# Message test — https://align.wal.app/ After reading your page, only 2 of 15 personas could name a reason to pick you over a similar option. - **Page tested:** https://align.wal.app/ - **Audience tested against:** Professionals anyone using AI to write in their name - **Personas:** 15 simulated - **Report:** https://grader.wynter.com/r/cocoon-bT-gT-4 > These answers are generated by AI, scored on Wynter's B2B Message > Layers framework using behaviorally-diverse simulated personas. The > methodology is real and the critique is directional. What a simulated > persona cannot have is a live budget, a renewal coming up, or a boss > asking about this quarter. --- ## 01 · The scores Every persona answered all four questions. These are four independent proportions of the same panel, not stages of a funnel. | Layer | Question | Cleared the bar | Strength | Of those who passed | | --- | --- | --- | --- | --- | | 1. Clarity | Do they understand what you do? | 15/15 | 68% | 1 without hesitation, 11 with reservations | | 2. Relevance | Can they tell what it solves, and who it's for? | 14/15 | 77% | 2 without hesitation, 12 with reservations | | 3. Value | Do they actually want it? | 6/15 | 44% | all with reservations | | 4. Differentiation | Is there a reason to pick you over the alternatives? | 2/15 | 30% | all with reservations | **Brand alignment** (a side metric, not one of the four layers) — 2/15, 31% strength (all with reservations). Does the page read like the company you actually are? **Fix first: Value.** Earliest failing layer, walking the sequence in order — not simply the lowest score. --- ## 02 · What to change, layer by layer Ordered worst-first. Specific edits, not a restatement of the score. ### Differentiation **Add a line near the price contrasting Align with feeding your writing to ChatGPT.** The page names workarounds but never says why a two-hour interview beats a free prompt or an inbox dump. State what the interview gets that uploaded documents cannot. *effort medium · impact high · tested against Give a reason to choose you* **Add a line under "Connect a Sui wallet" explaining why payment is wallet-based.** The wallet requirement reads as an arbitrary obstacle for a $200 purchase. Tie it to the benefit already claimed later: no account, no email, no record of who you are, and say it at the point of payment. *effort low · impact high · tested against Answer the live objection* **Rewrite the opening "Cocoon is architected as a space" line to say what Align produces.** "A cryptographically secure room" tells a comparison shopper nothing about what they receive. Lead with the concrete outcome: a two-to-three hour interview that produces an editable voice file for any AI tool. *effort low · impact high · tested against Concrete over abstract* ### Value **Add a short excerpt of a real aligned.md file under "What you leave with".** Buyers cannot tell what the $199.99 file actually looks like or how it differs from a paragraph of notes. Show ten lines of a real voice profile so the output is visible before payment. *effort medium · impact high · tested against Proof next to the claim* **Add a before/after writing sample beside "experience the difference straight away".** "Experience the difference straight away" asks the reader to take voice-capture quality on faith. Put the same email drafted with and without aligned.md side by side so the difference is readable, not promised. *effort medium · impact high · tested against Proof next to the claim* **Replace the "Refund" paragraph with plain terms: what triggers a refund and when.** "Cocoon delivers a begin. You deliver a completion" leaves buyers unable to tell what failure entitles them to money back. State the condition and the window in ordinary words. *effort low · impact medium · tested against Plain language* ### Relevance **Add a concrete deliverable example to the opening problem paragraph.** "Your communications, your strategy and planning" could describe any knowledge worker. Name the artefact the buyer actually writes, such as a client proposal or board update, so they recognise themselves immediately. *effort low · impact medium · tested against Lead with the use case* ### Clarity **Add one sentence after "Align asks. You answer." describing what the interview extracts.** Readers cannot tell how the interview captures voice rather than just transcribing answers. Say plainly that questions surface how you phrase, structure and hedge, and that the answers become a reusable markdown profile. *effort low · impact medium · tested against Concrete over abstract* ### Brand alignment (side metric) **Add a named audience line in the opening section, e.g. the role and deliverables.** Readers must deduce who this is for from tone alone. Name the buyer outright: consultants, advisors and partners who send client-facing writing under their own name. *effort low · impact high · tested against Name the audience* **Add company name, operator and contact to the footer beside the how-it's-built link.