What is Amazon's Future Press Release method?

It's a one-page document — written before any code — that announces the product as if it had already launched. It contains a headline, a sub-headline, the customer problem, the solution, a customer quote, and an FAQ. Amazon has used a version of it (the PRFAQ) since the early 2000s as the first artifact in any new product proposal. The point isn't the document — it's the forced clarity that writing it produces.

How long should writing a Future Press Release take?

Thirty to ninety minutes for a first draft. If it takes longer than three hours, the offer is too vague to validate — that's itself the signal. The exercise should reveal vagueness quickly. If you can't fit a coherent version on one page, you don't yet have an idea sharp enough for a landing page.

Does a Future Press Release replace a landing page test?

No. A Future Press Release is a thinking tool, not a market test — it tells you whether you have a clear offer to validate, not whether the offer resonates with strangers. The PR comes first; the landing page (with paid traffic) comes second. Skipping the PR is how founders end up running ads against a confused offer.

What should be removed from the Future Press Release?

Anything aspirational ("revolutionary", "AI-powered", "next-generation"), anything technical ("built on a transformer architecture"), anything internal-facing (roadmap, milestones), and anything the customer wouldn't say out loud. The five rules — no jargon, customer language, specific benefits, no aspirational tech, fits on one page — exist to strip the document down to what actually matters.

Can I share my Future Press Release publicly to test the idea?

You can, but the signal is weak. Friends will tell you it sounds great. The real test is converting that PR into a landing page headline, running €150–€250 of paid traffic, and watching whether strangers click and sign up. The PR sharpens the message; the ad spend tells you whether the message has a market.

Amazon's Future Press Release Method: Validate Your Startup Idea Before Building

We've watched founders spend three weeks building a prototype for an idea they couldn't describe in two sentences.

It happens like this. Someone has a hunch. The hunch becomes a Notion doc. The Notion doc becomes a Figma file. The Figma file becomes a Next.js repo. Three weeks in, you ask them "what does this product actually do for the customer?" and you get a thirty-second pause followed by four sentences that contradict each other.

The fix is older than most of us realize. Amazon has been using it since the early 2000s. It's called the Future Press Release — sometimes the PRFAQ — and it kills bad ideas in about thirty minutes for the cost of an empty page.

Our thesis: before you build anything, before you even buy a domain, you write the launch announcement. If you can't draft a credible one-page press release for your product today, you don't have an idea — you have a vibe. And vibes don't convert at 4%.

What is the Future Press Release method?

Amazon calls it "Working Backwards." Every new product proposal at Amazon starts with a fictional press release dated the day of launch, plus an internal FAQ. No slide deck, no spreadsheet — a one-page narrative announcement, written from the customer's point of view, before a single engineer is staffed. The practice traces back to the early Kindle and AWS days; Lenny Rachitsky's PRFAQ deep-dive is the cleanest public write-up.

The mechanism is brutal and simple. A press release has no room for hand-waving. You can't announce a product that does "a lot of things for a lot of people." You have to name the customer, name the problem, name the benefit, and quote a believable user. Anything else gets cut. The discipline of writing under those constraints surfaces every weak assumption in your idea, fast.

The one-page template

The Future Press Release fits on a single page. Six sections, in this order:

Headline — one line, max twelve words. What launched, for whom, with the core benefit. No adjectives. "Stripe Atlas lets non-US founders incorporate a Delaware C-corp in under ten minutes."
Sub-headline — one or two lines. Who it's for and the secondary benefit. Anchors the audience.
Problem paragraph — three or four sentences. The pain in the customer's words, with specifics. "Founders outside the US used to spend €2,500 and six weeks setting up a US entity, dealing with three lawyers and a notary."
Solution paragraph — three or four sentences. What the product does (not how). The benefit, the trigger, the outcome. Concrete verbs only.
Customer quote — two or three sentences attributed to a fictional but realistic customer. First person. The quote captures the "before" and the "after" in a single voice.
FAQ — five to seven internal questions you'd expect a skeptical reader to ask. Pricing. Differentiation. Why now. What it doesn't do. Honest answers, not marketing.

That's the entire document. One page. If yours runs to two pages, you haven't cut yet.

The five rules

The template is the easy part. The rules are where founders fail.

1. No jargon. If your mom couldn't read the press release and explain what your product does, the press release is broken. "Leveraging AI-powered workflows to streamline B2B SaaS GTM" is not a sentence. It's an ad for the writer's LinkedIn.

2. Customer language only. Use the words your customer would actually say. Not the words you wish they'd say. "I hate setting up Make zaps" beats "workflow automation friction." Steal phrases from real Reddit threads, support tickets, sales calls.

3. Specific benefits, with numbers. "Saves time" is not a benefit. "Cuts payroll reconciliation from 4 hours to 12 minutes" is. If you can't put a number on the benefit, you don't yet know what the product does.

4. No aspirational tech. The press release describes the product as it ships on day one — not the AGI version you'll build in year three. If a feature requires research, it doesn't go in the PR. Amazon's rule of thumb: if it isn't in the v1, don't announce it.

