erom.io had a category bet (Governed AI Delivery) and a complex platform — but no clarity on who to sell to or how. A B2B Customer Intelligence Sprint surfaced the buyer's language, the real alternatives, and the unifying insight. Then April Dunford's Sales Pitch structure turned that evidence into a positioning narrative.
erom.ioOctavian, Founder/CEO of erom.io, came to the Sprint with six hypotheses — his POV on what mid-market manufacturers were really struggling with. Each one was paired to a specific buyer type. The Sprint took those hypotheses to the buyers and let their words decide.
Hypotheses 01 & 02 framed the economic buyer's concerns. 03 & 04 named what the user buyer lives daily. 05 & 06 surfaced what the technical buyer sees. Each hypothesis below carries the buyer voices behind it, plus a confidence grade for how much weight those voices can bear. Because this is synthetic research, the grade matters as much as the quotes.
Result: 3 High · 2 Moderate · 1 Low. One ceiling applies to every grade: synthetic research tells you where to look, not what is proven. High means strongest signal and first in line for real-buyer interviews, not validated.
What it supportsGrowth has outpaced systems built for a smaller company. The evidence shows strain, not a hard ceiling: growth hasn't stopped, and only Marcus ties it to a specific change in headcount.
Real-buyer testAsk real CEOs and COOs for the last moment a process broke because of growth, and what they did next.
What it supportsFragmentation survives the purchase of category tools. Only the CEO still runs on spreadsheets and email; the CCO and CISO bought dedicated platforms and still reconcile between them by hand.
Real-buyer testMap each buyer's actual stack and count the manual handoffs between systems in one audit cycle.
What it supportsAuditors and customers now check that controls are implemented, not just documented: Ramesh's SOC 2 deficiency and the client who questioned Marcus's training records. No buyer reported an outside party demanding real-time proof. Real-time is what buyers want, not what they are asked for.
Real-buyer testAsk real compliance leads what their last auditor or customer actually requested, and in what form.
What it supportsThe cost of manual verification is real and measurable in days and weeks. It is less hidden than unbudgeted: buyers can name it, but it never shows up as a line item.
Real-buyer testHave real buyers estimate hours per audit cycle spent gathering and verifying evidence, then check it against calendars or tickets.
What it supportsThe strongest single evidence in the Sprint: an actual SOC 2 deficiency, and a GRC platform reporting controls as closed that were never implemented. ISO and SOC 2 are supported. DORA was never mentioned, and as an EU financial-services regulation it likely sits outside this ICP. The “existential” framing rests on a feared revenue loss, not one that happened.
Real-buyer testRecruit real buyers with an audit finding in the last 18 months and reconstruct the timeline from first signal to fix.
What it supportsBuyers want their time back for strategic work, and Eleanor already uses workflow tools to catch issues early, three to four times in six months. No buyer mentioned KPIs, bottlenecks, or quality gains. Aspirational answers are where synthetic participants agree most readily, so this pull is still a hypothesis.
Real-buyer testFind real buyers who already tried continuous monitoring: what they kept, and what they abandoned.
The six hypotheses above were taken to synthetic buyers — AI research participants modeled on the four committee archetypes — using Synthetic Users. Each verbatim reflects population-level signal across the profile, not a single voice, which is how a 7-day sprint reaches buying-committee depth that recruited research would take months to surface.
Octavian didn't start with the Insight. He started with six hypotheses. That's a critical distinction for pre-PMF founders. A hypothesis is a falsifiable claim you're willing to be wrong about. An "insight" stated upfront — without buyer evidence behind it — is a vendor opinion in disguise.
The buyer-pair structure (Economic / User / Technical) wasn't decorative. It made the hypotheses *committee-routable*: each pair was framed in the language and worldview of a different stakeholder. When Sprint findings came back, Octavian could see which hypotheses survived contact with which buyer — and which ones needed sharpening.
