# Message test — https://spd.tech/ai-studio/ After reading your page, only 4 of 15 personas could name a reason to pick you over a similar option. - **Page tested:** https://spd.tech/ai-studio/ - **Audience tested against:** tech and non-tech senior decision makers at product-led companies, scaleups and enterprises looking to build an AI-powered product or build AI-native engineering discipline into their in-house tech team - **Personas:** 15 simulated - **Report:** https://grader.wynter.com/r/ai-studio-by-spd-technology-from-first-idea-to-fQTWU48 > 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? | 5/15 | 79% | 1 without hesitation, 14 with reservations | | 2. Relevance | Can they tell what it solves, and who it's for? | 15/15 | 84% | 4 without hesitation, 11 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? | 4/15 | 37% | all with reservations | **Brand alignment** (a side metric, not one of the four layers) — 7/15, 48% strength (all with reservations). Does the page read like the company you actually are? **Fix first: Clarity.** Earliest failing layer, walking the sequence in order — not simply the lowest score. ### What they thought you sell 3 of the personas who named a category got it wrong: - 1× “AI code audit / software engineering consultancy” - 1× “AI codebase audit / architecture consultancy” - 1× “AI codebase audit/remediation consultancy” --- ## 02 · What to change, layer by layer Ordered worst-first. Specific edits, not a restatement of the score. ### Differentiation **Expand "We Transform, Not Rebuild" into a headline claim with a proof line beneath.** Keeping the existing codebase is the one promise readers repeat back, but it sits as item three in a list of six. Move it up and attach a named engagement showing the repo that was retained. *effort medium · impact high · tested against Give a reason to choose you* **Add a deliverables list under "Get an Expert Architecture Review" naming the artifacts a client receives.** Buyers comparing vendors want to see the audit trail: what report, what findings format, what remediation plan. Name the documents and the turnaround. *effort medium · impact high · tested against Proof next to the claim* **Replace "Senior Engineering Throughout" with a line stating who staffs the engagement.** Every agency claims senior people. Give years of experience, the ratio of seniors to juniors, and whether the same engineers stay through the project. *effort low · impact medium · tested against Specifics beat superlatives* ### Clarity **Replace the six benefit labels under "Why Founders and CTOs Work With" with outcome sentences.** Labels like "Startup Speed + Enterprise Discipline" and "Structured Paths, Clear Outcomes" could sit on any dev shop's page. Say what each one lets the buyer do, such as keeping the existing repo and shipping without pausing the in-house team. *effort medium · impact high · tested against Concrete over abstract* **Define "AI-Native Studio" in one sentence directly under the H1.** The term repeats across headings without ever saying what the team does day to day. Write a plain line such as: senior engineers audit, harden and scale codebases built with AI coding tools. *effort low · impact high · tested against Plain language* **Add engagement and pricing structure under the three stage tracks.** Readers cannot tell what a Vibe-to-Scale engagement costs, how many engineers show up, or how it is billed. State team size, fee model and what the 90 days include. *effort medium · impact medium · tested against Answer the live objection* ### Value **Label each case study tile with the starting condition and the work performed.** Tiles reading "70%" and "10x Less" do not say what was measured, over what baseline, or whether the engagement was a rescue. Add a line per tile naming the client situation, the intervention and the timeframe. *effort medium · impact high · tested against Specifics beat superlatives* **Add source and date next to each statistic in the six-item risk list.** Numbers like "242.7% Surge in Production Incidents" and "8,000+ Impending Rebuilds" appear with no study, sample or year, and readers discount everything near them. Put the publisher and year inline under each figure. *effort low · impact high · tested against Proof next to the claim* ### Brand alignment (side metric) **Cut generic services language from "Success Stories with Global Impact" and the service tag list.** Phrases like "transformative software solutions" and the API development / Mobile App Development tag row make the page read as a standard outsourcing shop. Replace with the specific work: auditing and hardening AI-generated codebases. *effort low · impact high · tested against Concrete over abstract* **Pick one reader for the section intros and rewrite founder-pitch lines to match.** The page swings between CTO due-diligence language and founder coaching like "staring at a blank page". Choose the technical buyer and keep the register consistent through every section opener. *effort medium · impact medium · tested against Name the audience* **Replace the PitchBook quote with a testimonial about AI-generated code work.