** The page is anonymous, which makes a prepaid crypto purchase look risky. Say who runs Cocoon and how to reach a human. *effort low · impact medium · tested against Answer the live objection* **Cut metaphors like "What happens in between is your instrument" and "a begin".** Atmospheric phrasing reads as a hobby project rather than a professional service. Use the same plain register as the "What you leave with" section throughout. *effort low · impact medium · tested against Plain language* --- ## 03 · What is working ### The opening problem statement lands for readers who already use AI Three respondents said the problem and reader are identified plainly in the first paragraphs and that the problem directly matches their existing AI usage. > The problem was clear fast — "It's extremely good at most things but it can't capture your voice... a generalised version of you" > > — Consultant using AI for writing, Management consulting, 11-50 > "It's extremely good at most things but it can't capture your voice... a generalised version of you" names the problem directly, and "your intellectual property is too precious to become everyone's training data" tells me who's reading: someone who already uses AI to write, already got spooked handing samples to a vendor > > — Content creator using generative AI, Professional services, 1001-5000 ### The reusable voice file is recognised as solving a real recurring pain Respondents named a reusable voice file as addressing recurring client communication work, and one described the value proposition as strong even while withholding purchase. > If it genuinely worked, I'd get a reusable "voice file" I could paste ahead of any drafting prompt so our comms don't read like generic AI output to clients or the board > > — Executive using AI for communications, Marketing and communications, 11-50 --- ## 04 · What the personas said ### Metaphorical, atmospheric language hides how the product actually works Six respondents said the mechanics are buried in spatial metaphor and vague abstraction, forcing them to infer what happens. Several could not tell how the interview encodes voice versus simply transcribing it. > the page buries the mechanics in mood-copy like "a cryptographically secure room, for the words that matter most," so I had to strip out the poetry to get to the actual nouns: interview, file, aligned.md, paste into any AI > > — Manager using AI for strategy and planning, Management consulting, 1001-5000 > it says nothing is stored or used for training, but doesn't explain how the file actually encodes "voice" versus just being a long Q&A transcript > > — Executive using AI for communications, Marketing and communications, 11-50 > the initial friction came from words like "Cocoon," "Align," and "a cryptographically secure room" up top, which are atmospheric rather than descriptive, so the first screenful reads more like a poem than a product spec > > — Business writer using AI assistance, Management consulting, 201-500 > phrases like "architected as a space" and "cryptographically secure room" - that's spatial metaphor doing the work a technical sentence should be doing > > — Professional using AI writing tools, Technology, 1001-5000 > It's the abstraction stacked on abstraction — "architected as a space," "a cryptographically secure room," "what a model needs to understand to carry your voice" — none of those tell me mechanically what happens in the interview or how the file is structured > > — Content creator using generative AI, Marketing and communications, 5000+ ### The audience is inferred from language cues, never named Five respondents said they worked out who the page is for from context rather than an explicit statement, adding translation effort. One wanted a specific failure mode and proof the product understands consulting deliverables. > The "who" is implied rather than stated outright - "professionals," "your clients, your partners or your board" - so I inferred it's for executives/consultants/client-facing people > > — Manager using AI for strategy and planning, Management consulting, 1001-5000 > The reader is inferred rather than named outright — there's no "this is for consultants" or "this is for executives" line — but the cues (board, clients, partners, agency, bios) point squarely at senior professionals > > — Business writer using AI assistance, Management consulting, 201-500 > The intended reader is inferred rather than named outright: nowhere does it say "for executives" or "for consultants," but between "your clients, your partners or your board" and the price tag of $199.99, I can piece together > > — Professional using AI writing tools, Technology, 1001-5000 > I'd need a line naming my actual failure mode — something like "your board deck sounds like every