The synthesized Insight at the top of this section rests on the High-confidence hypotheses (02, 04, 05) and the narrower, supported form of 03. Hypothesis 06, the aspirational pull, stays directional. That grading is the move worth copying: it tells a pre-PMF founder which claims can lead the pitch today, and which to take into real-buyer interviews first.
About the four archetypes you'll meet below: Marcus, Eleanor, Ramesh, and Fiona aren't single people — they're validated behavioral archetypes generated through Synthetic User Research. Each one represents the distribution of hundreds to thousands of real buyers matching that profile (mid-market US/UK manufacturing, 200–1,000 employees, <$20M ARR, ISO/SOC 2 contexts). One interview with a synthetic archetype reflects population-level signal, not a single voice. That's how a 7-day sprint surfaces buying-committee patterns that traditional research would take months to reach.
Nine alternatives surfaced in research — but they cluster into three fundamentally different ways buyers have tried to solve this. Each cluster leaves a distinct gap, which means each one suggests a distinct positioning opening for erom.io.
erom.io doesn't compete with GRC platforms, workflow tools, or security stacks. Those are hired tools owned by different buyers — ripping them out is a non-starter. erom.io competes with the manual labor people do on top of those tools to make them deliver verifiable, real-time, governed compliance. That reframes erom.io from "another platform to buy" to "the absorption layer your committee is already paying for in headcount and burnout."
This is what "Governed AI Delivery" should actually mean to a buyer: not replace-your-stack, but orchestrate-and-govern-the-work-across-it, with verifiable evidence as the byproduct.
Six dimensions emerged across all four archetypes, in their own language. Together, they describe the destination — and prime the buyer to see your solution as the bridge.
Marcus (CEO): "A fully integrated system that captures everything we need in real-time." Wants to "stay audit-ready" and "spend more time developing strategy."
Eleanor (COO): "A centralized compliance dashboard… holistic view of compliance status at any given moment." Documentation "automatically updated concurrently with any operational changes."
Ramesh (CCO): "Automatically pull evidence directly from the source system and verify it against the control objective." Compliance "wouldn't be 'Ramesh's problem.'"
Fiona (CISO): "Single pane of glass view" identifying risks "as they happen." Security as "enabler of innovation and efficiency, not just a barrier."
There's the category-language frame — the one analysts, the board, and investors use. It works for one audience. The buyer needs a frame they already half-believe.
Why not lead with the analyst category: Octavian's "Why Now" surfaced the risk — the category is real, but pre-PMF buyers don't shop by category name, they shop by problem. Category language is right for analyst conversations and Octavian's strategic narrative. Buyer-facing, the introduction has to mirror language Ramesh, Eleanor, Marcus, and Fiona already use: "verification," "evidence," "single pane of glass," "audit-ready."
The dual frame: category-level ("Governed AI Delivery") → buyer-level ("the layer that absorbs the manual work and produces evidence as a byproduct"). Both stay true. They serve different conversations.
The buying committee has four roles. The pitch can carry one unified value story — but the emphasis flexes by who's in the room. JTBD makes this routable, not improvised.
Marcus, Eleanor, Ramesh, and Fiona are behavioral archetypes generated with Synthetic Users, each carrying its own OCEAN personality profile. The value emphasis flexes by archetype because the underlying research did — same job, four worldviews.
The functional job is shared (integrated, real-time, auto-verifying) — that's where category-level value lives. But the emotional and social jobs differ, and those are what actually close deals. Selling "audit-ready" to Marcus is right; selling "audit-ready" to Ramesh is not — he wants "verification I can trust without playing detective."
The Cycle of Progress stage matters too: Marcus is the only buyer Actively Looking. He's the entry point. Ramesh, post-Ordeal, is the deepest champion candidate — his SOC 2 wound is the most visceral. Eleanor and Fiona are validators in Continuous Use, not initiators; they need a re-orchestration story, not greenfield.
