# Message test — https://www.siena.cx/ After reading your page, 11 of 15 personas could quickly tell what problem it solves and who it is for. - **Page tested:** https://www.siena.cx/ - **Audience tested against:** Heads of CX and heads of ecommerce Financial buyers are CEO/CFO, champion buyers are CX leaders, day-to-day users are less technical support people - **Personas:** 15 simulated - **Report:** https://grader.wynter.com/r/empathic-ai-agents-for-commerce-siena-ai-kTcabmQ > 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 | 79% | 1 without hesitation, 14 with reservations | | 2. Relevance | Can they tell what it solves, and who it's for? | 11/15 | 64% | 1 without hesitation, 10 with reservations | | 3. Value | Do they actually want it? | 15/15 | 78% | all with reservations | | 4. Differentiation | Is there a reason to pick you over the alternatives? | 13/15 | 72% | all with reservations | **Brand alignment** (a side metric, not one of the four layers) — 10/15, 59% strength (all with reservations). 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 an integrations block naming the helpdesk platforms the product connects to.** Nothing on the page says how this sits on top of an existing helpdesk stack, so buyers assume a migration. List the named helpdesks it plugs into and what happens to existing tickets and routing rules. *effort medium · impact high · tested against Answer the live objection* **Add an audience line under the H1 naming the roles and company type served.** Buyers had to work out who the product is for from the logo wall and testimonial job titles. State it outright: DTC and ecommerce support teams, and the CX or ops leader who owns the ticket queue. *effort low · impact high · tested against Name the audience* **Add a section stating which team owns the purchase and who gets involved.** Readers could not tell whether CX, ops or marketing buys this, so nobody knew if the page was addressed to them. Name the owning team and the stakeholders it drags in. *effort medium · impact medium · tested against Name the audience* ### Differentiation **Add a European customer case study with the same metrics as the US ones.** Every reference is US-based and each case study reports a different measure, so buyers outside the US cannot judge whether it works at their scale. Report the same two or three numbers for every named customer, including one in Europe. *effort high · impact medium · tested against Proof next to the claim* ### Value **Add pricing model, implementation time and escalation handling beside the case studies.** The case studies win a meeting, then the page goes silent on the three things that actually decide between vendors. State how pricing works, how long onboarding takes and what happens when the AI hands off to a human. *effort medium · impact high · tested against Answer the live objection* ### Brand alignment (side metric) **Move a product screenshot or resolved-ticket view above the logo wall.** The category only becomes clear once a reader reaches the demo transcript far down the page. Show the product resolving a real ticket near the top so the thing is visible before the claims. *effort medium · impact high · tested against Show the product early* **Replace the 'operating system' framing in the H1 with the plain function.** The category label reads as marketing abstraction, and one reader decoded it as a chatbot with analytics bolted on. Say what it does in plain words: AI support automation that resolves customer tickets. *effort low · impact high · tested against Lead with the use case* **Rewrite section headings so the first three words carry the meaning.** Read alone, the headings describe themes rather than outcomes, so a scanning reader learns nothing. Lead each with the result: tickets resolved without an agent, CSAT held steady, go live in weeks. *effort medium · impact medium · tested against Front-load the meaning* --- ## 03 · What is working ### The buyer persona and problem are legible from the opening section One respondent said the tagline and brand logos in the first section made the persona and problem statement clear immediately. > the tagline "The AI CX operating system for consumer brands" plus "One intelligence layer powering agents across every customer surface" tells you the problem (fragmented CX tooling/manual support) and the buyer (CX/ops leaders at consumer brands) in the first two lines > > — Director of Customer Experience, Ecommerce, 201-500 ### Named case studies with attributed automation percentages are the one thing that lands Six respondents cited case studies with named brands and specific metrics (68% automation, CSAT holding) as the most convincing and most differentiating content, more persuasive than the product description itself. > The named case studies with hard numbers attached to specific brands — "Kitsch automates over 68%," "Fresh Clean Threads Achieves 91% CSAT," "Terra Kaffe... save over $3,000 per month" — that's the one thing that differentiates this from a generic AI-CX pitch > > — Director of Customer Experience, Ecommerce, 201-500 > "Kitsch automates over 68%," "98% CSAT" at Simple Modern, "65% AI Automation" at Coterie and HexClad — that's what would