# Message test — https://vervoe.com/ After reading your page, only 7 of 15 personas could name a reason to pick you over a similar option. - **Page tested:** https://vervoe.com/ - **Audience tested against:** SMB SaaS Talent Acquisition Leaders, HR Leaders, People Operations Leaders who own recruitment and talent acquisition - **Personas:** 15 simulated - **Report:** https://grader.wynter.com/r/vervoe-ai-powered-skills-assessments-Xb23z2o > 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 | 82% | 3 without hesitation, 12 with reservations | | 2. Relevance | Can they tell what it solves, and who it's for? | 13/15 | 70% | all with reservations | | 3. Value | Do they actually want it? | 11/15 | 63% | all with reservations | | 4. Differentiation | Is there a reason to pick you over the alternatives? | 7/15 | 50% | all with reservations | **Brand alignment** (a side metric, not one of the four layers) — 4/15, 37% 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 **Add a mid-market reference customer beside the results strip.** Every name in the proof strip — Australia Post, BOQ Group, dentsu, iSelect — is far larger than the companies reading, so the numbers read as irrelevant rather than impressive. Add at least one result from a company in the 51–200 range with headcount named ('120-person…'), placed in the same 'You didn't just hear it from us' block so it is seen with the others. *effort medium · impact high · tested against Proof next to the claim* **Replace "the only platform dedicated to" with a testable difference.** 'Vervoe is the only platform dedicated to testing real job skills in the context of your role and company' is the page's sole reason-to-choose and it is an unverifiable superlative that a competitor could rewrite verbatim. Swap it for the concrete thing rivals cannot copy: simulations built per role and company rather than generic test libraries, open-ended tasks like spreadsheets and presentations auto-scored (not multiple choice), and independent bias audit by Holistic AI. State how many… *effort low · impact high · tested against Specifics beat superlatives* **Show a staffing or high-volume screening use case.** The proof is uniformly internal enterprise hiring, so agencies and high-volume recruiters see nothing that resembles their work. The 'Interactive Experiences' section already claims 'high-volume screening' — attach an actual scenario to it: applicants screened per requisition, time from application to ranked shortlist, and a named customer doing volume placement rather than corporate hiring. *effort medium · impact medium · tested against Name the audience* ### Value **Add baselines to the customer results strip.** '3× fewer interviews per hire', '2,340 hours saved annually' and '75% reduction in employee attrition' arrive with no before figure, no role type and no measurement window, so they are read as marketing rather than evidence. Give each a baseline and period — from what to what, over how long, for which roles and volume — either inline or in a one-line footnote beneath the strip. *effort medium · impact high · tested against Proof next to the claim* **Publish validity evidence behind the Assessment Validity claim.** 'developed and validated by experts, tested for accuracy' is the page's answer to whether the scores predict job performance, and it says nothing measurable. Buyers asked for predictive validity before crediting the outcome claims. State the correlation between assessment score and post-hire performance rating, the sample size, and link the technical documentation next to that paragraph. *effort medium · impact medium · tested against Proof next to the claim* ### Clarity **Explain how AI grading scores video and presentation answers.** Under 'Efficient Hiring', the line 'our platform simulates day-to-day work and scores candidates instantly' is the page's biggest unanswered question — readers wanted the mechanics of grading open-ended formats, and the FAQ answers with metaphor rather than method. Add two or three sentences of method next to that claim: what the model is scored against (role-specific rubrics/answer keys built by IO psychologists), whether humans review or override, and how scores map to a ranked shortlist… *effort medium · impact high · tested against Proof next to the claim* **Replace the hero line with what the product actually does.** "Better talent starts with skills" plus "The most trusted skills-based AI platform for hiring" tells a reader nothing about the mechanic, and the purpose only becomes clear several screens down. The clearest reading anyone reached was 'replace resume screening with auto-scored job simulations' — write that in the hero: e.g. 'Replace resume screening with job simulations your candidates complete and AI scores.' Keep the category word ('skills assessments') in the subhead, not the headline. *effort low · impact high · tested against Lead with the use case* **State ATS integration by name in the Efficient Hiring section.** 'Automate hiring workflows, save hours' and '2,340 hours saved annually' rest on the assessment fitting into an existing pipeline, but the page never mentions the ATS. Readers said that without it the time-to-hire argument does not hold. Name the integrations (Greenhouse, Workday, SmartRecruiters, Lever) and state what passes through — invite candidates from the ATS, scores and ranked shortlist written back to the candidate record. *effort low · impact high · tested against Answer the live objection* ### Brand alignment (side metric) **Name the company size the product is built for.