# Message test — https://www.plepic.com/ After reading your page, only 6 of 15 personas could name what kind of product this is, unprompted. - **Page tested:** https://www.plepic.com/ - **Audience tested against:** Estonian software agencies with 10-50 employees and dev teams in larger orgs where claude code has been allowed. - **Personas:** 15 simulated - **Report:** https://grader.wynter.com/r/claude-code-course-and-agentic-ai-training-ple-2P19n4E > 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? | 6/15 | 97% | 13 without hesitation, 2 with reservations | | 2. Relevance | Can they tell what it solves, and who it's for? | 15/15 | 94% | 11 without hesitation, 4 with reservations | | 3. Value | Do they actually want it? | 13/15 | 70% | all with reservations | | 4. Differentiation | Is there a reason to pick you over the alternatives? | 10/15 | 61% | 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: Clarity.** Earliest failing layer, walking the sequence in order — not simply the lowest score. ### What they thought you sell 5 of the personas who named a category got it wrong: - 3× “Corporate technical training / upskilling course” - 1× “Corporate developer training / upskilling course” - 1× “Developer training / upskilling program” --- ## 02 · What to change, layer by layer Ordered worst-first. Specific edits, not a restatement of the score. ### Clarity **Define how the 3–5× speedup is measured, and against what.** The FAQ line 'letting developers ship 3–5× faster' floats free of any baseline. Readers had nothing to verify it against. Replace with a stated basis and conditions, e.g. 'Squads measure cycle time on their own backlog tasks in week one and week six; the median improvement across 300+ trained engineers was Nx on tasks of type X.' If a baseline cannot be published, say what is measured and who measured it rather than leaving a bare multiple. *effort medium · impact high · tested against Specifics beat superlatives* **Spell out the guarantee terms where the guarantee is claimed.** The money-back-at-3× promise is referenced as the headline reassurance but its conditions never appear on the page, so readers discounted it themselves. Add the mechanics next to the claim: what is measured, who measures it, by when, and what the refund covers (full fee, per dev, per squad). *effort medium · impact high · tested against Answer the live objection* **Explain 'Scopeful' and 'squad' on first use.** 'We're building Scopeful: pricing tied to outcomes, not time' and '~20 per squad' / '8.7/10 squad avg' use internal terms a first-time reader must decode. Either define squad as 'a cohort of up to 20 engineers from your team' at first mention, or drop it for 'cohort'. Give Scopeful one clause explaining it is a separate venture, not part of this training, so it stops reading as an undefined product name mid-page. *effort low · impact medium · tested against Plain language* **Rewrite 'The Shift' and 'The Bigger Picture' as meaning-carrying headings.** Section labels 'The Shift', 'The Bigger Picture', 'Meet the Team' and 'Common Questions' say nothing when scanned. Use the conclusion instead: 'An agent is an LLM using tools in a loop' already exists as the subhead — promote it. 'When developers ship faster, hourly billing breaks' likewise. A reader skimming only the headings should get the offer, the method, the proof and the price. *effort low · impact medium · tested against Headings stand alone* ### Differentiation **Attach an output number to at least one named client quote.** Helmes, Holm Bank and the rest are the main proof asset, but every quote is a mindset testimonial — 'biggest value was a mindset shift', 'broadens your thinking'. None reports a result. Add one line per testimonial with a concrete before/after from that team: task cycle time, PRs merged, time to first agent-shipped production change, with role and company beside it. *effort high · impact high · tested against Proof next to the claim* **State plainly what this offers that generic AI bootcamps cannot.** The real separators — your own repository, senior engineers sitting beside the team, a refund tied to a measured speed threshold — are scattered across hero, hackathon block and FAQ. Pull them into one short 'Why us, not a bootcamp' list high on the page, each phrased as a contrast: 'Your codebase, not sample projects.' 'Refund tied to a measured threshold, not a satisfaction promise.' *effort medium · impact high · tested against Give a reason to choose you* **Source the '300+ trained · 8.7/10 squad avg' bar.** The top-of-page stats are asserted with no basis and were read as unverified. State the period, the sample and how the score is collected, e.g. '8.7/10 average from 300+ engineers rated at the end of each six-week cohort, 2025'. *effort low · impact medium · tested against Proof next to the claim* ### Value **Promote 'bring one real task from your backlog' into the hero.