# Message test — https://www.salesforce.com/ After reading your page, only 1 of 15 personas could name a reason to pick you over a similar option. - **Page tested:** https://www.salesforce.com/ - **Audience tested against:** revenue leaders at enterprise SaaS - **Personas:** 15 simulated - **Report:** https://grader.wynter.com/r/salesforce-the-1-agentic-ai-crm-Yfy-Gis > 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? | 10/15 | 59% | all with reservations | | 3. Value | Do they actually want it? | 11/15 | 61% | all with reservations | | 4. Differentiation | Is there a reason to pick you over the alternatives? | 1/15 | 24% | with reservations | **Brand alignment** (a side metric, not one of the four layers) — 5/15, 41% 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 **Say how Data 360 reconciles conflicting records across systems.** 'Data 360 connects your systems, manages your data, and unifies customer context' is a sentence any data vendor could publish unchanged, so the offering reads as the existing stack with an AI layer on top. Replace it with the mechanism: how duplicate and conflicting records from separate systems are matched and resolved, what happens when two systems disagree, and what an agent sees as a result. The technical method is the differentiator; the phrase 'trusted business context' is not. *effort high · impact high · tested against Concrete over abstract* **Promote the Volkswagen call-handling result into the hero area.** The only evidence anyone believed on this page is the Volkswagen reduction in call handling time, and it sits buried mid-page under a broken '0%'. Move that named-customer result up beside the hero and write it out in full: the customer, the workload it handled, the measured drop, the timeframe. A concrete named outcome near the top does more competitive work than '#1 Agentic CRM' ever will. *effort medium · impact high · tested against Proof next to the claim* **Replace '#1 Agentic CRM' with a claim a competitor cannot copy.** '#1' and 'trusted source of truth' are claims every CRM vendor makes, and nothing on the page gave a reader a reason to pick this over an incumbent they already run. Use the H1 to state something specific and checkable — the measured outcome from a named customer, or what agents can do here that a bolted-on AI add-on cannot, with the reason why. *effort medium · impact high · tested against Give a reason to choose you* ### Relevance **Open with the problem the reader already feels.** The page starts at architecture — '#1 Agentic CRM', 'System of Context', 'System of Work' — and the business pain never gets stated. Add a short line above or beneath the H1 naming what is actually wrong today: customer data split across systems so reps retype it, service queues no one can staff, answers that differ depending on which tool you open. Name the pain in the reader's words before introducing the product names. *effort medium · impact high · tested against Problem before solution* **Name the buyer and their team in the hero subhead.** 'Your agents, your team, your data' leaves the reader working out from customer logos whether the page is for them. Rewrite the subhead to name a role and situation — a RevOps or service operations leader running a CRM their reps work in daily — so the intended buyer can point at a line and recognise themselves instead of inferring it from PepsiCo. *effort low · impact high · tested against Name the audience* **Define or drop 'System of Context', 'System of Work', 'System of Agency'.** These internal labels sit as headings above each product with no definition anywhere, so a first-time reader has to decode five coined terms before learning what any product does. Replace each with a plain description of the job that product does for the reader, and let 'Data 360', 'Customer 360' and 'Agentforce' follow as the product names. *effort low · impact medium · tested against Plain language* ### Value **Attribute performance claims to independent, named sources.** Three of the four stat blocks cite '2025/2026 Salesforce Success Metrics Global Highlights', which readers discounted as vendor-graded. Lead with the evidence that is not self-sourced — the Valoir analyst figure and the Volkswagen customer result — and give each the detail that makes it checkable: which customers, over what period, measured how. *effort medium · impact high · tested against Proof next to the claim* **Fix the '0%' statistics or remove them entirely.** Every stat block on the page renders as '0%increase in employee productivity', '0%reduction in call handling time', '0xfaster delivery of agents'. Broken numbers read as either poor quality control or reluctance to publish the real figures, and they contaminated the sources that did render. Either ship the actual values with the measurement period and sample, or pull the blocks until they work. *effort low · impact high · tested against Proof next to the claim* **Add a migration story from an incumbent CRM.** Nothing on the page tells a reader already running a competing CRM what moving looks like, so the value claims float free of their situation. Add a short named-customer account of a switch: what they came from, how long the move took, what happened to their data and their reps during it, and the result afterwards. *effort high · impact medium · tested against Answer the live objection* ### Clarity **Give the H1 a verb describing what the product does.