# Message test — https://www.elephan.ai/ After reading your page, only 0 of 14 personas could quickly tell what problem it solves and who it is for. - **Page tested:** https://www.elephan.ai/ - **Audience tested against:** B2B Sales Managers, RevOps Managers and B2B Sales Directors working in companies with 500 up to 2000 employees. - **Personas:** 15 simulated - **Report:** https://grader.wynter.com/r/inteligencia-de-receita-para-vendas-e-cs-eleph-Vwdyn7o > 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 | 44% | — | | 2. Relevance | Can they tell what it solves, and who it's for? | 0/14 | 3% | — | | 3. Value | Do they actually want it? | 0/15 | 4% | — | | 4. Differentiation | Is there a reason to pick you over the alternatives? | 0/15 | 1% | — | **Brand alignment** (a side metric, not one of the four layers) — 0/15, 44% strength. Does the page read like the company you actually are? **Fix first: Relevance.** 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. ### Relevance **Add a problem paragraph above the fold describing the forecast pain in the buyer's words.** The page never names a problem, so a reader cannot tell whether it applies to them. State the situation: deal data scattered across calls and CRM, forecasts that miss. *effort medium · impact high · tested against Problem before solution* **Add a subhead under the headline naming the buyer: RevOps and sales leaders.** Nothing on the page says who it is for; readers guessed from a single headline. Write a line naming the role and team, such as RevOps and sales leaders running a forecast. *effort low · impact high · tested against Name the audience* **Replace the rhetorical headline with one naming the job the product does.** "ChatGPT ou Claude substituem uma plataforma de Revenue Intelligence?" poses a question instead of saying what the buyer gets. Lead with the job: keeping every deal's real status visible before the forecast call. *effort low · impact high · tested against Lead with the use case* ### Value **Fix the broken SVG so body copy, features and screenshots render as readable text.** The page ships raw markup, so only a headline and a login button are readable. Restore the body copy and product visuals so a reader can see what the thing is. *effort medium · impact high · tested against Show the product early* **Add a three-step section explaining how the product captures deal data and produces output.** There is no explanation of how the product works, so nothing justifies a meeting. Describe what it ingests, what it analyses and what the user sees. *effort medium · impact high · tested against Tie the feature to the outcome* **Add a customer outcome with a number next to the main claim.** No metrics, pricing or before-and-after results appear anywhere. Put a named customer result beside the claim, such as forecast accuracy moving from one figure to another in a stated timeframe. *effort medium · impact high · tested against Proof next to the claim* ### Differentiation **Add a section stating what this does that a general chatbot cannot.** The headline raises the ChatGPT comparison and then drops it, leaving the reader to answer it alone. Answer it directly: the specific deal and call data the product holds that a chat tool never sees. *effort medium · impact high · tested against Answer the live objection* **Add a short list of the failure modes this product catches that others miss.** Nothing separates this from established revenue intelligence tools. Name two or three concrete situations it detects, such as a champion going quiet or a deal with no economic buyer engaged. *effort medium · impact high · tested against Give a reason to choose you* **Add a line stating the narrow segment or motion this is built for.** A reader comparing options sees no reason to pick this one. State the focus plainly, for example B2B SaaS teams selling in Portuguese-speaking markets. *effort low · impact medium · tested against Specifics beat superlatives* ### Brand alignment (side metric) **Add customer logos and a founder or company line below the hero.** With no copy rendering, the page reads as an unfinished experiment rather than a working vendor. Show who uses it and who builds it so the product looks operational. *effort medium · impact high · tested against Proof next to the claim* **Replace "Acessar conta" as the only action with a single primary demo CTA.