# Message test — https://www.zealos.io/ After reading your page, only 9 of 15 personas could name a reason to pick you over a similar option. - **Page tested:** https://www.zealos.io/ - **Audience tested against:** SMB founders and small b2b/b2c sellers & sales teams - **Personas:** 15 simulated - **Report:** https://grader.wynter.com/r/ai-sales-assistant-for-founders-sellers-zealos-kP4Kg1Q > 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 | 81% | 2 without hesitation, 13 with reservations | | 2. Relevance | Can they tell what it solves, and who it's for? | 14/14 | 98% | 13 without hesitation, 1 with reservations | | 3. Value | Do they actually want it? | 12/15 | 67% | all with reservations | | 4. Differentiation | Is there a reason to pick you over the alternatives? | 9/15 | 56% | all with reservations | **Brand alignment** (a side metric, not one of the four layers) — 11/15, 63% strength (all with reservations). Does the page read like the company you actually are? **Fix first: Differentiation.** 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 **Source or drop the tool-replacement pricing comparison.** The cost comparison against a replaced stack was the single most attacked element on the page: unsourced list prices, and a stack readers said nobody buys in full. It actively cost the page credibility. Either cite each price with a link and date and scope the comparison to the two or three tools a seller genuinely pays for, or cut the table and let the consolidation argument rest on the 'All your work in one window' section, which readers accepted on its own. *effort low · impact high · tested against Proof next to the claim* **Lead differentiation with promise-tracking from call transcripts.** The only differentiator readers articulated in their own words was that Zealos catches things you said you'd do — the Zoom transcript moment where 'I'll get you both by Thursday' becomes a tracked obligation. That mechanic currently appears last, under a truncated 'You promise something on a call' item, after generic-sounding LinkedIn and website-visitor scenarios that other AI SDR tools also claim. Move promise-tracking to the top of 'What it actually does' and give the section heading that… *effort low · impact high · tested against Front-load the meaning* **Say drafts wait for your approval, and name the integration model.** Readers who found real separation from competitors pointed to the human approval checkpoint versus rivals who only say 'AI-powered', and to MCP/API integration meaning no lock-in for people already in Slack or Claude. Neither is stated as a differentiator in the copy — 'Drafted' and 'Sent' labels in the mockup are left to imply the approval gate, and integration is only hinted at by 'brings your tools together'. Add a short line under the hero: nothing sends until you approve it, and Zealos… *effort low · impact high · tested against Give a reason to choose you* ### Value **Add named customers with role and outcome beside the automation claims.** The page asserts "12 handled today" and "11 things ran while you slept" with nobody's name attached to it. Readers treated the total absence of customer names, logos or usage evidence as a reason to read the page as pre-traction — good copywriting without validation. Put two or three attributable lines directly under the 'What it actually does' section: a named seller or founder, their company, and what changed (e.g. follow-ups sent same day instead of three days late). Evidence needs to sit… *effort medium · impact high · tested against Proof next to the claim* **State how often drafts go out unedited, with a number.** "Zealos automatically follows up" and "Tasks completed before they're even requested" are the claims readers refused to take on faith. What they wanted before spending even half an hour was output quality: how many drafted actions need editing before sending, and whether the product invents commitments that were never made. Replace or support those lines with a stated figure — e.g. the share of drafts sent without edits, and that every promise is quoted from the call transcript it came from. A… *effort medium · impact high · tested against Specifics beat superlatives* **Make the trial CTA a concrete first-week test, not "Start for free".** "Start for free · 7-day free trial · Setup in minutes" asks for time without saying what the reader will have proved by the end of it. Readers said they wanted to verify the automation on their own inbox with an approval gate before trusting it. Rewrite the CTA support line to name that test: connect your inbox and calendar, and see the promises you made last week drafted for your approval before you send anything. That converts the free trial from a generic offer into the verification step… *effort low · impact medium · tested against Tie the feature to the outcome* ### Clarity **Change "does most of them for you" to state the approval step.** The hero's 'does most of them for you' and 'Tasks completed before they're even requested' hide the one thing readers most wanted to know: whether Zealos sends on its own or drafts for review. The mockup uses both 'Drafted' and 'Sent' badges without explaining who decided. Rewrite the subheader to say it drafts every reply and follow-up in your voice and sends only what you approve, so the reader does not have to infer the automation's boundary from badge labels. *effort low · impact high · tested against Plain language* ### Brand alignment (side metric) **Pick one buyer per section: solo seller or team rollout.