** The strongest risk-reducer on the page — bringing a real backlog task so the team can test the speed claim on their own work — is buried below the fold. Put it directly under the hero subhead, beside the price, so the first screen shows how a buyer proves the claim themselves before committing. *effort low · impact medium · tested against Tie the feature to the outcome* ### Relevance **Separate the price from the Estonian subsidy line.** '€2,520/dev · 🇪🇪 Töötukassa covers 80%' makes the offer look Estonia-only at the exact moment a buyer elsewhere in the EU is deciding whether it applies to them. Lead with the standalone per-dev price and delivery mode (remote or on-site), then note the Töötukassa reimbursement as an Estonia-specific bonus underneath. *effort low · impact medium · tested against Name the audience* ### Brand alignment (side metric) **Reframe founding-year and team size as delivery capacity.** 'Kaido Koort founded Plepic in 2025' next to a three-person instructor roster reads as execution risk for a buyer committing a whole dev team for six weeks. Keep the 23 years and 50+ combined years, but add the operating facts that answer the doubt: cohorts delivered to date, squads run in parallel, and what happens if an instructor is unavailable. *effort medium · impact high · tested against Answer the live objection* **Move the Scopeful section below the program and proof.** 'We're building Scopeful: pricing tied to outcomes, not time' and 'hours × rate → scope × value' put the founder's next venture and a critique of agency billing ahead of what the buyer is being asked to buy. It reads as a manifesto interrupting the offer. Demote it or cut it to a single line in the About area. *effort low · impact medium · tested against Front-load the meaning* --- ## 03 · What is working ### The offer itself — six Fridays of Claude Code training on the team's own codebase — is… Respondents consistently read back the same offer: a paid six-week/six-Friday cohort teaching agentic coding on the customer's own repository, aimed at senior engineers rather than beginners. Eight points name the hero line, subhead, or FAQ as doing this work explicitly, and several noted the audience is stated more than once in the opening screens. Nobody mistook it for a product or tool. > It's a paid corporate training course teaching senior devs to use Claude Code (agentic AI coding) on their own codebase — six Fridays, certificate at the end, run by a small Estonian outfit called Plepic. > > — Head of Engineering, Professional Services, 501-1000 > it's stated twice in slightly different words within the first screen and again in the FAQ > > — Head of Engineering, Professional Services, 501-1000 > The subhead itself says "Claude Code Training for Dev Teams" and right under it "senior engineers beside you, your real codebase," then the FAQ nails the audience directly: "Senior engineers with 2+ years of professional experience and access to a real codebase... Not for learning to code." > > — CTO, Software Development, 1001-5000 > Six Fridays, senior engineers beside you, your real codebase. You direct the agent, it ships production code," and right under it "For engineers with 2+ years experience · ~20 per squad." That's a clear problem statement > > — Software Developer, Software Agencies, 1001-5000 > It's a six-Friday training program that teaches senior dev teams to use Claude Code (agentic AI coding) on their own codebase, run by Plepic > > — CTO, Software Development, 1001-5000 > the hero line "Six Fridays, senior engineers beside you, your real codebase" plus "For engineers with 2+ years experience · ~20 per squad" tells you exactly who this is for within the first screen > > — Software Developer, Software Agencies, 11-50 ### 'Bring one real task from your backlog' was named as a low-friction way in One respondent picked out the invitation to bring a real backlog task as a low-effort entry point that lets a buyer pressure-test the speed claim themselves. It was the only specific mechanism on the page that a respondent said reduced their own risk of trying it. > they're framing it as "bring one real task from your backlog," which is low-effort and lets me pressure-test the claim on our own code rather than a demo > > — CTO, Software Development, 1001-5000 --- ## 04 · What the personas said ### Key terms are used without definitions or baselines One respondent said the page's key terms lack definitions or baselines against which anything could be verified. This sits alongside the unexplained measurement basis for the 3x claim raised elsewhere. > the terms "agentic harness," "CLAUDE.md architecture," and the "3-5x faster"/"3x time savings" figures — each is used as if it's a settled, load-bearing concept, but none is defined with a baseline or a worked example > > — Software Developer, Software Agencies, 1001-5000 ### Estonia-specific framing and missing enterprise commercial terms limit