Scribe-X MMF V4

A single, authoritative source of truth that defines how Scribe-X explains, teaches, and scales the Clinical Intelligence Platform ... across people, platforms, and AI systems.

This is not a traditional messaging document. It is Scribe-X's verbal identity and narrative operating system.

How this framework is used

• Strategic narrative framework: the core story, category definition, and point of view that shape how Scribe-X is understood.
• Messaging playbook: practical guidance for sales, marketing, leadership, and external communications.
• North-star messaging guide: the reference standard for training internal teams, vendors, and partners.
• Context engineering for AI: structured training data so AI systems reason, communicate, and prioritize in alignment with Scribe-X's principles and operating philosophy.

Tone, style, and output

This framework defines what Scribe-X believes, how it frames problems, and how it guides decisions. The expression of that content ... tone, style, format, and voice ... can and should be customized at the output level based on audience and use case.

The writing in this document is intentionally professional, clear, and direct. The goal is shared understanding, not persuasion through flourish. That neutral, disciplined style lets the framework be used reliably by sales teams, operators, executives, partners, and AI systems alike. Tone may evolve in downstream outputs (marketing copy, sales decks, AI-generated content), but the underlying narrative logic and meaning stay consistent.

Purpose

This framework articulates Scribe-X's transformational story, not as marketing language alone, but as a coherent system of beliefs, definitions, and decision logic. It keeps the narrative consistent across every customer-facing channel, aligned internally across teams and partners, and structured so AI systems reinforce the strategy rather than dilute it. It feeds the website, the Practice Health Scorecard, the MQL plan, and the AI Brand Twin. At its core, it defines how Scribe-X helps practices move from the old way (AI alone, stopping at the note) to a new way of winning (AI plus human intelligence, across the whole patient journey), and how that transformation is clearly articulated, consistently reinforced, and scaled.

---

Primary ICP

Enterprise-level FQHCs and their lookalikes ... larger, multi-site community health organizations, not small private practices.

• Why this vertical: roughly 1,600 FQHCs plus ~150 lookalikes (orgs without the designation, paid the same way). The barrier to entry is high, which is exactly why it becomes a moat once we're in. "If we had 150 of them, we'd be a $100M company."
• Why now: budgets are tighter every year, and these orgs are forced to think differently or they won't survive ... so they're finally open to change instead of riding the old playbook.
• Who we deprioritize (on purpose): health systems (longer cycles, heavier politics, real risk of stalled execution) and small practices (can't carry the model). Large private outpatient groups have the provider volume and the operational pain to carry the model ... a real secondary pathway, but enterprise-level FQHCs and their lookalikes are where we lead.

The buying committee (six roles, two jobs)

We sell into a buying committee, not a single buyer. Six roles matter, and they split into two jobs. The clinical leaders (CMO and CMIO) are the doors in ... they feel the pain first and carry us into the room. The economic buyers (CEO, CFO, and the COO / Chief Quality Officer) control whether dollars get spent and on what, with the CIO clearing risk and governance. We don't rank them ... a real FQHC deal needs the clinical champion AND the economic owners aligned. The three clinical doors below; all six in full in §6.

The cast around the buyer

---

---

Our Identity ... why Scribe-X exists

We exist so patients get the provider's full attention, and the care that follows the visit doesn't fall through the cracks. That's the point ... caring for patients is the action, and better health is the result. Everything else, the notes, the codes, the orders, the inbox, is the work around the visit that too often takes the provider away from the patient. We believe a clinician's time belongs with the patient, not the chart, and that the practice, the care team, and the patient are all better off when it does.

This is our mission layer, not our sales trigger. A buyer rarely arrives searching for our mission ... they arrive with a pain (a provider about to quit, revenue leaking, a stalled AI rollout). Identity is what we stand for once we're in the room, and what every piece of content ultimately rolls up to. It is not the hook. The hook is the pain. The tie-back is always the patient: more patients seen, better care delivered, a healthier care team, because the providers are freed to do the work only they can do.

Foundational Belief ... "Providers want patient time, not paperwork time."

"Providers" is shorthand for the clinicians a practice hinges on ... MDs, DOs, NPs, PAs.

We believe what these clinicians want is, in good faith, what's best for the patient ... because caring for patients is why they went into medicine. Giving a provider their patient time back isn't a perk for the provider. It's the patient getting cared for. Same thing.

Paperwork steals that time. Most of these practices already turned on an AI scribe to get that time back. It helped with the note and left the rest, so the charting still follows them home ... the "pajama time" they put in at 10pm and on the weekend just to close their notes ... and the inbox never empties. The providers are still maxed out, and that's what burns the good ones out until they quit. Then the patient gets less of their provider, the practice loses revenue and sees fewer patients than it could, and the follow-ups that get dropped start putting patient care at risk.