** The page opens 'AI Sales Assistant for Founders & Sellers' and speaks to one person — 'Good morning, Mark', 'you never opened a single tab' — then the 'All your work in one window' and task sections shift into team-wide language with colleagues blocked on you and shared pipelines. Readers noticed the persona moving between a solo rep and a team deployment depending on where they looked. Keep the individual seller as the named buyer throughout the narrative sections, and if team use matters… *effort medium · impact medium · tested against Name the audience* **Replace "Never log into Salesforce again" with a line about writing back.** 'Never log into Salesforce again' reads as a rip-and-replace promise, and it collides with the mockups that show Salesforce and HubSpot fields being filled in. Readers already unsure whether the product writes into their existing tools or replaces them will read this heading as the answer and get the wrong one. Rewrite it to say Salesforce and HubSpot stay your system of record — Zealos fills the empty next-step and decision-maker fields the mockup already shows blank. *effort low · impact medium · tested against Answer the live objection* --- ## 03 · What is working ### The hero line, subheader and mockup tell people who this is for and what it does on… Nine respondents said the header, subheader, problem statement and mockup immediately conveyed the product's function, the pain point and the intended buyer. Several stated they understood the audience and positioning without re-reading. This was the most consistently praised part of the page. > Yes, it was clear fast — the header does the work: "AI Sales Assistant for Founders & Sellers" plus "the to-do list that does itself," and the "Good morning, Mark" mock inbox with tasks like "Sent Betty the SOC 2 doc" makes the problem obvious in one screen > > — Sales Manager, E-commerce, 51-200 > the hero line "AI Sales Assistant for Founders & Sellers" plus "the first to-do list that does itself" told me the problem (dropped follow-ups, tools scattered across Gmail/Slack/Salesforce/Zoom) and the buyer in about two seconds > > — Founder & CEO, Software & SaaS, 1-10 > the header does the job in one line: "AI Sales Assistant for Founders & Sellers" plus "The first to-do list that does itself," and by the third sentence ("Zealos watches every call, inbox and message for the tasks you owe — then does most of them for you") I know exactly what problem it's targeting: scattered follow-ups across tools that fall through the cracks > > — Sales Development Representative, Software & SaaS, 1-10 > "For Founders & Sellers" tag at the very top and the whole "Chaos before / Calm after" mock-up with Betty, Kestrel, Northwind deals spells out the reader directly — this is for someone running sales personally > > — Founder & Sales Lead, Professional Services, 11-50 > the header literally says "AI Sales Assistant for Founders & Sellers" and the subhead "watches every call, inbox and message for the tasks you owe — then does most of them for you" tells you the problem (scattered follow-ups across tools falling through the cracks) and the audience straight away. > > — Sales Manager, Software & SaaS, 1-10 ### Promise-tracking and the human approval checkpoint are the differentiators respondents… Three respondents named specific mechanics as real differentiation: promise-tracking as recognising an actual failure mode, the human approval checkpoint versus competitors claiming only 'AI-powered', and MCP/API integration as avoiding vendor lock-in for existing Slack or Claude users. These were the only differentiators respondents articulated in their own words. > The "You promise something on a call" mechanic — where a spoken "I'll send you..." becomes a task that "won't tick off until the thing has gone out" — is the one specific hook that would tip me toward this over a competitor > > — Account Executive, Professional Services, 11-50 > it means I'm not locked into their UI if I already live in Slack or Claude > > — Founder & Sales Lead, E-commerce, 51-200 > "nothing sends without your yes" line next to the Dana/LinkedIn example is the specific thing that would pull me toward this one over a competitor — it shows they've thought about the actual failure mode > > — Founder & CEO, Software & SaaS, 1-10 --- ## 04 · What the personas said ### The page never says whether the product acts on its own or waits for human approval Five respondents could not tell from the copy whether the product autonomously sends follow-ups or only drafts them for review, and flagged vague action verbs that hide whether the automation is real or just reminders. Related uncertainty was raised about whether the product writes into existing tools or replaces them. One respondent read the mechanic as reading CRM and transcripts to auto-draft or execute, without… > the page never says whether it's writing into my existing CRM/inbox or replacing them, so terms like "drafted" and "logs the deal" leave me guessing what stays in Gmail/HubSpot versus what moves into Zealos itself > > — Founder & CEO, Professional Services, 11-50 > catches every commitment you made ("I'll send you the SOC 2 doc") and either drafts or actually executes the follow-up before you have to think about it > > — Founder & CEO, Software & SaaS, 1-10 > words like "watches," "handled for you," and "does