** A ten-year outsourcing relationship with "there is nothing they can't do" reinforces the legacy dev-shop read. Use a client whose vibe-coded product was audited and scaled. *effort medium · impact medium · tested against Proof next to the claim* --- ## 03 · What is working ### The problem statement and audience segmentation let readers self-select immediately Six respondents said the problem and target audience were explicit and understood fast, crediting named tools, buyer personas, and the three maturity-stage tracks for signalling who the page is for. > The "Where Are You in the Journey?" section with the three tracks — Zero-to-One, Vibe-to-Scale, MVP-to-Platform — and lines like "running a product whose foundation can't keep up with its own traction" > > — VP of Engineering, Software Development, 5000+ > I use Bolt, so I saw myself in that sentence within the first screen, no hunting required > > — Chief Technology Officer, Technology Services, 201-500 > the "Where Are You in the Journey?" section with the three named stages (Zero-to-One, Vibe-to-Scale, MVP-to-Platform) told me immediately this is about AI-generated code that's shaky or unscalable > > — VP of Engineering, AI/Machine Learning, 5000+ > The Problem with AI-Generated Products at Scale" and the line about AI-generated code lacking "governance, architectural discipline, and production-grade rigor" told me the problem within the first screen > > — Chief Product Officer, Technology Services, 11-50 ### 'Retain, don't rebuild' is the one claim respondents could repeat back Three respondents named the promise to keep the existing codebase as a concrete differentiator against competitors. Two of them qualified it as differentiating only if backed by case study proof, which the page does not supply. > most clients retain the majority of their existing codebase on a stronger production-grade architecture" — that's a concrete, checkable promise > > — VP of Engineering, Software Development, 5000+ > "Most clients retain the majority of their existing codebase on a stronger production-grade architecture" — that's a concrete, checkable promise about how they work (audit/harden, not rip-and-replace), which matters because I don't want a rebuild that resets our roadmap. But it's still just a sentence with no case study attached to it directly > > — Chief Product Officer, SaaS, 11-50 > "Most clients retain the majority of their existing codebase on a stronger production-grade architecture" line in the FAQ - if true, that's a real differentiator > > — CTO, Software Development, 501-1000 --- ## 04 · What the personas said ### Marketing vocabulary is repeated without definition and obscures the offering Three respondents said key terms recur undefined and that layered marketing language hides basic services work. One added that pricing and engagement structure are never explained. > "Studio," "AI-Native," and "governance" — all three get used repeatedly without a definition > > — VP of Engineering, Software Development, 5000+ > "value-based, not body-based" sounds nice but doesn't say whether I'm being quoted fixed price, milestone billing, or something else > > — Chief Technology Officer, Technology Services, 201-500 > It's the constant relabeling — 'AI-Native Studio,' 'value-based outcomes,' 'architecture governance,' 'senior technical co-founder' — none of those are load-bearing terms with a fixed meaning, they're marketing wrappers stacked on top > > — Head of Engineering, SaaS, 1001-5000 ### The case studies do not show the work the page says the company does Nine respondents said the named case studies — gift-advisor search, document processing AI/ML — demonstrate product outcomes, not vibe-code rescues or governance retrofits. They also lacked denominators, cost breakdowns, success rates, and timeline data. > no before/after architecture diagram, no named client saying "we kept 80% of our code," just vague stats like "45% Vulnerability Rate" with no source. Not worth a meeting off this alone > > — Founder, AI/Machine Learning, 51-200 > the case studies show conversion lifts and cost savings on client products, not a governance retrofit on an existing Replit-based codebase at our scale > > — CTO, Software Development, 501-1000 > The case studies listed (gift-advisor search, document processing AI/ML) don't obviously map to "rescued a vibe-coded app," > > — Founder, AI/Machine Learning, 51-200 > PitchBook and Mogami are named for the quotes but not tied to any of those specific metrics, so I can't tell if the 12.5% conversion lift is the same client as the CPO testimonial or a different, unverifiable one > > — Chief Technology Officer, Technology Services, 201-500 > no named client at our size, no detail on what "AI coding guardrails" actually means day-to-day, and the case studies (12.5% conversion lift, 1M+ users) read as generic dev-shop wins > > — VP of Engineering, AI/Machine Learning, 5000+ > But the case studies are thin: "+12.5%" and "10x less" aren't tied to enough detail for me to know if that's a Bolt-to-production story like ours or something else entirely. > > — Founder, Software Development, 51-200 ### The statistics are unsourced, and respondents stopped trusting them Eight respondents flagged that cited statistics carry no source, methodology, or evidence. They read the absence as a rigor problem that undermined the surrounding claims rather than as a minor omission. > "1.7x Higher Issue Rate," "242.7% Surge in Production Incidents," "45% Vulnerability Rate" — have no source, no methodology, no baseline, so I can't tell if they're industry data or something SPD made up to scare me > > — Chief Product Officer, SaaS, 11-50 > stats like "1.7x higher issue rate" or "242.7% surge in production incidents" have zero sourcing, so I'd want the methodology behind those > > — VP of Engineering, Software Development, 5000+ > the stat-dump (1.7x, 43%, 242.7%) without sourcing made me a bit skeptical of how rigorously the "problem" was actually measured > > — VP of Engineering, AI/Machine Learning, 5000+ > the numbers cited (1.7x issue rate, 242.7% surge in incidents) are unsourced scare stats, not proof of their own delivery > > — CTO, Software Development, 501-1000 > 1.7x issue