscribers (27% against 28%). Trials of 17–32 days convert about 70% better than three-day trials. Test it deliberately.
- Rank your landing pages by the share of visitors who reach checkout, not by visits45m
The page with the most traffic is often the worst buyer. On one audited site the “free, no sign-up” pages brought half the organic visitors and reached checkout at 6.5%, against 26.7% for a page about a feature. Write the next page for the intent that pays.
- Show the free tool’s answer on the page, and send the full version by email2h
Gate the whole answer and most visitors leave; give all of it away and you capture nobody. Show enough to be useful, then offer the complete result or plan to their inbox. That email is one they asked for, so it gets opened, and it starts the list.
- Put a click-through demo on the homepage, nine to twelve steps, with no form in front2h
In Arcade’s benchmark of 14 million demo sessions, completion first peaked at nine to twelve steps, and in its controlled comparison screenshot demos were finished about twice as often as video-heavy ones. That is a vendor measuring its own product: judge yours by the signup rate of visitors who opened it against those who did not.
- Send one recovery email to people who left checkout3h
Someone who reached checkout already decided once. A link that reopens the same checkout removes the work of starting again.
- Send a one-person, one-week code after the first result2h
A deadline only works if it is real. A code tied to one customer and one date cannot be passed around or quietly extended.
Strongest first
Plays
Reusable patterns that explain how: backed by read sources first, then practitioner advice, then our own ideas.
- Backed / 1 sourceTrade your first five deliveries for an honest review and a write-upA new service has no proof, so give your first few deliveries free in exchange for an honest review and an anonymised before-and-after you may publish. Each one becomes a case card and a short thread, the label keeps it honest, and five real results persuade more than any claim on the homepage.Read the play
- Our ideaSay the same sentence everywhere a machine looksWrite one sentence naming your category and your buyer, then propagate it verbatim across your homepage, listings, review profiles and community answers. Retrieval systems assemble a picture of you from many sources at once; when they all agree, the claim reads as consensus, and when they contradict each other the model hedges or leaves you out.Read the play
- Our ideaThe "what is [product]" page, answered in one sentence firstPublish a page whose first sentence answers "what is [product]?": the category, who it is for, and the one thing it does differently, followed by what it is not and who should pick something else. Branded searches and AI answers pull from this page, and if you do not write it, a directory or a competitor writes it for you.Read the play
5 checked claims
What the evidence says
- Research on how to launch a product effectively on Product Hunt
- Storing cookies requires internet users’ active consent (Press release 125/19, Case C-673/17 Planet49)
- Cookies and similar technologies (Guide to PECR)
- Subscription app trends and benchmarks 2026 (State of Subscription Apps)
- Hard paywall vs freemium
- 2026 Arcade benchmarks
Related topics
In the launch plan
Where it sits in the plan
Share this topic