most of them for you" are doing all the mechanical work without explaining it: watches how, handled via what trigger, most meaning what percentage. Those verbs are vague enough to cover either a real automation pipeline or a glorified reminder list > > — Sales Development Representative, Software & SaaS, 1-10 > Whether it actually "does" the tasks reliably or just drafts them for approval is unclear from the copy — it hedges between "does most of them for you" and "nothing sends without your yes." > > — Sales Manager, Software & SaaS, 1-10 > feels like it's trying to be both a thin layer and an all-in-one, and I'd want to see which one it really is before I pitch it internally > > — Account Executive, Professional Services, 11-50 ### There are no named customers, logos or proof points, and respondents treat that as a… Six respondents pointed to the absence of customer names, logos or evidence the product works at scale. Two said this alone blocks a purchase decision or makes them read the page as pre-traction, and one said the brand risks looking like good copywriting without validation. Respondents wanted verification via reference calls or a live inbox test with an approval gate before trusting the claims. > No named customers, no case study, just a fake "Betty" and "Northwind" demo, so I can't tell if it actually works at 50+ seats — that's what would move me. > > — Sales Manager, E-commerce, 51-200 > there's no customer name or case study anywhere, just mockups labeled "Mark" and "John," which is what you'd expect pre-traction > > — Founder & CEO, Software & SaaS, 1-10 > "obsessive/hands-on/intense" as a badge of honor is trying hard to flatter me into ignoring that none of the drafting-accuracy claims are backed by a number > > — Founder & Sales Lead, Software & SaaS, 1-10 > there's no named customer anywhere, no logo, no "here's a professional services firm your size using this," which is exactly what I'd want before ruling a competitor out in its favour > > — Founder & CEO, Professional Services, 11-50 > I've been burned by "AI does your CRM" promises before, so the category doesn't scare me off but it does make me want a real proof point, not just mockup screenshots, before I trust the "it just works" framing. > > — Sales Development Representative, E-commerce, 51-200 > A rep on a real team my size stops manually re-typing CRM fields and chasing signatures for a full week, with zero wrong drafts sent to a live customer — if I can verify that with a reference call, that's worth a meeting. > > — Sales Manager, E-commerce, 51-200 > What would move me from meeting to buyer is watching it run live on my own inbox and CRM for a week with the "nothing sends without your yes" gate on, so I can see the false-positive rate on what it thinks I owe people > > — Sales Development Representative, Software & SaaS, 1-10 ### Respondents want evidence on output quality before spending time on the product Two respondents said they needed proof of how many drafted actions require editing before sending, and proof the product does not hallucinate commitments, before committing even 30 minutes. Both framed this as the precondition for believing the automation claim. > how many of those drafted actions actually need editing before they go out, because if it's most of them, I've just added a review step, not removed one > > — Founder & Sales Lead, E-commerce, 51-200 > I'd want a live demo on my own Salesforce/Zoom data, not the mockup, before I give this 30 minutes > > — Founder & CEO, Software & SaaS, 1-10 ### The pricing comparison reads as a strawman because it uses unsourced list prices and… Four respondents attacked the tool-replacement cost comparison directly: the competitor pricing is unsourced, it uses list prices, and it assumes a stack no one actually buys in full. They described this as misleading and as undermining the page's credibility. One respondent, by contrast, read the comparison table as a concrete differentiator and one accepted the multi-tool consolidation savings. > those competitor per-seat prices have no source, and stacking unsourced numbers like that just makes me trust the rest of the page less > > — Sales Manager, E-commerce, 51-200 > the $415-vs-$49 comparison is the thing that would make me rule it out if a competitor showed real usage data instead of a hypothetical stacked tool list — right now it's their assumption > > — Founder & Sales Lead, E-commerce, 51-200 > laying out Superhuman, DocSend, Otter, Calendly etc line by line at $415/user/month versus $49 is a concrete, checkable claim, not just marketing fluff > > — Founder & CEO, Professional Services, 11-50 ### The page speaks to solo founders and individual sellers, not to managers or enterprise… Six respondents said the tone and positioning read as founder-led seed/Series A and aimed at solo operators, not enterprise buyers with RevOps, IT approval or team and finance justification. One respondent noted the buyer persona shifts between solo rep and team-wide rollout depending on the section. Two framed this as appropriate fit rather than a flaw. > the problem is crystal clear, but the buyer persona shifts depending on which section you're reading > > — Account Executive, Professional Services, 11-50 > They're clearly selling to other founders and small sales teams, not to a company my size — the whole "Founders & Sellers" framing, the Betty/Northwind/Kestrel mockups, and the per-seat pricing table calibrated