5. Fits on one page. Hard cap. One page in 11pt, single-column, with whitespace. If you're cheating with 8pt font and zero margins, you're cheating yourself. The constraint is the point.

Founders we've worked with break rule 4 most often. The PR is so easy to fill with the dream version of the product that you forget you're supposed to ship the day you publish it.

Where it fits in the pre-MVP validation flow

The Future Press Release is the first artifact, not the only one. It's the cheapest filter in the stack — and the one most founders skip.

The full pre-MVP sequence we use:

Future Press Release (30–90 minutes, €0). Forces offer clarity. Kills vibes-only ideas before they cost anything.
Landing page from the PR (2–4 hours, €0–€20). The headline becomes the hero. The customer quote becomes social proof. The FAQ becomes the FAQ.
Paid traffic test (10–14 days, €150–€250). Reddit, Meta, or Google Search ads. Pre-committed conversion threshold. Real strangers, real money.
Conversation with the first 5 signups (2 hours). If your PR was honest, these calls will sound like the customer quote. If it was wishful, they won't.

Step 1 is the gatekeeper for steps 2–4. If the press release isn't crisp, the landing page won't be either, and you'll be running ads against a confused offer. Confused offers convert at 0.6%. We've watched it happen more than once. Related: why validation comes before the MVP and the broader 2026 pre-MVP validation playbook.

A worked example: a fictional SaaS

Here's a Future Press Release we drafted for a fictional but plausible idea. Time on draft: 38 minutes.

The idea: a tool that auto-generates monthly investor updates for early-stage founders by pulling from Stripe, HubSpot, and Linear.

FOR IMMEDIATE RELEASE — Berlin, May 2 2026
Headline: Stipend lets seed-stage founders send a polished investor update in under 5 minutes a month.
Sub-headline: For solo and small-team founders who hate writing investor updates and ship them late or not at all.
Problem: Investor updates are the single most common task seed-stage founders skip. The average founder we surveyed spends 2 to 4 hours per month writing one, dreads it, and ends up sending it 8 days late or not at all. Late updates erode investor trust at exactly the stage where trust converts into the next round.
Solution: Stipend connects to Stripe, HubSpot, and Linear, pulls the month's revenue, pipeline, and shipped features, and drafts a clean update in the founder's voice. The founder spends 5 minutes editing, hits send, and the email goes out before the 5th of the month. Existing investors get a consistent rhythm. New investors get a clean paper trail at the next raise.
Customer quote — Léa K., founder, B2B SaaS, Paris: "I used to dread the first of the month. I'd rewrite the same paragraphs from memory and ship the update a week late. With Stipend, I review what it pulled, edit two lines, and the update goes out by 9 a.m. on the 1st. My lead investor told me I look more on top of things than founders three rounds ahead of us."
FAQ:
Pricing? €29/month flat. No seat-based billing.
Why now? Stripe and HubSpot APIs are stable; LLM-drafted prose is now good enough to ship without heavy editing.
What it does NOT do: custom dashboards, board decks, fundraising CRM. Stipend is one workflow, done well.
Differentiation vs Visible.vc? Visible is a portfolio dashboard for VCs and a heavyweight platform for founders. Stipend is a 5-minute monthly email.
Data security? Read-only API keys, EU hosting, zero data retention beyond the current month.

Notice what the exercise did. The headline forced us to commit to one user (seed-stage founders, not Series B), one job (the monthly update, not all reporting), and one promise (5 minutes a month, not "saves time"). The FAQ forced us to admit what the product doesn't do — which is what kills 60% of pre-MVP ideas, because the "does not do" list usually reveals the actual market is too small.

From here, the landing page hero writes itself. The headline goes in 80pt, the sub-headline below, the problem becomes the "why now" section, the customer quote becomes the social-proof block, the FAQ becomes the FAQ. Two hours of work, €12 in domain + Vercel, and you're ready to run ads on the next step of the pre-MVP flow.

What to do when the PR fights back

The PR will resist you. That's the feature, not the bug. A few patterns we see when the exercise is working:

You can't write the customer quote. If you can't put words in a believable customer's mouth, you don't know your customer. Stop. Go talk to three people who match your imagined ICP before continuing.

The FAQ "what it does NOT do" list is empty. Every product worth building does a small set of things and refuses to do most others. If your no-list is empty, the offer is too broad to validate.

You can't finish in 90 minutes. That's the strongest signal of all. The exercise should be uncomfortable but doable. If hour three rolls around and you're still rewriting the sub-headline, the underlying idea isn't sharp enough yet.

How LemonPage fits

The Future Press Release is the thinking tool. LemonPage is what you do with it next: the headline becomes the page hero, the customer quote becomes the testimonial slot, the FAQ becomes the FAQ section, and the page is live with paid traffic running against it inside a day. The PR sharpens the message; LemonPage measures whether the message converts. The two pair so naturally that we now ask every founder we work with for their PR before they touch the editor.

If you can't announce it on one page, you can't ship it. Write the press release first. The rest gets easier.