If your product still changes every week, your landing page feels like a lie within ten days. You ship a clearer flow, kill a half-baked feature, rename a plan—and the homepage still promises last month’s product. Visitors bounce. You rewrite the whole page again. Two weeks later you are back in the same loop.
That loop is not a writing problem. It is an ops problem. Early products move. Landing pages that pretend they are permanent monuments break under that motion. This playbook is for indie and early SaaS founders who need a page that still converts while the product is honestly unfinished—and a ritual for updating it without a full rewrite every ship.
This is not another positioning workshop. If you still need a one-page who/job/why brief before you touch copy, do that first (IndieHunt’s positioning brief before a homepage rewrite is the right prep). Here we assume you already know the job. The question is how to structure and maintain the page when that job’s surface area keeps shifting.
Freeze one promise for two weeks
The page needs a frozen promise even when the product is not frozen. Pick one sentence you will not rewrite for fourteen days:
[ICP role] can [complete one job] without [the painful workaround they use today].
Examples:
Solo founders can collect structured waitlist intent without a spreadsheet and five Zapier zaps.
Freelance designers can send a client-ready weekly status without rebuilding it in Docs.
Ops leads can hand off a recurring checklist without Slack archaeology.
Everything on the page either supports that promise or gets parked. New features that do not change the promise become a changelog note or a “what’s ready” bullet—not a new hero. If you cannot hold the promise for two weeks, you do not have a landing page problem; you still have a product focus problem.
Structure the page around one job-to-be-done
Build the page as a path through one job, not a catalog of capabilities.
Hero: who it is for + the frozen promise + one primary CTA (signup, demo, waitlist—pick one).
Before / after: the painful workaround vs the job completed in your product. Two short paragraphs beat a feature grid.
How it works: three steps max. Name the steps after user actions, not product modules.
What’s ready now: an honest list of what a new user can complete this week. Update this weekly; leave the hero alone.
Proof: one quote, one defended number, or a short Loom—whatever you can stand behind. Thin proof beats fake logo walls; for honest proof when you have almost no users, see how to write social proof for an indie launch when you have almost no users.
Who this is not for: one blunt line. It filters bad-fit traffic and keeps you from stretching the promise.
CTA again: same action as the hero. Do not invent a second conversion goal mid-page.
If a section does not help someone decide whether they can complete the job, cut it. Comparison tables, twelve integrations, and “AI-powered” adjectives are usually theater at this stage.
Use modular sections you can update without a rewrite
Treat the page like Lego, not a novel. Each section should be editable in isolation.
Section | Update cadence | What you change |
|---|---|---|
Hero promise | Every 14 days max | Only if the core job or ICP actually changed |
What’s ready now | Weekly | Bullets for ships that affect the job; remove dead claims |
How it works | When the flow changes | Step names and screenshots—not the whole narrative |
Proof | When you earn a better quote or number | Swap the artifact; keep placement |
Pricing / plans | When buyers ask | One clear offer; hide “coming soon” tiers |
Operational rule: when you ship, open the landing page checklist before you open Twitter. Ask: does this ship change the promise, the steps, or only “what’s ready”? Update the smallest section that keeps the page honest. Do not rewrite the hero because you renamed a button.
Honest “what’s ready” framing beats fake completeness
Visitors smell “coming soon” walls. Early buyers prefer clarity.
Say what they can finish today: “You can invite three teammates and run the weekly checklist end to end.”
Say what is intentionally later: “SSO and custom roles ship after we stabilize the core handoff.”
Avoid apologizing for every missing feature. Ambition without a date reads unfinished; a named cut reads intentional.
If messaging still feels mushy, pause and fill a positioning brief before another homepage rewrite—how to write a one-page positioning brief before you rewrite your homepage is the prep step. Then come back and update modules, not the entire page voice every Monday.
Measure conversion, not vanity
A pretty rewrite that does not move signups is a hobby. Track a short list for two weeks:
Unique visitors → primary CTA clicks (hero + bottom CTA combined)
CTA → completed signup (or demo booked—pick one)
Signup → first win in-product (your activation event)
Qualitative: three stranger replies that mention the promise in their own words
Ignore session duration theater and homepage bounce rate alone. If CTA click rate is fine but signup completion collapses, the form or auth is the bug—not the headline. If signups arrive and nobody reaches first win, stop polishing the landing page and fix the first ten minutes in-product. That split matters: I stopped obsessing over the landing page and fixed the first 10 minutes is the complementary fix when the page is “good enough” and activation is not.
A 14-day landing-page sprint
Days 1–2 — Freeze and sketch. Write the one-sentence promise. Outline the seven sections above. Delete any section that does not serve the job. Screenshot the current page so you can see what you are replacing.
Days 3–5 — Ship a modular v1. Hero, before/after, how-it-works, what’s-ready, one proof artifact, not-for line, CTA. Use real product screenshots even if they are ugly. Ugly and true beats polished and stale.
Days 6–9 — Traffic with intent. Send the page to warm ICP people and one public channel where buyers already hang out. Tag the links. Do not A/B-test five headlines; you do not have the volume.
Days 10–12 — Update ritual only. After each ship, update “what’s ready” and how-it-works if needed. Leave the hero alone unless a stranger repeatedly misunderstands the job.
Days 13–14 — Decide with numbers. Keep, tighten, or rewrite one section based on CTA→signup and signup→first win. Schedule the next promise freeze date. Do not start a full redesign because one friend said the colors feel “off.”
Traps that keep the page broken
Feature laundry lists. Twelve bullets that name modules do not explain the job. Visitors skim and leave. Lead with the outcome.
Rewriting the whole page every ship. That is how you never learn which line converted. Modular updates create history you can compare week over week.
Hero that tracks the roadmap. If the headline changes whenever you merge a PR, you trained the page to be unstable. Freeze the promise; park features below.
Fake completeness. “Coming soon” grids and grayed-out enterprise tiers signal vaporware. Honest scope converts better at this stage.
Two CTAs fighting each other. Book a demo and self-serve signup on the same hero usually means you have not chosen a motion. Pick one primary action for fourteen days.
Ignoring post-click reality. A landing page cannot save an empty dashboard. Pair page work with activation work or you will optimize the wrong layer.
Closing
Freeze one promise. Structure the page around one job. Keep sections modular so weekly ships update “what’s ready,” not the whole narrative. Measure CTA→signup→first win. Run the ritual for two weeks before you declare the page “dead.”
Early products will keep moving. That is fine. The page’s job is not to be perfect—it is to stay honest long enough for the right people to try the job you actually support this week.
When that early product is ready for people who hunt for tools still taking shape, boards like EarlyHunt help you get in front of early adopters without pretending you shipped a finished cathedral.