other consultant's LinkedIn post" — plus proof they understand consulting deliverables specifically (proposals, board memos, client emails) > > — Consultant using AI for writing, Management consulting, 11-50 > The intended reader is inferred rather than stated outright — no line says "this is for executives" or "for consultants," but the repeated references to "your clients, your partners or your board" and "your agency" made it obvious > > — Manager using AI for strategy and planning, Technology, 201-500 ### No sample output, case study or before/after leaves the core claim unverified Six respondents said they cannot judge voice-capture quality without evidence: a sample output, a side-by-side against plain ChatGPT or free DIY prompts, or a documented client-noticed change. Absence of named customers blocks purchase. > the page never shows me an example output, and "carries your voice, not your IP" is asserted, not demonstrated, so that's the proof gap that would decide it > > — Manager using AI for strategy and planning, Management consulting, 1001-5000 > I'd take a meeting if someone showed me one comparable firm's aligned.md actually changing how a client or board reacted to their comms > > — Professional using AI writing tools, Financial services, 5000+ > I need proof the aligned.md output is meaningfully better than what I already get from just pasting 3-4 samples of my own writing into a prompt, which is free and takes five minutes, not $199.99 and 2-3 hours. Show me a side-by-side > > — Content creator using generative AI, Technology, 1-10 > But it's a two-to-three hour investment plus $199.99 plus a crypto wallet just to get one markdown file, with no before/after proof shown on the page. > > — Manager using AI for strategy and planning, Financial services, 501-1000 > That's a real, narrow problem and worth solving. But "exactly as promised" is doing a lot of work here - the page never explains what's structurally in that file, what model does the extraction > > — Professional using AI writing tools, Technology, 1001-5000 ### Crypto-only payment is a hard disqualifier, not a friction point Five respondents rejected the Sui wallet and USDC requirement outright, citing enterprise procurement and compliance rules and disproportionate friction for a $200 purchase versus card-paying competitors. > the crypto-wallet-only payment ("Connect a Sui wallet to continue... You pay with a Sui wallet holding USDC or USDsui") would actually count against it on a shortlist — that's friction and a compliance headache for a consulting firm's procurement > > — Manager using AI for strategy and planning, Management consulting, 1001-5000 > That's a real disqualifier for a financial-services rollout — I'm not asking a board member or a partner to go set up a crypto wallet > > — Professional using AI writing tools, Financial services, 5000+ > I don't have a Sui wallet sitting around, and adding "buy USDC on a chain I don't use" as a prerequisite step is friction a $199.99 one-time tool doesn't earn, especially when a competitor probably just takes a card. > > — Content creator using generative AI, Technology, 1-10 > crypto wallet payment and no track record kills it for me first > > — Consultant using AI for writing, Professional services, 51-200 > Sui wallet only, USDC/USDsui payment. No card option, no company track record - too much friction and risk for $200 > > — Consultant using AI for writing, Professional services, 51-200 ### The brand reads as a crypto-native solo operation aimed at individuals, not enterprise… Five respondents said the tone signals a small early-stage founder team selling to solo professionals, misaligned with enterprise and management-consulting buyers. One flagged the anonymous page with no company verification as a long-term trust risk. > probably crypto-native rather than management-consulting-native, given the choice to gate a $199.99 writing tool behind a Sui wallet instead of Stripe > > — Consultant using AI for writing, Management consulting, 11-50 > written to one person deciding for themselves, not to someone who has to build a business case, get procurement and infosec sign-off > > — Professional using AI writing tools, Financial services, 5000+ > Small, early-stage outfit — reads like a two-or-three-person crypto/AI crossover team rather than an enterprise vendor, probably selling to individual consultants, coaches, or personal-brand types rather than companies like mine. > > — Manager using AI for strategy and planning, Financial services, 501-1000 > it still doesn't tell me if there's a real company behind it that'll exist in six months when I need support or a refund honored > > — Content creator using generative AI, Professional services, 1001-5000 ### Respondents who got past the metaphor described the product accurately as a… Four respondents read the product back correctly: it extracts a markdown voice