** 'Get started with the #1 Agentic CRM' names a category and an award, not a function, and ten product labels — Slack, Tableau, Agentforce, Customer 360, Data 360 — stack up before any of them is explained. Rewrite the H1 so it states the job the product does for a customer-facing team, and let the product names appear only once the reader knows what they are for. *effort medium · impact high · tested against Lead with the use case* ### Brand alignment (side metric) **Rewrite section copy for the practitioner, not the contract signer.** Lines like 'Transform your business with our step-by-step guide to becoming an Agentic Enterprise' and 'infinitely scalable workforce' are boardroom register, and the skeptical daily user reads them as marketing rather than description. Rewrite each product section to describe what a rep, service agent or admin does differently on Monday morning — what they stop doing by hand, what the agent does instead, what they still control. *effort medium · impact high · tested against Plain language* **Add a mid-market proof point beside the enterprise logo wall.** PepsiCo, Siemens and Live Nation stacked twice signal a product built for enterprises with implementation teams, which reads as overbuilt to a mid-market buyer. Pair the logo strip with one named smaller customer and their result — company size, what they deployed, how long it took, what changed — so the claim 'Customers of every size' is carried by an example instead of an assertion. *effort medium · impact high · tested against Proof next to the claim* **Cut 'Any business, any size, any industry' and name one segment.** That line, sitting under 'Every business can be an Agentic Enterprise', tells the reader the page was written for nobody in particular and clashes with the enterprise imagery around it. Replace it with the segment this section actually serves — for example a revenue team at a growing SaaS company already running Salesforce — so the tone and the audience finally agree. *effort low · impact medium · tested against Name the audience* --- ## 03 · What is working ### The Volkswagen customer reference is the one proof point respondents believed Three respondents singled out the Volkswagen call stat as the only concrete, credible evidence of agent value on the page, and one said a named customer reference with verifiable data would be the deciding factor. It is the sole element that earned positive reactions. > at least it names a real company and a specific metric, even though the number itself rendered blank. If I could click through and see the actual before/after numbers, industry, and team size, that's the one piece of proof on this page I'd chase > > — Revenue Operations Manager, Technology, 1001-5000 > The one thing that would actually move me is the Volkswagen line — "reduction in call handling time" tied to a named customer story — because that's a real company I could theoretically go verify, unlike the generic "0% increase in employee productivity" stats sourced to Salesforce's own "Success Metrics" report > > — Director of Sales, Enterprise Software, 1001-5000 > The one thing that would actually pull me toward Salesforce over a rival is the Volkswagen line — "0% reduction in call handling time" is broken on this page, but a named, checkable enterprise customer story is real proof > > — VP of Sales, Software as a Service (SaaS), 1001-5000 > the Volkswagen call-handling reduction stat and the Data 360 pitch of "one trusted business context for every human and every AI agent" would matter to me > > — Chief Revenue Officer, Enterprise Software, 5000+ --- ## 04 · What the personas said ### Placeholder stats rendering as zeros destroy the credibility of every number on the page Nine respondents flagged the unrendered productivity statistics showing as zero percent. They read them not as a minor bug but as evidence of poor QA or a lack of confidence in the underlying numbers, and several said the broken figures cast doubt on the metrics that did render. > the stats like "0% increase in productivity" and unlabeled multipliers are useless without a defined baseline or methodology — that's the kind of thing that would stall a renewal conversation with me > > — Revenue Operations Manager, Technology, 1001-5000 > the page gives me stat soup with no baseline — "0% increase in productivity," "0hrs saved per rep" — those placeholders (literally rendering as 0%) > > — VP of Sales, Software as a Service (SaaS), 1001-5000 > a company this size shipping unrendered placeholder stats on their flagship page tells me either their QA is sloppy or they don't actually trust the numbers enough to hardcode them > > — Chief Revenue Officer, Enterprise Software, 5000+ > the stats like "0% increase in productivity" and "0hrs saved" are placeholders, not proof > > — Chief Revenue Officer, Enterprise Software, 5000+ > The percentages ("0% increase in productivity") were literally unrendered placeholders, so I can't tell you what the actual lift is — that's the kind of gap that would make me close the tab if I were actually evaluating this. > > — Revenue Operations Manager, Technology, 1001-5000 > "0% increase in productivity" with the number stripped out is a placeholder, not proof. > > — Head of Revenue, Software as a Service (SaaS), 5000+ > The stats like "0% increase in productivity" are broken/placeholder on this read, which undercuts the pitch > > — Head of Revenue, Software as a Service (SaaS), 5000+ ### Jargon and stacked product names obscure what the product actually does Four respondents said the hero line lacks a verb, ten