how far the page… One respondent said the Estonia-only subsidy makes the offer look Estonia-first and limits applicability across EU hiring. Another said they would need enterprise volume pricing and confirmation that multiple cohorts can be run before the page is actionable for them. > I'd need a line item aimed at multi-team procurement — something like volume pricing per squad, a named enterprise client of our size (1000+ devs, not a 20-person shop), and confirmation this scales beyond one cohort of ~20 > > — Software Developer, Software Agencies, 1001-5000 ### The 3x claim is the page's central promise and almost nobody believed it as written Six respondents flagged the 3x/3-5x speed claim as unproven: no before/after numbers, no named customer attached to a result, no explanation of how the speedup is measured or under what conditions. This is the most repeated negative reaction on the page and it attaches to the single biggest claim being made. > I'd want one concrete before/after example (a real PR, real time logged) rather than the 8.7/10 squad average number, which is meaningless without a source > > — Staff Engineer, Software Development, 51-200 > the urgency is asserted ("agentic AI changes how code gets written") but not proven with a concrete before/after case study, just testimonials that are more about mindset shift than measurable output. > > — Engineering Manager, Software Agencies, 201-500 > that 3x refund guarantee has "conditions apply" doing a lot of unexplained work, and the only proof offered is three quote-blurbs from unnamed-scale companies, not hard before/after numbers > > — Head of Engineering, Professional Services, 501-1000 > If the "3x time savings on typical development tasks" claim actually held, that'd be worth something — but I already use Claude day to day, so the real question is whether six Fridays of structured coaching gets me meaningfully further than just messing with it myself, and the page doesn't prove that. > > — Senior Software Developer, Professional Services, 11-50 > Devs ship 3x faster, they claim. Refund guarantee is a nice hook, but no named customer results shown. > > — Staff Engineer, Software Development, 201-500 > right now that's the one unsupported qualifier standing between a real guarantee and marketing fluff > > — Senior Software Developer, Professional Services, 51-200 > the promise of 3-5x speed with a money-back guarantee if you don't hit 3x savings > > — CTO, Software Development, 1001-5000 ### Named clients appear but the testimonials carry no output metrics, so the proof stops… Respondents noticed the named Estonian client list — Helmes, Holm Bank — and treated it as the main proof asset, then said the accompanying quotes lack case studies or output numbers. Two respondents specifically described the testimonials as unnamed or unsupported, and one called an on-page statistic unverified. > the named testimonials right after it (Kristina Krist, Erich Jagomägis, Merilin Sellik) all talk about "mindset shift" and "clear direction," not a single output metric > > — Head of Engineering, Professional Services, 501-1000 > Tone's aimed at senior devs like me, but the "8.7/10 squad avg" stat with no source undercuts it. > > — Staff Engineer, Software Development, 201-500 > Real named clients like Bitweb, Mindworks give quotes with names attached — that's a plus. But "8.7/10 squad avg" with no source, and no named case study on the 3x claim, rules it out > > — Staff Engineer, Software Development, 201-500 ### The company reads as a small, founder-led, 2025-vintage Estonian shop, and several… Respondents described a three-person or founder-led operation founded in 2025 with roughly 20 per squad. Two framed this plainly as execution risk and as misaligned with enterprise expectations; others noted it neutrally as a consultancy that productized training. No respondent read the size as an asset. > it also means I'd be betting on three people's bandwidth, not an institution > > — Staff Engineer, Software Development, 51-200 > Small Estonian startup, founder-led, one guy's story front and center — feels early-stage, not enterprise. > > — Staff Engineer, Software Development, 201-500 > probably under 10 people, launched in 2025 given Kaido's "founded Plepic in 2025 after 23 years in software" line > > — Senior Software Developer, Professional Services, 51-200 ### The refund guarantee reads as the strongest differentiator but its terms are missing Respondents singled out the money-back-at-3x guarantee as what separates this from generic bootcamps and from competitors offering only satisfaction promises. Two respondents then undercut it themselves, saying the conditions are unspecified and the lack of stated terms weakens it as a differentiator. > "we work on your actual codebase, the code you'll ship Monday" combined with the refund clause — that's a concrete mechanism, not just marketing fluff > > — Staff