AI alone won't give the time back ... it speeds up the note and leaves the rest. It takes a Clinical Intelligence Platform: one system that carries the whole patient journey ... before, during, and after the visit ... so the provider's time goes where it belongs. With the patient.

---

Position 1 ... "AI alone isn't enough."

They've stopped believing an ambient or autonomous scribe can carry the work by itself. Usually they learned it the way it really happens: the AI they bought isn't broken, it's just "good enough" ... and good enough is quietly costing them. A provider 300 notes behind. A provider ready to quit. The codes that stopped getting captured. The quality gaps that are missed. Because AI alone isn't enough to handle the hard cases, go past the note (orders, the inbox, care coordination, the whole journey), capture the revenue (missed codes leak reimbursement), or be trusted with the judgment (people forgive a human's mistake and can't forgive the AI's). Earns "AI + Human Intelligence," not AI only.

The tell: they tried ambient AI and felt it fall short, or they're already wary of "automate the provider." If they believe AI will simply keep improving until it replaces the work, they're not ready ... they're a top-of-funnel education target.

Position 2 ... "The real job is the whole patient journey, not just the note."

Documentation relief is a symptom. The prize is owning the journey end to end ... chart prep, the visit, orders, the inbox, care coordination ... and the people problem around it: the expensive clinician bookended by transient, churning staff who don't set them up to succeed. "Notes done faster" buys a commodity scribe; the journey is a platform. Earns the whole-journey scope of the CIP.

The tell: they talk in journey and operations terms (chart prep, inbox, retention), not just charting speed. These leaders already think this way ... the patient journey is in their strategic plan.

Position 3 ... "The clinical and legal risk never leaves the clinic ... so put human intelligence on reducing it, not AI on adding to it."

They know the clinical and legal risk can never be outsourced ... it's always the clinic's, and liability is shifting harder onto the provider who used the AI. The smart move, then, isn't the cheapest tool; it's human intelligence actively reducing the risk they'll always carry ... owning the accuracy, catching the misses, keeping the data onshore ... instead of artificial intelligence on its own, quietly adding to it. Earns the human at the center and the toggle bar: the more human intelligence you set into the work, the more of the risk you carry gets reduced.

The tell: they ask "who's accountable / what's the risk / where's our data handled," not just "what's it cost." The opposite belief ... "good enough, just give me the cheapest" ... is our Adversary, specifically the CFO (see §6 for the CFO/CEO distinction).

---

Three states, not two. The OLD way was fully manual ... human-only, before AI touched the visit at all. Nobody defends it anymore. The CURRENT way is AI-only: a clinic puts an AI scribe on the encounter to draft the note and assumes the practice's problem is handled. It isn't. AI alone covers one step, the note, and the rest of the work around the visit stays manual or undone: chart prep before, the orders and codes during, the referrals, the inbox, and the follow-ups after. The note gets faster and little else changes ... "good enough," but not enough.

The NEW way completes the whole process around the visit, before, during, and after, with AI carrying the volume and a Remote Healthcare Assistant owning the judgment and closing the loops. The clinic sets how much human, per provider, with the toggle bar. That is what actually frees the provider, because the work around them is genuinely handled, not just the note auto-drafted.

The shift in one line: from AI-only covering the note and assuming the rest follows, to completing the whole process around the visit ... AI for the volume, human intelligence for the judgment, context, and follow-through.

Current way · AI only	What matters most	New way · AI + Human Intelligence
Note	Scope	Journey
None	Judgment	Human
Leaks	Revenue	Captured
Dropped	Follow-ups	Closed
Generic	Fit	Configured
Stalls	Adoption	Sticks
Added	Risk	Reduced
Paperwork	Time	Patients

Why the current way is breaking now

Two forces are ending it. The clinics that tried ambient AI have hit its limit and felt what it leaves undone. And the accountability for an AI note is landing on the provider who used it, the same provider the practice is fighting to keep. States are already writing that into law. The tool that was supposed to help is quietly becoming a new exposure on the person you can least afford to lose, so "set it and forget it" is now a risk, not a shortcut.

---

A practice where the provider's day belongs to the patient again.

Picture a day in a practice that got this right.

Before the visit, the chart is already prepped. The recent labs, the relevant imaging, the gaps in care are pulled up and waiting, so the provider walks in knowing the patient instead of scrambling to catch up.

During the visit, the provider is with the patient. Eyes up, not buried in a keyboard. The note, the orders, the coding, the endless EHR clicks are handled around them, not by them.

After the visit, the loop actually closes. The note is finished before the provider leaves the room. The referral goes out with what the insurer needs to approve it. The abnormal lab gets its callback. The patient doesn't bounce back in three months because something got missed.