profile upstream of existing AI writing tools rather than being a standalone writer. The plain description does land once it appears. > spits out a "voice profile" file to feed other AI tools so your writing sounds like you > > — Consultant using AI for writing, Professional services, 51-200 > an AI interview tool that spends a couple of hours grilling you with questions to extract "your voice," then spits out a single markdown file you paste into ChatGPT or whatever else you use > > — Executive using AI for communications, Professional services, 1-10 --- ## 05 · The hardest read An adversarial pass over the findings. Every claim below was checked against the panel's own answers; unsupported ones were dropped. - **The page cannot survive a procurement conversation: payment mechanics alone kill the deal before messaging is even evaluated.** *(high)* Five respondents rejected the Sui wallet and USDC requirement outright on compliance and procurement grounds, and five more read the brand as a crypto-native solo operation misaligned with enterprise buyers. Two independent signals point at the same… - **The central claim — that the product captures voice — is asserted and never demonstrated, so the entire page rests on trust it has not earned.** *(high)* Six respondents said no sample output, side-by-side against plain ChatGPT, or named customer left the claim unverified, and absence of proof blocked purchase. One respondent called the value proposition strong while still withholding purchase. - **Metaphor is doing the work that a mechanism explanation should do, and it costs the page its most valuable readers.** *(high)* Six respondents said mechanics are buried in spatial metaphor, several unable to tell encoding from transcription — while four who reached the plain description read the product back correctly. The clear copy works; it arrives too late. - **The page converts interest into refusal rather than indifference — readers who understand and want it still walk.** *(high)* Respondents named the reusable voice file as solving recurring client work and called the proposition strong, yet withheld purchase; four correctly described the product and still rejected the crypto payment. Comprehension and desire are not the bottleneck. - **Readers are made to do the qualification work themselves, which hands the page's fate to their patience rather than its argument.** *(medium)* Five respondents inferred the audience from language cues rather than any explicit statement, adding translation effort, and one wanted a named failure mode plus proof the product understands consulting deliverables. Only three found the reader plainly… - **Positive signals are thin and confined to the first few paragraphs, so the page has no reserve of goodwill to spend on its harder asks.** *(medium)* Only three respondents endorsed the opening problem statement and two the reusable voice file, against five to six on each of metaphor, unnamed audience, missing proof and payment. Every positive theme is outnumbered. --- ## 06 · Who answered | # | Role | Industry | Company size | | --- | --- | --- | --- | | 1 | Manager using AI for strategy and planning | Management consulting | 1001-5000 | | 2 | Professional using AI writing tools | Financial services | 5000+ | | 3 | Content creator using generative AI | Technology | 1-10 | | 4 | Executive using AI for communications | Marketing and communications | 11-50 | | 5 | Consultant using AI for writing | Professional services | 51-200 | | 6 | Business writer using AI assistance | Management consulting | 201-500 | | 7 | Manager using AI for strategy and planning | Financial services | 501-1000 | | 8 | Professional using AI writing tools | Technology | 1001-5000 | | 9 | Content creator using generative AI | Marketing and communications | 5000+ | | 10 | Executive using AI for communications | Professional services | 1-10 | | 11 | Consultant using AI for writing | Management consulting | 11-50 | | 12 | Business writer using AI assistance | Financial services | 51-200 | | 13 | Manager using AI for strategy and planning | Technology | 201-500 | | 14 | Professional using AI writing tools | Marketing and communications | 501-1000 | | 15 | Content creator using generative AI | Professional services | 1001-5000 | --- ## 07 · Before you act on this The methodology is real, and the critique is directional. What a simulated persona cannot have is a live budget, a renewal coming up, or a boss asking about this quarter. **Validate anything you're betting on with real ICPs who are actually in-market.** Being wrong is more expensive than you think. Finding out is cheaper than you'd guess. Wynter runs message testing with verified B2B professionals — trusted by HubSpot, RingCentral, Shopify, Cognism, Paddle, Veeam, Rippling and Miro. This report is kept for 60 days from 2026-09-25, then deleted along with the personas and their answers.