Engineer, Software Development, 51-200 > The one thing that would actually tip me toward picking this over a generic competitor is the refund line: "If you don't experience at least 3x time savings in typical development tasks after completing the training, we'll refund your contribution in full." > > — Engineering Manager, Software Agencies, 201-500 > that 3x refund guarantee has "conditions apply" doing a lot of unexplained work, and the only proof offered is three quote-blurbs from unnamed-scale companies, not hard before/after numbers > > — Head of Engineering, Professional Services, 501-1000 > the promise of 3-5x speed with a money-back guarantee if you don't hit 3x savings > > — CTO, Software Development, 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 central promise is dead on arrival: the 3x claim is the entire value proposition and it collapses under the lightest scrutiny.** *(high)* Seven of 15 respondents flagged the 3x/3-5x speed claim as unproven — no before/after numbers, no named customer attached to a result, no measurement basis. It is the most repeated negative reaction on the page and it lands on the single biggest claim. One respondent additionally said key terms carry no definitions or baselines at all, so there is nothing on the page a buyer could verify the number against. - **The guarantee cannot rescue the unproven claim, because the guarantee is written on the same unverifiable number.** *(high)* Five respondents named the money-back-at-3x guarantee as the strongest differentiator, but two immediately undercut it by noting the conditions are unspecified. The guarantee is denominated in the same 3x metric that seven respondents said has no measurement basis or before/after numbers. Unverifiable metric plus unstated terms means the page's best asset is unenforceable in the reader's mind. - **The page has proof assets and wastes them.** *(high)* Three respondents found the named Estonian clients — Helmes, Holm Bank — and treated that list as the main proof on the page, then said the quotes carry no case studies or output numbers; two described the testimonials as unsupported and one called an on-page statistic unverified. Named logos were available to substantiate the 3x claim that seven respondents rejected, and the page attached no result to any of them. - **The page actively broadcasts execution risk instead of managing it.** *(high)* Four respondents read back a three-person or founder-led shop founded in 2025 with roughly 20 per squad; two named this plainly as execution risk and misaligned with enterprise expectations, and no respondent read the small size as an asset. Alongside this, two respondents said the Estonia-only subsidy makes the offer look Estonia-first and that enterprise volume pricing and multi-cohort confirmation are absent — so the page reads as a small local shop that has not stated it can serve at scale. - **Clear comprehension is being mistaken for persuasion — respondents understood the offer and still had no reason to buy.** *(high)* Eight of 15 respondents read back the offer identically — a paid six-Friday cohort on the customer's own repo for senior engineers — and nobody mistook it for a tool. Yet seven rejected the value claim, three found the proof empty, two said the guarantee terms were missing, and two said commercial terms were missing. Comprehension is the page's only strength; every claim built on top of it failed. - **Only one risk-reducing mechanism landed on the whole page, and a single respondent found it.** *(medium)* 'Bring one real task from your backlog' was named by exactly one respondent as a low-friction entry point, and that respondent said it was the only specific mechanism on the page that reduced their own risk of trying it. Against seven respondents rejecting the headline claim, the page is carrying one under-surfaced de-risking device that 14 respondents did not mention. --- ## 06 · Who answered | # | Role | Industry | Company size | | --- | --- | --- | --- | | 1 | Software Developer | Software Agencies | 1001-5000 | | 2 | Senior Software Developer | Professional Services | 11-50 | | 3 | Staff Engineer | Software Development | 51-200 | | 4 | Engineering Manager | Software Agencies | 201-500 | | 5 | Head of Engineering | Professional Services | 501-1000 | | 6 | CTO | Software Development | 1001-5000 | | 7 | Software Developer | Software Agencies | 11-50 | | 8 | Senior Software Developer | Professional Services | 51-200 | | 9 | Staff Engineer | Software Development | 201-500 | | 10 | Engineering Manager | Software Agencies | 501-1000 | | 11 | Head of Engineering | Professional Services | 1001-5000 | | 12 | CTO | Software Development | 11-50 | | 13 | Software Developer | Software Agencies | 51-200 | | 14 | Senior Software Developer | Professional Services | 201-500 | | 15 | Staff Engineer | Software Development | 501-1000 | --- ## 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.