** The lone login button speaks to existing users and gives a first-time visitor nothing to do. Make the primary action a demo or trial request, with login demoted to the nav. *effort low · impact high · tested against One clear next action* --- ## 03 · What is working ### The headline question was provocative enough to register on its own Two respondents called the single legible headline compelling or provocative, though both noted nothing behind it supported the claim. > "ChatGPT ou Claude substituem uma plataforma de Revenue Intelligence?" which is a sharp, provocative question and honestly a better hook than most of the bland "AI-powered revenue platform" copy I usually see > > — Sales Manager, Sales and Revenue Operations, 501-1000 > a provocative title with a broken or unrendered page behind it is a mark against them, not neutral — if I can't get the page to show me their argument on the first click, that's a data point about their execution > > — Director of Sales, Sales and Revenue Operations, 1001-5000 --- ## 04 · What the personas said ### The page shipped as raw SVG code and almost nothing was readable Nine respondents saw raw SVG markup instead of rendered marketing copy. Only a single headline and a login button were legible as text. > the page loaded as a pile of raw SVG path code and coordinates, not actual copy > > — Sales Manager, Sales and Revenue Operations, 501-1000 > the page didn't render as content for me, it was just raw SVG/path markup (coordinates, fill colors, transform values) > > — Senior Sales Manager, Software as a Service (SaaS), 1001-5000 > page was just SVG/icon code, no actual copy loaded > > — B2B Sales Director, B2B Technology, 501-1000 > the page dissolved into raw SVG/vector code — path coordinates, fill colors, transforms. There's no actual product description, no mechanism, no proof points, nothing I could evaluate. > > — Director of Sales, Sales and Revenue Operations, 1001-5000 > the page was just raw SVG/vector code (path data, coordinates, fill colors) with a single headline > > — RevOps Manager, Software as a Service (SaaS), 501-1000 > The only human text that came through was a headline asking whether ChatGPT or Claude can replace a "Revenue Intelligence" platform, plus an "Acessar conta" login button > > — Revenue Operations Manager, B2B Technology, 1001-5000 > the page didn't render as actual content for me, just a wall of SVG path coordinates and colour codes > > — Revenue Operations Manager, B2B Technology, 1001-5000 > The page that loaded for me was basically just raw SVG/vector code — path coordinates, fill colors, transform values — not actual marketing copy. > > — Director of Sales, Sales and Revenue Operations, 1001-5000 > the page loaded as raw SVG/vector code (path coordinates, fill colors, transform values) instead of actual content > > — RevOps Manager, Software as a Service (SaaS), 501-1000 ### Audience and problem were never stated, only inferred from the headline Eight respondents said no body copy named the problem, use case, or buyer. Relevance to RevOps or revenue teams had to be guessed from a single Portuguese headline or the page title. > That headline implies the problem is "can general LLMs replace a Revenue Intelligence platform," and it implies the reader is someone currently using or considering ChatGPT/Claude for revenue work — but that's me inferring from a single title, not something the page spelled out. > > — RevOps Manager, Software as a Service (SaaS), 501-1000 > there was no body text explaining the problem, no named use case, and no stated audience > > — Revenue Operations Manager, B2B Technology, 1001-5000 > no body copy, no stated audience, no problem statement — so if the question is "who is this for," I genuinely can't say > > — Sales Manager, Sales and Revenue Operations, 501-1000 > Only clue was the title: "ChatGPT ou Claude substituem uma plataforma de Revenue Intelligence?" — implies it's for revenue/sales teams > > — B2B Sales Director, B2B Technology, 501-1000 > Not obvious at all — I had to infer it from a single headline: "ChatGPT ou Claude substituem uma plataforma de Revenue Intelligence?" > > — Senior Sales Manager, Software as a Service (SaaS), 1001-5000 > I had to infer "who it's for" is probably sales/RevOps leaders from the term "Revenue Intelligence" alone — it's never actually spelled out > > — B2B Sales Director, B2B Technology, 501-1000 > I'd need a line early on naming my role or workflow directly — something like "Built for RevOps teams running pipeline reviews" > > — RevOps Manager, Software as a Service (SaaS), 501-1000 > The only legible content