** The tone claims enterprise ('Enterprise-grade data protection', eight G2 Enterprise badges) while the register elsewhere addresses a generalist HR buyer, and readers found the two mismatched. Resolve it by saying who this is for near the hero — e.g. 'Built for talent teams hiring at volume, from 50-person scaleups to 10,000-person enterprises' — so the enterprise logos read as range rather than as an entry requirement. *effort low · impact high · tested against Name the audience* **Give the AI audit a link and a published result.** 'Our platforms AI is independently audited for fairness and bias by Holistic AI. With results openly reported, every decision is explainable' is the strongest credibility asset on the page but is stated and left unsupported — and 'every decision is explainable' is a large claim with nothing behind it. Link the audit report, date it, and say what explainability the buyer sees: which sub-skills produced a score and where a candidate lost points. Also fix the typo 'platforms'. *effort low · impact high · tested against Proof next to the claim* **Rewrite "Never make another bad hire" in the closing CTA.** 'Never make another bad hire with AI-powered skills assessments' is an absolute promise no assessment vendor can keep, and it sits directly above the demo button, so it is the last thing a technical evaluator reads. Replace it with something the page can support — e.g. 'See how a role-specific simulation is built and scored, in 20 minutes' — which also makes the single next action concrete. *effort low · impact medium · tested against Specifics beat superlatives* --- ## 03 · What is working ### The core mechanic — replacing resume screening with auto-scored job simulations — is… One respondent could restate the product plainly as replacing resume screening with skills-based job-simulation assessments, and another said auto-scoring of skills tasks directly addresses their screening bottleneck. One also read the Australia Post and BOQ Group proof points as clear evidence of an enterprise-ready tool. > candidates do job-simulation tasks (coding challenges, spreadsheet tests, video responses) and AI grades and ranks them, replacing resume screening > > — Head of People Operations, Software as a Service (SaaS), 11-50 > screening moves from resume/keyword triage to candidates doing actual role tasks that get auto-scored — the "instantly" scoring on spreadsheet tasks, coding challenges, presentations and video responses is the mechanism that would directly hit my stated problem > > — Talent Acquisition Leader, Staffing and Recruiting, 201-500 > The names that stuck with me were Australia Post and BOQ Group citing "2,340 hours saved annually," which is the kind of concrete detail that makes me think it's a real enterprise tool, not vaporware > > — Director of Talent Acquisition, Software as a Service (SaaS), 51-200 --- ## 04 · What the personas said ### How the AI actually scores candidates is never explained Three respondents said the mechanics of AI grading — specifically how video and presentation responses are scored — are obscured by marketing language rather than explained. One added that the FAQ answers with generic metaphors instead of methodology. > I can't tell is the actual scoring mechanism — how does the AI grade a video response or presentation objectively, what's the model, and where's the accuracy/validity data behind "assessments are developed and validated by experts"? > > — HR Leader, Staffing and Recruiting, 501-1000 > Phrases like "immersive question types," "conversational AI makes high-volume screening efficient," and "job fit assessment is akin to peering through a window" — that's marketing fluff dressed up as explanation, and it dodges the actual mechanics of how the AI grades or what "immersive" means in practice. > > — Senior Talent Acquisition Manager, Staffing and Recruiting, 11-50 > Phrases like "immersive question types," "surfaces your top candidates," and "conversational AI makes screening efficient" are the vague marketing-speak that slowed me down — they sound nice but don't tell me the integration mechanics > > — Director of Talent Acquisition, Software as a Service (SaaS), 51-200 ### ATS integration is unclear, which breaks the time-to-hire argument Two respondents said integration with existing ATS workflow is left unclear and untested, and one stated directly that without it the skills assessment does not actually address time-to-hire. The gap undercuts the page's central operational claim. > What I can't yet tell is how it actually integrates with our existing ATS workflow > > — Talent Acquisition Leader, Staffing and Recruiting, 201-500 ### The hero line is too abstract to convey what the product does One respondent said the hero line is abstract enough that the product's purpose is unclear without scrolling. This compounds the broader complaint that vague marketing phrases stand in for technical specifics. > the hero line "Better talent starts with skills" is too abstract to tell me what's