When the day runs like that, everyone is better off at once:

• the provider practices at the top of their license instead of being a paper jockey, leaves the pajama time behind, and stays in medicine;
• the care team isn't constantly covering for what didn't get done, and can actually support the visit;
• the practice sees more patients, keeps its people, and stops leaking revenue to the codes that used to get missed;
• the patient gets a clinician who is present, and care that doesn't get missed.

That is the whole point of the Clinical Intelligence Platform. Faster notes are the smallest part of it. The promised land is the whole patient journey, handled end to end ... so the practice runs the way the provider always wanted, and the patient feels the difference.

---

We sell into a buying committee, not a single buyer. Six roles matter, and they split into two jobs. The clinical leaders (CMO and CMIO) are the doors in ... they feel the pain first and carry us into the room. The economic buyers (CEO, CFO, and the COO / Chief Quality Officer) control whether dollars get spent and on what, with the CIO clearing risk and governance. We don't rank them ... we build the case for every one, because a real FQHC deal needs the clinical champion AND the economic owners aligned.

The providers underneath the buyer (who they're buying for → the solutions, §10-§11)

The buyer doesn't set one level for everyone ... each provider is set to the solution they need on the toggle bar: how much human intelligence their work needs, and how much risk the clinic wants handled. Three kinds:

• Pulse (left of the bar · AI only): the cost-conscious or tech-comfortable provider who's fine owning more of the workflow. AI drafts from the encounter with minimal human review; they accept the most risk, because they figure they and their staff can catch whatever slips. (Pulse is the AI-only floor ... grayed out until it's operationally ready. We show it; we don't promote it.)
• Assist (middle of the bar · AI + human): the provider who's been burned before and wants AI efficiency with a trained Remote Healthcare Assistant's review where quality and context matter ... usually younger, comfortable with technology, the ones who'll be running these clinics in a few years. The biggest untapped group, and the most likely to move. Available now.
• Live (right of the bar · real-time human): two faces ... the high producer you don't dare slow down, and the senior clinician who just wants to see patients and never touch the computer. A remote medical scribe supports documentation live during the visit. Both want the most human, both have the least tolerance for a critical miss, and both pay for the assurance. Available now.

---

---

Nearly every practice saying these things already has an AI scribe running. This is the pain that's left after the AI did its narrow part.

The injustice

The grievance that turns these problems into a fight worth having.

• You should never have to choose between being present with your patient and finishing your notes.
• You should never have to lose a great provider because the paperwork beat them.
• You should never have to settle for "good enough" and call it a strategy.
• You should never have to leave revenue you earned on the table because a code got missed in the rush.
• You should never have to wonder whether a dropped follow-up put a patient at risk.
• You should never have to tell your best people to just absorb the pain because the budget said so.
• You should never have to defend a technology investment you can't actually measure.
• You should never have to choose between growing the organization and protecting the people already in it.

---

Standing still feels safe. It isn't. Leaving the work around the visit half-handled doesn't hold steady ... the cost compounds, quarter after quarter, across the provider, the practice, and the patient. And it's countable. Here's what the industry data says each piece costs.

• You lose your best people, and replacing them is expensive. Burnout takes the most conscientious providers first. They already spend close to two hours on the record for every hour with a patient, and the providers doing the most after-hours charting are about twice as likely to burn out. Burnout roughly triples the odds a provider is looking to leave, and the AMA puts the cost of replacing one physician at $500,000 to $1 million once you count recruiting, lost billings, and the ramp. A family-medicine seat now takes about five months to fill, and every month it sits empty is production you don't get back.
• Your providers can leave for a practice that fixed it. The burned-out provider has options. As more organizations pair AI with human intelligence and hand the day back, the practice still running its providers into the ground is the one they leave ... and the patients who counted on that provider lose their access along with them.
• Revenue keeps leaking, mostly through the codes. Every under-coded or missed visit is reimbursement you earned and didn't collect. A federal audit found that miscoded and under-documented Evaluation and Management visits drove $6.7 billion in improper Medicare payments in a single year, with about 42% of those claims coded wrong. And the prevention and coordination work your providers already deliver often never gets billed ... modeling in the Annals of Internal Medicine put that at around $40,000 a provider a year left on the table. It never shows up as a crisis. It comes off the bottom line every month.
• Patients pay for the dropped loops. A lab with no callback, a referral that never went out, a follow-up that slipped ... these become repeat visits, worse outcomes, and real risk to patient safety.
• Patients wait longer for less. As providers fall behind and capacity tightens, appointment slots shrink and wait times grow ... patients get less access to their provider, not more.
• The money you already spent stays wasted. The AI pilot that didn't stick was paid for. Leave it at "good enough" and you keep paying for a tool most providers don't use, while the problem it was meant to fix sits frozen.
• The records become a liability. Notes that AI alone got thin or wrong are a compliance exposure ... inaccurate documentation the clinic is accountable for, in a climate where that accountability is landing harder on the provider who used the AI.
• The margin for error is already gone. Community health centers ran an aggregate operating margin of about 1.6% in 2023, and nearly half operated at a loss. At that thinness, lost visit revenue or added labor cost doesn't dent the budget ... it decides whether the center ends the year in the black. Budgets get tighter every year, and the liability keeps moving onto the provider who used the AI. Doing nothing isn't holding position ... it's falling behind a problem that's getting harder.