was the headline, "ChatGPT ou Claude substituem uma plataforma de Revenue Intelligence?" — which implies this is a Revenue Intelligence tool pitching itself as better than generic LLMs for sales/revenue analysis > > — Revenue Operations Manager, B2B Technology, 1001-5000 > The only real text I got was the title, "ChatGPT ou Claude substituem uma plataforma de Revenue Intelligence?", which at least hints at a comparison between AI chat tools and revenue intelligence platforms, so I could infer this is aimed at someone like me evaluating that exact decision. But the rest of the page was just SVG code > > — Director of Sales, Sales and Revenue Operations, 1001-5000 ### No mechanism, metrics, or customer proof were available to evaluate Eight respondents found no explanation of how the product works, no features, no cost, and no ROI or before/after forecast-accuracy data. Several said this left nothing to justify taking a meeting. > the page gives me zero mechanism for how that substitution would actually work — no workflow, no proof it captures call/deal signals, nothing on data security or integration with our stack. > > — Director of Sales, Sales and Revenue Operations, 1001-5000 > the page never gave me a feature list, a workflow, or a claim to evaluate, just a rhetorical headline asking if ChatGPT/Claude can replace a Revenue Intelligence platform. There's nothing to say yes to. > > — RevOps Manager, Software as a Service (SaaS), 501-1000 > the page never told me what "worked as promised" would even mean — there's a headline asking whether ChatGPT or Claude can replace a Revenue Intelligence platform, and then nothing > > — Revenue Operations Manager, B2B Technology, 1001-5000 > A hard number showing it beats or meaningfully complements what I already run — like "catches 20% more at-risk deals than Gong/Clari" > > — Sales Manager, Sales and Revenue Operations, 501-1000 > A clear before/after on forecast accuracy or pipeline visibility — something like 'teams using X caught deal risk 3 weeks earlier than manual review' > > — Senior Sales Manager, Software as a Service (SaaS), 1001-5000 > One named account at our scale telling me they cut a specific manual reporting or pipeline-review task and can show the hours or dollars saved — that's the only thing that gets me past 'interesting idea' into 'worth a meeting.' > > — Director of Sales, Sales and Revenue Operations, 1001-5000 > There's no description of what the product actually does, no mechanism, no named customer, no numbers — just a headline asking whether ChatGPT or Claude could replace a Revenue Intelligence platform. > > — Senior Sales Manager, Software as a Service (SaaS), 1001-5000 > there's nothing here to evaluate. The page gave me one headline question (does ChatGPT/Claude replace a Revenue Intelligence platform) and no actual claim of what changes, what it costs, or what breaks if I don't switch > > — B2B Sales Director, B2B Technology, 501-1000 > I'd need at minimum a concrete "here's what it does that an LLM can't" comparison and a customer number before I'd put it in front of my team > > — RevOps Manager, Software as a Service (SaaS), 501-1000 ### Respondents could name zero differentiators against Gong, Clari, or ChatGPT Seven respondents said the page offered no comparison, feature set, or failure mode that separated it from incumbents. Two said competitor pages that simply explain how they capture call and deal data would win by default. > If a competitor's page even minimally explains how it captures call/deal signals or shows one customer result, that competitor wins by default — this one gives me literally nothing to weigh it against > > — Director of Sales, Sales and Revenue Operations, 1001-5000 > If I got a working version, I'd need it to show a concrete failure mode of doing this in ChatGPT today plus a specific thing their tool catches that I'd otherwise miss — right now it's a title with nothing behind it > > — RevOps Manager, Software as a Service (SaaS), 501-1000 > I was mentally stacking this against Gong and Clari, since those are the incumbents I already live with day to day; for this to leapfrog them on my shortlist I'd need to see a direct head-to-head > > — Director of Sales, Sales and Revenue Operations, 1001-5000 > there's literally nothing on the page but a rendered logo/graphic and one headline question. No feature list, no pricing, no case study, no number I could compare against the other options on my shortlist > > — B2B Sales Director, B2B Technology, 