product labels are stacked before any function is explained, and terms like "Agentic CRM", "System of Context", Data 360 and Agentforce go undefined. One added that the business pain point is buried behind architecture jargon. > The hero line itself — "your agents, your team, your data — working together so every customer interaction reflects your business at its best" — is pure brand fluff with no verb telling me what the product actually does > > — Director of Sales, Enterprise Software, 1001-5000 > the copy jumps straight to architecture ("System of Context," "System of Work," "System of Agency") without ever naming the pain point in plain language > > — Chief Revenue Officer, Enterprise Software, 5000+ > phrases like "System of Context" and "System of Agency" as labels for Data 360 and Agentforce are jargon layered on top of jargon, and it took a second pass to realise they're just naming which layer of the stack does what. > > — Head of Revenue, Software as a Service (SaaS), 5000+ > It's the layering of jargon without definitions — "Agentic CRM," "System of Context," "System of Agency," "System of Work" all get thrown around like I already know what distinguishes them > > — VP of Sales, Software as a Service (SaaS), 1001-5000 ### The page never names who it is for, leaving buyers to infer the audience from logos Six respondents said the target buyer is inferred from customer logos and verticals rather than stated. Several specifically noted the absence of a RevOps decision-maker, of SaaS company size positioning, and that the page claims to serve any business of any size. > they're not speaking to me specifically as a RevOps director, they're speaking to every department head at once > > — Revenue Operations Manager, Technology, 1001-5000 > it's "any business, any size, any industry" plus a grab-bag of verticals (Financial Services, Retail, Healthcare, Construction...), so I had to infer it's aimed at enterprise buyers already juggling multiple systems, not a focused ICP > > — Director of Sales, Enterprise Software, 1001-5000 > the tone isn't written "for someone like me" specifically — it's generic top-of-funnel marketing sweep, all breadth ("any business, any size, any industry") > > — Senior Vice President of Revenue Operations, Technology, 5000+ > I'd have appreciated one explicit "built for RevOps/CX leaders at enterprise scale" line rather than piecing it together from the layout. > > — Head of Revenue, Software as a Service (SaaS), 5000+ > Who it's for is inferred rather than stated outright: no line says "for RevOps directors at mid-market companies," but the industry cards (Financial Services, Retail, Healthcare, Manufacturing, etc.) and enterprise logos like PepsiCo and Siemens signal this is aimed at large-org buyers > > — Revenue Operations Manager, Technology, 1001-5000 > there's no line saying "built for RevOps leaders at X-size companies." I had to piece the problem together from the five product blocks > > — Senior Vice President of Revenue Operations, Technology, 5000+ ### Remaining performance claims are vendor-graded, so respondents discounted them Four respondents noted the proof points are sourced to Salesforce's own reports or vendor-commissioned benchmarks rather than independent ones. They said value claims rest on assertions they cannot verify, and generic superlatives further eroded trust in the surrounding claims. > The stats they dangle — "reduction in call handling time," "hrs saved per rep, every week," "16 times faster agent delivery" — are exactly the numbers I'd need to hit my cost-reduction mandate, but they're sourced to Salesforce's own "Success Metrics" reports and a vendor-commissioned Valoir study, not an independent benchmark I can trust > > — Director of Sales, Enterprise Software, 1001-5000 > "0x faster delivery," "#1 Global Software Company" — is an unsourced superlative that rules nothing in, and actively makes me discount the surrounding claims > > — Senior Vice President of Revenue Operations, Technology, 5000+ > the page's own proof points (the "0% increase in productivity" stats) were blank placeholders, and the one real number I could see, "16 times faster agent delivery," comes from a vendor-commissioned Valoir report, not an independent benchmark. That's not enough to take a meeting over > > — Revenue Operations Manager, Technology, 1001-5000 ### There is no proof or path for a buyer at the reader's own scale or current CRM Two respondents asked for a mid-market case study with real numbers tied to a comparable company size, and for a migration story covering a move off an incumbent CRM such as Dynamics. Without these, the page gives no evidence a reader can map to their own situation. > A real case study from a company our size showing a specific, non-zero number — like X% drop in call handling time or Y hours saved per rep per week > > — VP of Sales, Software as a Service (SaaS), 1001-5000 > the page gives me architecture and a customer logo, not a migration story: nobody tells me what happens to five years of Dynamics workflows and custom fields, or how many quarters this takes to land > > — Senior Vice President of Revenue Operations, Technology, 5000+ ### Respondents read the offering as the existing CRM stack with an AI layer, not something… One respondent stated the core offering is a rebrand of the existing stack rather than fundamental innovation, and another said Data 360 would only address data fragmentation if the technical reconciliation method were demonstrated. Combined