Pre-PMF proof is honest about what it has and what it's building. erom.io has one published case study (Clear Assured) — and three independent peer voices that speak to the dimensions a buyer most needs to believe: technical credibility, governance credibility, and orchestration credibility.
Each quote speaks to a different reason a buyer might doubt erom.io. Together they cover the technical, the governance, and the orchestration claims the Sales Pitch makes.
Yes — erom.io has one published case study. That's the honest state of the proof layer, and it's exactly where most pre-PMF B2B companies live. The temptation is to dress up logos and testimonials to look like a wall of social proof. The Dunford-grade move is the opposite: pick the proof points that map to distinct dimensions of doubt, and let the absence of a full case study stable be a stage-of-business fact, not a story problem.
Why these three peer voices, not five: Each one answers a different question a buyer might raise. Can erom.io build the hard thing? (Cirdan, banking). Can it govern AI without creating a black box? (Sandulescu). Can ops and engineering both live in it? (Stanisor). Three covers the proof surface; five would just dilute the structure.
And one more thing: Ramesh's SOC 2 deficiency story from the Sprint is itself a proof point — not of erom.io yet, but of why the gap erom.io closes is real, audit-relevant, and currently unaddressed by any category tool. That diagnostic proof — buyer-side evidence that the problem is real — carries weight while the case-study stable builds.
When you know your buyer's OCEAN profile, leadership style, and past Hiring/Firing moves, the objections are no longer surprises. They're predictable, and the response is rehearsable.
Each objection maps to the OCEAN profile of the archetype it comes from. High-Conscientiousness, low-Neuroticism buyers don't object on emotion — they object on evidence, risk, and operational fit. The pattern is consistent across all four: show me proof, show me a path to scale that limits my exposure, show me how this fits with what I already own.
What this unlocks: erom.io doesn't need 40 objection-handling slides. It needs four — one per archetype — rehearsed cold.
A single CTA wastes three of the four archetypes. The Ask varies by where the buyer is in the Cycle of Progress — and the entry-point Ask should be calibrated to the buyer most likely to say yes.
Cycle of Progress stage is the lever: Marcus is the only Actively Looking buyer, with a defined trigger (audit notice, client documentation flag). He gets the lowest-friction, highest-value Ask — a diagnostic, not a demo. The other three are already in Continuous Use with adjacent tools; sending them to a generic "book a demo" wastes their committee weight.
The Ask is positioning's last mile. If positioning is the answer to "why us, why now," the Ask is the answer to "what's the smallest step that proves the rest." Each Ask above is calibrated to the smallest step that creates value AND moves the deal.
erom.io Subject of this reportBefore working with John, we were struggling to decide which market to focus on. Our platform serves multiple industries, and doing our own research was slow and unfocused. The Sprint gave us clarity, speed, and structure — the kind every startup needs.
John's real superpower isn't the tools or data — it's how he untangles the noise and frames everything through proven sales and discovery methodologies. The insights suddenly make sense in context; it's not just information, it's actionable intelligence.
The biggest shift for me as a founder was gaining a glimpse into how our customers actually think about us. We learned who really champions our solution internally — not just the COO we'd been targeting, but also the Chief Security Officer and Data Protection Officer, who feel the pain day-to-day. Now, our messaging speaks to what each role values most.
The ROI? Clarity and time. The Sprint saved us weeks of manual research and gave us a repeatable method to evaluate direction and fit. If you don't yet have crystal clarity on who you're targeting or why they buy — this Sprint will give you that. It's like seeing what happens behind closed doors in your buyers' world."
April Dunford's Sales Pitch model splits in two halves: The Setup orients the buyer in the market — insight, alternatives, the perfect-world destination. The Follow-Through brings your solution into focus — introduction, value, proof, objections, ask.
Each block carries a job. Each job has a JTBD input that fills it with buyer evidence — or doesn't, in which case the block fills with vendor opinion.
This page took erom.io through all eight, with the Sprint outputs as the source material.