501-1000 > I was mentally stacking it against Gong and Clari, since those are the names that come up whenever "Revenue Intelligence" is said in my world; to pick this one over them I'd need at least one thing they can't claim > > — RevOps Manager, Software as a Service (SaaS), 501-1000 ### The broken page itself disqualified the vendor from consideration Three respondents treated the execution failure as a credibility signal, saying it removed the product from their shortlist against Gong and Clari. One read it as early-stage testing rather than an established vendor with a sales motion. > it reads like a small, early-stage team — maybe a newsletter or blog series ("Builders #02" suggests a series, not a product page) testing an idea rather than an established vendor with a sales motion. > > — Director of Sales, Sales and Revenue Operations, 1001-5000 > I can't shortlist something where I never saw a mechanism, a proof point, or even a feature list. If they fixed the render and came back with something concrete — what it ingests, how it scores calls or deals differently than a generic LLM, one named customer or benchmark — I'd reconsider > > — Revenue Operations Manager, B2B Technology, 1001-5000 > a provocative title with a broken or unrendered page behind it is a mark against them, not neutral — if I can't get the page to show me their argument on the first click, that's a data point about their execution > > — Director of Sales, Sales and Revenue Operations, 1001-5000 --- ## 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. - **This test measured a rendering failure, not messaging — the results cannot be read as a verdict on the positioning.** *(high)* Nine respondents saw raw SVG markup with only a headline and login button legible. Every downstream negative on relevance, value, and differentiation describes absent copy rather than weak copy. - **One Portuguese headline is carrying the entire page, and it is the wrong language for the buyer being tested.** *(high)* Eight respondents had to infer audience and problem from a single Portuguese headline or the page title, with no body copy naming the problem, use case, or buyer. - **The page gives a prospect no reason to take a meeting, so traffic to it is wasted spend.** *(high)* Eight respondents found no mechanism, features, cost, or ROI data, and several said explicitly that nothing justified a meeting. There is no conversion path beyond a login button. - **Incumbents win this category by default on basic explanation alone.** *(high)* Seven respondents named zero differentiators against Gong, Clari, or ChatGPT, and two said competitor pages that merely explain how they capture call and deal data would beat this one. - **The failure costs more than a lost visit — it actively brands the vendor as unfit to buy from.** *(medium)* Three respondents treated the broken execution as a credibility signal and removed the product from their shortlist; one read it as early-stage testing rather than an established vendor. - **The only positive signal is unusable evidence, since a provocative question with nothing behind it converts curiosity into distrust.** *(medium)* Two respondents called the headline compelling, and both noted nothing on the page supported the claim. Two of fifteen at 0.43 confidence is not a finding to build on. --- ## 06 · Who answered | # | Role | Industry | Company size | | --- | --- | --- | --- | | 1 | Director of Sales | Sales and Revenue Operations | 1001-5000 | | 2 | RevOps Manager | Software as a Service (SaaS) | 501-1000 | | 3 | Revenue Operations Manager | B2B Technology | 1001-5000 | | 4 | Sales Manager | Sales and Revenue Operations | 501-1000 | | 5 | Senior Sales Manager | Software as a Service (SaaS) | 1001-5000 | | 6 | B2B Sales Director | B2B Technology | 501-1000 | | 7 | Director of Sales | Sales and Revenue Operations | 1001-5000 | | 8 | RevOps Manager | Software as a Service (SaaS) | 501-1000 | | 9 | Revenue Operations Manager | B2B Technology | 1001-5000 | | 10 | Sales Manager | Sales and Revenue Operations | 501-1000 | | 11 | Senior Sales Manager | Software as a Service (SaaS) | 1001-5000 | | 12 | B2B Sales Director | B2B Technology | 501-1000 | | 13 | Director of Sales | Sales and Revenue Operations | 1001-5000 | | 14 | RevOps Manager | Software as a Service (SaaS) | 501-1000 | | 15 | Revenue Operations Manager | B2B Technology | 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 2026-09-23, then deleted along with the personas and their answers.