against tools like Superhuman and Calendly all read like they're built for a 5-20 person team > > — Sales Development Representative, E-commerce, 51-200 > the "zeal·os" manifesto page with the "they'll call you obsessive... good" copy reads like founders writing about themselves, not a mature enterprise vendor with a brand book. > > — Founder & CEO, E-commerce, 51-200 > the "unreasonable commitment" branding is pitched more at solo founders wearing the sales hat themselves than at someone managing a team, so it's aimed slightly to my left rather than dead at me. > > — Sales Manager, Professional Services, 11-50 > the "you" in the copy is a solo operator drowning in their own inbox, not a manager who has to justify a switch to a team and a finance line, so it half-lands > > — Sales Development Representative, Professional Services, 11-50 > the tone — a bit evangelical, a bit "us against the tool-bloat world" — is pitched more at a solo seller or founder-who-also-sells than at an AE who has to build a business case and get IT/RevOps sign-off > > — Account Executive, E-commerce, 51-200 > at our size (11-50 people, still fairly hands-on) matters more than it would at a bigger firm with dedicated ops > > — Founder & CEO, Professional Services, 11-50 --- ## 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's one strength is cosmetic: people grasp the pitch instantly and then cannot verify a single claim in it.** *(high)* Nine respondents said the hero, subheader and mockup conveyed function, pain and buyer without re-reading (theme 5), yet six flagged the total absence of named customers, logos or scale evidence (theme 1), with two saying that alone blocks a purchase and one calling the brand good copywriting without validation. Comprehension without proof produces an audience that understands the offer and still cannot act. - **The core automation claim is unproven and unspecified at the same time, so the central promise of the page is unbuyable.** *(high)* Five respondents could not tell whether the product acts autonomously or only drafts for review, and flagged vague action verbs hiding whether automation is real or just reminders (theme 0). Two more demanded proof of how many drafted actions need editing and that it does not hallucinate commitments before giving even 30 minutes (theme 2). The mechanic is neither described nor evidenced. - **The page destroys its own credibility with the pricing comparison, and that damage spreads to claims respondents had no other way to check.** *(high)* Four respondents attacked the cost comparison as unsourced, list-price-based and assuming a stack nobody buys, calling it misleading and credibility-undermining (theme 3). With six already noting no logos or proof points (theme 1), the one quantified argument on the page is the one respondents distrust most. Only one read the table as a concrete differentiator. - **The differentiator respondents value most is the one the copy fails to state clearly, so the page's best asset is invisible.** *(high)* Three respondents named the human approval checkpoint as real differentiation versus competitors claiming only 'AI-powered' (theme 6). But five respondents could not tell from the copy whether the product acts on its own or waits for human approval (theme 0). The page buries the exact mechanic that would separate it from the field. - **The page cannot survive contact with a buying committee: it addresses one operator while the deal requires several.** *(high)* Seven respondents read the tone and positioning as founder-led seed/Series A aimed at solo operators, not enterprise buyers with RevOps, IT approval or team and finance justification, and one noted the persona shifts between solo rep and team-wide rollout by section (theme 4). Combined with no logos or scale evidence (theme 1) and no sourced cost case (theme 3), there is nothing for anyone to take upward. - **Only three of fifteen respondents could name a differentiator in their own words, which means the page reads as category description rather than a reason to switch.** *(medium)* Theme 6 records three respondents naming promise-tracking, the approval checkpoint and MCP/API integration as the only differentiators articulated unprompted. Against that, four attacked the explicit differentiation attempt as a strawman (theme 3). The page's stated differentiation is rejected and its real differentiation is under-registered. --- ## 06 · Who answered | # | Role | Industry | Company size | | --- | --- | --- | --- | | 1 | Founder & CEO | Professional Services | 11-50 | | 2 | Founder & Sales Lead | E-commerce | 51-200 | | 3 | Sales Development Representative | Software & SaaS | 1-10 | | 4 | Account Executive | Professional Services | 11-50 | | 5 | Sales Manager | E-commerce | 51-200 | | 6 | Founder & CEO | Software & SaaS | 1-10 | | 7 | Founder & Sales Lead | Professional Services | 11-50 | | 8 | Sales Development Representative | E-commerce | 51-200 | | 9 | Account Executive | Software & SaaS | 1-10 | | 10 | Sales Manager | Professional Services | 11-50 | | 11 | Founder & CEO | E-commerce | 51-200 | | 12 | Founder & Sales Lead | Software & SaaS | 1-10 | | 13 | Sales Development Representative | Professional Services | 11-50 | | 14 | Account Executive | E-commerce | 51-200 | | 15 | Sales Manager | Software & SaaS | 1-10 | --- ## 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 2026-08-25, then deleted along with the personas and their answers.