actually being sold, and I had to scroll down to "Skills Validation" and "Efficient Hiring" sections before I understood > > — Director of Talent Acquisition, Software as a Service (SaaS), 51-200 ### The enterprise customer logos read as proof the product is not built for mid-market or… Eight respondents read the named customers — Australia Post, dentsu, BOQ Group — plus the overall tone as signals that the page is addressed to enterprise TA directors and HR generalists, not to 51-200 person companies, staffing agencies, or smaller recruiting shops. Several stated the page never explicitly names a target company size at all. > it reads like it's built for enterprise HR/TA teams with real hiring volume — logos like Australia Post, dentsu, BOQ Group — not obviously a scrappy 11-50 person SaaS company. Nothing on the page said "for smaller teams" > > — Head of People Operations, Software as a Service (SaaS), 11-50 > the page is written more for enterprise buyers than an 11-50 person agency — the logos (Australia Post, Lumen, Tennis Australia) and stats (2,340 hours at BOQ Group, a bank) all read big-company, not boutique staffing firm. > > — Senior Talent Acquisition Manager, Staffing and Recruiting, 11-50 > It doesn't feel written for someone like me at a 51-200 person recruiting shop — there's no mention of agency or staffing use cases anywhere > > — Head of Recruitment, Staffing and Recruiting, 51-200 > the customer logos (Australia Post, dentsu, BOQ, Tennis Australia) suggest big enterprise, and the "2,340 hours saved annually" stat sounds like a large operation with high-volume hiring, not obviously a 51-200 person HR company like mine > > — Director of People Operations, Human Resources, 51-200 > Who they normally sell to is obvious from the tone and proof points: enterprise TA teams with the budget and headcount to run "Enterprise" G2 tiers and handle compliance reviews > > — Director of People Operations, Staffing and Recruiting, 11-50 > feels like it's aimed at a broader HR generalist audience rather than someone with my narrow mandate > > — Head of Recruitment, Human Resources, 201-500 > nothing on the page explicitly says "built for mid-market," so that's a gap I'd need a sales rep to close > > — Director of Talent Acquisition, Software as a Service (SaaS), 51-200 > The tone isn't written for someone like me. It's written for a TA director at a company doing thousands of hires a year, not a 51-200 person HR firm — there's nothing here that says "we work at your scale," > > — Director of People Operations, Human Resources, 51-200 > Who it's for is less explicit; there's no "for TA leaders at 200-2000 employee firms" segmentation > > — Talent Acquisition Leader, Staffing and Recruiting, 201-500 ### The stated results are given without baselines or methodology, so respondents treat them… Four respondents flagged that the fewer-interview-rounds and case study statistics arrive with no baseline, no role-type breakdown, and no methodology, and one called them outright unsupported. Two asked for before/after numbers or predictive validity data before they would credit the claims. > fewer interviews compared to what baseline, over what role types, and does that saving survive once you add in the time candidates spend on these job-simulation tasks upfront? > > — Head of Recruitment, Human Resources, 201-500 > if the other tool on my shortlist has a published case study breaking down time-to-hire specifically, with before/after numbers and company size similar to mine > > — Head of Recruitment, Human Resources, 201-500 > those look like they're doing the heavy lifting for the pitch but I'd need methodology, not just a company name attached to a stat > > — Talent Acquisition Leader, Staffing and Recruiting, 201-500 > the FAQ section drifts into generic explainer copy ("a job fit assessment is akin to peering through a window") that reads like SEO filler aimed at someone earlier in their research than I am > > — Talent Acquisition Leader, Staffing and Recruiting, 201-500 ### The proof points are all much larger companies, so there is no reference customer… Four respondents said the statistics and case studies lose credibility because every referenced customer is far larger than they are, and asked for a same-size or mid-market reference. Two others noted the proof is all internal enterprise hiring rather than staffing or agency placement. > I'd need one case study from an 11-50 person company to actually believe the setup and cost work for us. If a competitor on the shortlist had a same-size customer story with a real number attached, that would beat this page > > — Head of People Operations, Software as a Service (SaaS), 11-50 > I'd need a case study from a staffing/recruiting firm placing candidates at volume for external clients — something showing throughput on hundreds of reqs a week, agency-side workflows, and how scores get communicated back to clients, not just an internal enterprise hiring its own headcount. > > — HR Leader, Staffing and Recruiting, 501-1000 > every named proof point — Australia Post, BOQ Group's "2,340 hours saved," dentsu's "75% reduction in attrition" — is a much bigger company than mine, and the G2 badges are all "Enterprise" tier > > — Director of Talent Acquisition, Software as a Service (SaaS), 51-200 ### The page reads as a mid-market competitor without the enterprise-grade proof its tone… One respondent said the positioning lands as a mid-market competitor while lacking enterprise-grade proof points, and another said the tone aims at a generalist HR buyer rather than a technical evaluator. The page's register and its evidence are read as mismatched. > The tone itself is written for a generalist HR buyer, not for someone like me — it's smooth, benefit-led copy ("Never make another bad hire") that answers the "why" fast but goes soft exactly where I'd push, like "assessments are developed and validated by experts" with zero methodology attached. > > — HR Leader, Staffing and Recruiting, 501-1000 > still feels like a company selling into mid-market/enterprise without quite having the enterprise-grade proof (predictive validity data) that would let me close the deal against Modern Hire or HireVue > > — Senior Talent Acquisition Manager, Software as a Service (SaaS), 501-1000 --- ## 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 disqualifies itself from every account that is not enterprise, and it does so with its own proof assets.** *(high)* Eight of 15 respondents read the named customers — Australia Post, dentsu, BOQ Group — and the overall tone as addressed to enterprise TA directors, not to 51-200 person companies, staffing agencies, or smaller shops (theme 3). Four more said the statistics and case studies lose credibility because every referenced customer is far larger than they are (theme 5). The logos are not neutral social proof; they are actively read as an exclusion notice, and no theme shows a single respondent outside… - **The page never states who it is for, so readers assign it a target audience and then rule themselves out.** *(high)* Several of the eight respondents in theme 3 stated the page never explicitly names a target company size at all. In that vacuum, the logos and register do the targeting: two respondents read the tone as aimed at a generalist HR buyer or a mid-market competitor without matching proof (theme 6). Leaving segment unstated does not broaden reach — it hands the decision to the strongest visual cue on the page. - **Every quantitative claim on the page is currently unusable as evidence.** *(high)* Four respondents flagged that the fewer-interview-rounds and case study statistics arrive with no baseline, no role-type breakdown and no methodology, one called them outright unsupported, and two asked for before/after or predictive-validity data before crediting them (theme 4). Four more discount the same numbers because the customers behind them are far larger (theme 5). The numbers are simultaneously unverified and un-analogous — there is no reading in which they persuade. - **An AI scoring product refuses to explain its scoring, and the FAQ makes it worse.** *(high)* Three respondents said the mechanics of how video and presentation responses are graded are obscured by marketing language rather than explained, and one noted the FAQ answers with generic metaphors instead of methodology (theme 0). The FAQ is where a skeptical reader goes for methodology; finding metaphors there converts an open question into an assumption that there is nothing to show. - **The central operational promise — faster time-to-hire — is asserted and then structurally undercut.** *(high)* Two respondents said ATS integration is left unclear and untested, and one stated directly that without it the skills assessment does not address time-to-hire at all (theme 1). Combined with results reported without baselines or methodology (theme 4), the speed claim has neither a mechanism nor a measurement behind it. - **The page loses readers before the evidence has a chance to work.** *(medium)* One respondent said the hero line is abstract enough that the product's purpose is unclear without scrolling (theme 2), and three said vague marketing phrases stand in for technical specifics throughout (theme 0). Everything the page needs to prove sits below a fold that readers have no stated reason to cross. --- ## 06 · Who answered | # | Role | Industry | Company size | | --- | --- | --- | --- | | 1 | Senior Talent Acquisition Manager | Staffing and Recruiting | 11-50 | | 2 | Director of Talent Acquisition | Software as a Service (SaaS) | 51-200 | | 3 | Head of Recruitment | Human Resources | 201-500 | | 4 | HR Leader | Staffing and Recruiting | 501-1000 | | 5 | Head of People Operations | Software as a Service (SaaS) | 11-50 | | 6 | Director of People Operations | Human Resources | 51-200 | | 7 | Talent Acquisition Leader | Staffing and Recruiting | 201-500 | | 8 | Senior Talent Acquisition Manager | Software as a Service (SaaS) | 501-1000 | | 9 | Director of Talent Acquisition | Human Resources | 11-50 | | 10 | Head of Recruitment | Staffing and Recruiting | 51-200 | | 11 | HR Leader | Software as a Service (SaaS) | 201-500 | | 12 | Head of People Operations | Human Resources | 501-1000 | | 13 | Director of People Operations | Staffing and Recruiting | 11-50 | | 14 | Talent Acquisition Leader | Software as a Service (SaaS) | 51-200 | | 15 | Senior Talent Acquisition Manager | Human Resources | 201-500 | --- ## 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 2026-08-20, then deleted along with the personas and their answers.