pull me toward this one over a competitor, because the numbers are tied to a named exec and a named brand > > — VP of Ecommerce, Digital Commerce, 1001-5000 > A verified case study from a brand our size and ticket complexity showing automation rate plus CSAT and handle-time change on the same before/after basis — without that comparable proof point, it's just a demo I'd take once and forget. > > — Head of Customer Experience, Retail, 201-500 > The names stuck with me more than the mechanics: Spanx, Kitsch, Coterie, HexClad, Prose all using it and citing things like "cut handle time in half" or "65% AI automation" is what would make me pitch this internally, not the product description itself. > > — Head of Ecommerce, Retail, 5000+ > The named case studies with hard numbers — Kitsch at "68% of support automated," Coterie at "65% AI Automation," HexClad "Boosts CX Efficiency 65%" — are what would make me pick this over a competitor with just a generic logo wall. > > — Director of Ecommerce, Ecommerce, 5000+ > The Kitsch, Coterie, and HexClad numbers (68%, 65%, 65% automation with CSAT holding or improving) are the specific proof points that make me believe this isn't just chatbot theater — CSAT usually craters when you automate, and they're claiming the opposite. > > — Director of Customer Experience, Ecommerce, 1001-5000 > The specific automation percentages tied to named brands — "68%" for Kitsch, "65%" for Coterie, "91% CSAT" for Fresh Clean Threads — are the only thing that would move this forward for me > > — VP of Customer Experience, Digital Commerce, 501-1000 --- ## 04 · What the personas said ### The product category is stated in jargon rather than plain function Four respondents called the positioning vague or unclear, one had to scroll to a demo transcript to work out the category, and one rejected the 'operating system' framing as chatbot-plus-analytics. > Phrases like "AI CX operating system" and "intelligence layer powering agents across every customer surface" are jargon that sound big but don't say plainly what the product does > > — VP of Ecommerce, Digital Commerce, 1001-5000 > the actual product looks like a chatbot-plus-analytics tool, not some new category > > — VP of Ecommerce, Digital Commerce, 201-500 > both are category-defining words but they're used without any definition of what layer sits where in the stack, so I couldn't tell if it replaces my helpdesk, sits on top of it, or just plugs into Gorgias/Zendesk-type tools until the chat transcript implied the latter. > > — Director of Ecommerce, Ecommerce, 501-1000 ### The page never states who it is for, leaving buyers to reverse-engineer the audience… Six respondents said the target buyer was inferred from the customer logo wall, testimonials and VP/COO titles rather than stated anywhere. Two also could not tell whether CX, ops or marketing owns the purchase. > the specific reader (CX lead? ecommerce VP? both?) I had to piece together from the testimonials rather than see spelled out upfront > > — Head of Ecommerce, Retail, 501-1000 > that's inferred from context (logos like Spanx, Kitsch, HexClad, Coterie) rather than spelled out as "this is for VPs of CX at DTC brands." > > — VP of Ecommerce, Digital Commerce, 1001-5000 > The reader is implied rather than stated outright, but it's obvious from the logo wall (Spanx, Kitsch, HexClad, Coterie, Grüns) and the quotes from VP of CX, COO, CX Lead titles — this is built for consumer/ecommerce brands' CX leaders, which is exactly my seat. > > — Director of Customer Experience, Ecommerce, 1001-5000 > The reader isn't spelled out explicitly anywhere, no "for VPs of CX at DTC brands" line — I inferred it from the customer logos and titles like "VP of Global CX" and "COO," so I had to piece it together rather than being told directly. > > — Head of Ecommerce, Retail, 5000+ ### Integration with existing helpdesk and infrastructure is unaddressed Two respondents wanted specifics on how the product fits existing helpdesk stacks and written confirmation of overlap with Cloudflare bot management tooling. > I'd need my ticket volume and channel mix (chat, email, social) named explicitly, plus a line on how it plugs into or replaces my existing helpdesk and Cloudflare setup > > — Director of Customer Experience, Ecommerce, 201-500 ### The proof shown is US-only and inconsistent, so buyers cannot map it to their own scale Three respondents flagged the absence of European references and metrics that vary across case studies, preventing size and complexity comparison. One raised a named customer churning to a competing enterprise platform. > every stat is US-based (Spanx, Coterie, HexClad, Kitsch) — nothing EU, no GDPR or EU-data-handling mention — and that alone would make me pause before shortlisting it seriously > > — Director of Ecommerce, Ecommerce, 5000+ > every number is a different metric (automation %, CSAT, efficiency) with no common baseline, so I can't tell if an 80%-automation brand is comparable in size or ticket complexity to us > > — Head of Customer Experience, Retail, 201-500 > they "were able to take Siena and all the training we did with it and move it to a more enterprise CX platform that better suited our needs" — that's a