rate, 242.7% surge in incidents) since they're presented with no citation, which is the one place the page loses rigor > > — Chief Product Officer, Technology Services, 11-50 ### Competitors are seen as winning on audit-trail evidence the page never shows One respondent said competitors win by showing concrete audit trails, a form of proof absent here. > To pick SPD over them I'd need to see the same kind of concrete audit trail here, not just the claim - right now the competitor wins purely because they show their work > > — CTO, Software Development, 501-1000 ### The page reads as a legacy dev shop relabelled as AI-native Six respondents concluded the service menu reveals standard outsourcing or decade-old dev-shop work under AI-Native Studio branding. Several described it as ordinary audit-and-rebuild engineering rather than anything proprietary. > closer to a specialized dev shop or fractional CTO practice than a "studio" or platform, despite the branding > > — VP of Engineering, Software Development, 5000+ > mid-size dev shop (50-200 people, probably 8-10+ years old going by the PitchBook quote saying "over the last 10 years") that's pivoted its pitch to ride the AI-coding wave > > — Founder, AI/Machine Learning, 51-200 > you can see the seams where old positioning ("comprehensive alignment between emerging technologies and established business processes") bumps into new positioning ("Vibe-to-Scale," "AI-Native Studio"). > > — Chief Product Officer, SaaS, 11-50 > strip away the branding and it's staff-aug/dev-shop work with a senior architect wrapper, billed as "value-based" instead of hourly > > — Head of Engineering, SaaS, 1001-5000 ### The voice switches between CTO buyer and founder coaching Three respondents found the tone inconsistent, shifting from technical due diligence and enterprise procurement language into founder-facing or generic pitch-deck register. > it's founder-coaching language ("acts as your AI-powered technical co-founder," "Where Are You in the Journey?") rather than enterprise procurement language > > — VP of Engineering, AI/Machine Learning, 5000+ > The tone does feel like it was written for someone in my seat in places — naming Cursor, Lovable, Bolt, Replit specifically, and lines like "no disruption to your team" and "no pause in delivery" are clearly aimed at someone worried about workload and control, not a founder just wanting a demo. But then it swings into generic case-study filler ("1M+ Users," "+40% Up") that reads like it was written for a pitch deck, not for me > > — Head of Engineering, Technology Services, 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. - **The page's evidence layer collapses: every proof element offered is either off-topic or unsourced.** *(high)* Seven respondents said the case studies demonstrate product outcomes rather than vibe-code rescues or governance retrofits, and six said statistics carry no source or methodology. Nothing on the page survives as usable proof. - **The one differentiator respondents retained is unusable, because the proof that would validate it is exactly what's missing.** *(high)* Three respondents named 'retain, don't rebuild' as concrete, but two conditioned it on case study proof — and seven said the case studies show unrelated product work. The page's strongest claim self-cancels. - **Readers self-select in, then find nothing that earns their attention — the page converts interest into disqualification.** *(high)* Six respondents understood the problem and audience immediately, yet four concluded the service menu is a relabelled dev shop and seven found case studies off-topic. Fast comprehension only accelerates the negative verdict. - **The AI-Native positioning is read as a rebrand, not a capability, and the page supplies nothing to reverse that reading.** *(high)* Four respondents identified standard outsourcing or decade-old dev-shop work under the AI-Native Studio label, describing ordinary audit-and-rebuild engineering. Undefined marketing vocabulary flagged by three more reinforces the sense of language… - **Missing sourcing is not a tidiness issue — it actively contaminates claims that were otherwise fine.** *(high)* Six respondents read the absence of methodology as a rigor problem that undermined surrounding claims rather than a minor omission. Unsourced numbers cost the page more credibility than omitting them would have. - **The page cannot survive a side-by-side comparison because it never enters the category of evidence buyers use to decide.** *(medium)* A respondent said competitors win by showing concrete audit trails, and six others flagged that no statistic carries evidence while seven found the case studies irrelevant to the stated work. The comparison is lost before it starts. --- ## 06 · Who answered | # | Role | Industry | Company size | | --- | --- | --- | --- | | 1 | VP of Engineering | Software Development | 5000+ | | 2 | Chief Product Officer | SaaS | 11-50 | | 3 | Founder | AI/Machine Learning | 51-200 | | 4 | Chief Technology Officer | Technology Services | 201-500 | | 5 | CTO | Software Development | 501-1000 | | 6 | Head of Engineering | SaaS | 1001-5000 | | 7 | VP of Engineering | AI/Machine Learning | 5000+ | | 8 | Chief Product Officer | Technology Services | 11-50 | | 9 | Founder | Software Development | 51-200 | | 10 | Chief Technology Officer | SaaS | 201-500 | | 11 | CTO | AI/Machine Learning | 501-1000 | | 12 | Head of Engineering | Technology Services | 1001-5000 | | 13 | VP of Engineering | Software Development | 5000+ | | 14 | Chief Product Officer | SaaS | 11-50 | | 15 | Founder | AI/Machine Learning | 51-200 | --- ## 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.