with the broken numbers, no respondent could articulate a competitive differentiator. ### Enterprise framing reads as overbuilt and misaddressed for mid-market and practitioner… Five respondents said the enterprise logos, scale signals and boardroom tone position the product as too heavy for mid-market companies and speak to contract-signers rather than skeptical daily users. One noted the page chases top-of-funnel volume instead of existing customer expansion, and another said it addresses the current Salesforce ecosystem rather than converting competitors' customers. > it's written for someone already inside the Salesforce ecosystem, fluent in their internal jargon — "System of Context," "System of Work," "System of Agency," "Agentblazer Champion/Innovator/Legend." > > — Chief Revenue Officer, Enterprise Software, 5000+ > It doesn't feel written for a skeptical end user who has to live with the tool daily — it feels written for whoever signs the contract. > > — Revenue Operations Manager, Technology, 1001-5000 > The tone is generic brand messaging, not written for someone like me specifically — it's written for a boardroom persona in the abstract, not a Head of Revenue with a fragmented-data problem. > > — Head of Revenue, Software as a Service (SaaS), 5000+ > marketing wrote this for top-of-funnel volume, not for a renewal or expansion conversation with an existing customer like me > > — Director of Sales, Enterprise Software, 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 cannot be evaluated on its argument because it fails basic production standards first.** *(high)* Nine respondents flagged unrendered statistics showing as zero percent, and rather than dismissing it as a rendering bug they read it as poor QA or a lack of confidence in the numbers, with several saying the broken figures contaminated the metrics that did render. Damage spread beyond the broken elements to the intact ones. - **Every quantitative claim on the page is now worthless: the broken ones are unreadable and the working ones are vendor-graded.** *(high)* Nine respondents said the zeroed stats cast doubt on the figures that did render, and four separately discounted the surviving proof points as sourced to Salesforce's own reports or vendor-commissioned benchmarks. The two failures compound — there is no number left that a reader both can read and will believe. - **A single third-party customer stat is carrying the entire evidentiary weight of the page.** *(high)* The Volkswagen call stat was the only element to earn positive reactions, named by three respondents as the sole concrete, credible evidence of agent value, and one said a named customer reference with verifiable data would be the deciding factor. Meanwhile nine respondents rejected the zeroed stats and four discounted the vendor-sourced claims. One proof point is doing the job of the whole page. - **The page never states what it does or who it is for, so the reader is doing the positioning work the copy should have done.** *(high)* Four respondents said the hero line lacks a verb and stacks ten product labels before explaining any function, with Agentic CRM, System of Context, Data 360 and Agentforce left undefined. Six respondents said the target buyer has to be inferred from customer logos and verticals because it is never named. Both the what and the who are left as inference tasks. - **Claiming to serve any business of any size is why the page lands as too heavy for mid-market — the breadth claim reads as an admission of no fit.** *(high)* Several respondents noted the page claims to serve any business of any size while never naming a buyer, and five said the enterprise logos, scale signals and boardroom tone position the product as too heavy for mid-market. Two more asked for a mid-market case study with real numbers tied to comparable company size and got none. Unbounded audience claims plus enterprise-only proof equals no reader seeing themselves. - **The page is unusable as a competitive displacement asset.** *(high)* One respondent said the page addresses the current Salesforce ecosystem rather than converting competitors' customers, two asked for a migration story covering a move off an incumbent CRM such as Dynamics and found none, and no respondent could articulate a competitive differentiator. A reader on another CRM is given no reason and no route to switch. --- ## 06 · Who answered | # | Role | Industry | Company size | | --- | --- | --- | --- | | 1 | Revenue Operations Manager | Technology | 1001-5000 | | 2 | Chief Revenue Officer | Enterprise Software | 5000+ | | 3 | VP of Sales | Software as a Service (SaaS) | 1001-5000 | | 4 | Senior Vice President of Revenue Operations | Technology | 5000+ | | 5 | Director of Sales | Enterprise Software | 1001-5000 | | 6 | Head of Revenue | Software as a Service (SaaS) | 5000+ | | 7 | Revenue Operations Manager | Technology | 1001-5000 | | 8 | Chief Revenue Officer | Enterprise Software | 5000+ | | 9 | VP of Sales | Software as a Service (SaaS) | 1001-5000 | | 10 | Senior Vice President of Revenue Operations | Technology | 5000+ | | 11 | Director of Sales | Enterprise Software | 1001-5000 | | 12 | Head of Revenue | Software as a Service (SaaS) | 5000+ | | 13 | Revenue Operations Manager | Technology | 1001-5000 | | 14 | Chief Revenue Officer | Enterprise Software | 5000+ | | 15 | VP of Sales | Software as a Service (SaaS) | 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 2026-08-22, then deleted along with the personas and their answers.