churn story sitting right there in the testimonials > > — Head of Ecommerce, Retail, 501-1000 > I'd take a meeting to get hard numbers from a brand my size in Europe, not another logo wall, before I'd spend real time on this. > > — Director of Ecommerce, Ecommerce, 5000+ ### Some respondents did read the core offer correctly Three respondents restated the product without prompting as AI-driven support automation for DTC and ecommerce brands, one noting it clearly positions away from enterprise. > It's an AI customer service agent platform for ecommerce brands—handles support tickets, chats, and some upsell/CX tasks > > — Head of Ecommerce, Retail, 501-1000 > It's an AI-powered customer service platform for ecommerce/retail brands — handles support tickets, chat, and some upsell conversations > > — Head of Customer Experience, Retail, 201-500 > It's an AI customer service platform for ecommerce brands — basically an AI agent that handles support tickets, chats, and some upsell conversations > > — Head of Customer Experience, Retail, 1001-5000 ### The case studies buy a meeting but not a decision — respondents want references and… Four respondents said they would only proceed after a reference call with a comparable customer, and one listed missing pricing, implementation time and escalation process as what actually decides between vendors. > It's worth a first meeting to see the mechanism and ask for a reference call with someone like Kitsch who moved off Siena > > — Head of Ecommerce, Retail, 501-1000 > I'd take a meeting to get hard numbers from a brand my size in Europe, not another logo wall, before I'd spend real time on this. > > — Director of Ecommerce, Ecommerce, 5000+ > A verified case study from a brand our size and ticket complexity showing automation rate plus CSAT and handle-time change on the same before/after basis — without that comparable proof point, it's just a demo I'd take once and forget. > > — Head of Customer Experience, Retail, 201-500 > That's worth a first meeting, yes — but only a scoping one, where I'd want them to walk me through what happens to my existing team's roles day-to-day and get a reference call with someone like the Kitsch or HexClad contact before I'd put my name on anything > > — Head of Ecommerce, Retail, 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 makes buyers do the qualification work it should be doing for them** *(high)* Six respondents reverse-engineered the audience from logos, testimonials and job titles, and two could not tell whether CX, ops or marketing owns the purchase. Only one found the persona legible from the opening. - **Jargon positioning means the case studies are carrying the category explanation the copy refuses to make** *(high)* Four respondents called the positioning vague, one had to reach a demo transcript to work out the category, and one rejected 'operating system' as chatbot-plus-analytics — while seven cited case studies as the clearest, most persuasive content. - **Only a third of readers can restate what the product does, so the headline messaging is not doing its job** *(high)* Three of fifteen restated the offer correctly as support automation for DTC and ecommerce, against four who found the category stated in jargon and six who could not find a stated audience. - **The strongest asset on the page is also its ceiling — proof that convinces cannot close** *(high)* Seven respondents named the case studies as the most differentiating content, but five said they would only proceed after a reference call, with pricing, implementation time and escalation process missing. - **The proof is not durable: inconsistent and geographically narrow evidence undercuts the one thing that lands** *(medium)* Three respondents flagged US-only references and metrics varying across case studies, blocking size comparison, and one raised a named customer churning to a competing enterprise platform. - **Silence on the existing stack leaves the buyer assuming rip-and-replace** *(medium)* Two respondents asked how the product fits existing helpdesk stacks and wanted written confirmation of overlap with Cloudflare bot management tooling; neither question is answered on the page. --- ## 06 · Who answered | # | Role | Industry | Company size | | --- | --- | --- | --- | | 1 | Director of Customer Experience | Ecommerce | 201-500 | | 2 | Head of Ecommerce | Retail | 501-1000 | | 3 | VP of Ecommerce | Digital Commerce | 1001-5000 | | 4 | Director of Ecommerce | Ecommerce | 5000+ | | 5 | Head of Customer Experience | Retail | 201-500 | | 6 | VP of Customer Experience | Digital Commerce | 501-1000 | | 7 | Director of Customer Experience | Ecommerce | 1001-5000 | | 8 | Head of Ecommerce | Retail | 5000+ | | 9 | VP of Ecommerce | Digital Commerce | 201-500 | | 10 | Director of Ecommerce | Ecommerce | 501-1000 | | 11 | Head of Customer Experience | Retail | 1001-5000 | | 12 | VP of Customer Experience | Digital Commerce | 5000+ | | 13 | Director of Customer Experience | Ecommerce | 201-500 | | 14 | Head of Ecommerce | Retail | 501-1000 | | 15 | VP of Ecommerce | Digital Commerce | 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 2026-09-30, then deleted along with the personas and their answers.