Put a number on it, for the finance conversation

The champion can turn this into a figure the CFO can't wave away, using the practice's own numbers:

• The providers you're at real risk of losing, times $500,000 to $1 million to replace each one.
• The visits a day your providers can't get to, times your per-visit rate, times the days you're open. For scale, the Medicare FQHC base rate alone is about $190 a visit, and most FQHC visits are reimbursed below what the care costs, so lost throughput hurts twice.
• The prevention and coordination work already being delivered and never billed, around $40,000 a provider a year, times your providers.

Add those up, and you have what the cheapest tool is quietly costing the practice. That's the number to set the price against.

The question was never whether to change. It's whether to change while it's still cheap to fix, or after the cost has compounded. And this isn't only getting more expensive. The accountability for AI notes is landing now, on the providers you can least afford to lose, in the states that already passed the laws.

The stakes, by role

The cost of doing nothing doesn't land evenly. Each role carries a different piece of it, and it compounds differently for each.

• CMO. Every quarter this stays half-handled, you lose more of the providers you can't replace, and the ones who stay carry a heavier load. The burnout you're tracking today becomes the resignation letter next year.
• CMIO. The tools already in place keep adding review work instead of removing it, and adoption erodes further with every rollout that doesn't stick. The governance question gets harder to answer, not easier.
• CIO. Every month of ungoverned AI use is another month of undocumented risk sitting on your books. When the audit or the breach comes, "we were still figuring it out" is not a defensible answer.
• CEO. Standing still doesn't hold your position, it cedes it. The organizations that move now set the standard the rest of the market gets measured against, and yours falls further behind the longer the decision gets deferred.
• CFO. The revenue that leaks this quarter doesn't come back next quarter, it compounds. Every month without a fix is a month the budget absorbs a cost nobody's put a number on yet.
• COO / Chief Quality Officer. Quality measures don't fail all at once, they slip a little every cycle until the org is chasing a gap it can't explain. Coverage that depends on no one calling in sick isn't a plan, it's a countdown.
• Provider. Every night this stays unfixed is a night that comes home with you. The providers who leave first are usually the ones you could least afford to lose.
• Patient. Every dropped follow-up is a patient whose outcome depends on nobody noticing what fell through. The cost of doing nothing doesn't show up on a balance sheet ... it shows up in someone's chart.

---

Every other tool forces the same question: AI, or no AI? The platform answers a better one ... how much human intelligence and judgment do you want in the work, provider by provider? That's the whole mechanism. One platform, one toggle bar ... and you decide which responsibilities to take on, and which to hand off.

The toggle bar, left to right

Set a provider on the left, at Pulse, and AI carries the documentation while the clinic keeps everything else in-house ... the AI-only floor, coming when it's operationally ready. Set them in the middle, at Assist, and AI drafts while a trained Remote Healthcare Assistant reviews the work and carries the highest-stakes stages ... the coding, the orders that drive revenue. Set them on the right, at Live, and a remote medical scribe supports the visit in real time, with human intelligence at the center of the whole journey, so the clinic hands off the most risk and the most manual work. Same platform. What changes is how far right each provider is set. There's no second product to buy and no rip-and-replace to move up.

It runs across the whole journey

The toggle bar doesn't sit on the note alone. It runs across the entire journey, organized as intelligence layers ... the platform's capability stacks as the clinic moves right, and we mark what's live and what's expanding honestly:

• Pre-Visit Intelligence. The provider walks into the room prepared ... chart prep, prior visit history, open orders, medication context, quality gaps, and outside records surfaced before the visit starts.
• Encounter Intelligence. The note, the after-visit summary, and the orders discussed in the room, captured as the visit happens, around the way the provider actually works.
• Revenue Intelligence. The diagnosis and billing codes captured accurately and completely ... documentation specificity, HCC visibility, charge capture ... so the practice collects the reimbursement it actually earned.
• Quality Intelligence. The care gaps caught across the panel ... preventive reminders, UDS/HEDIS visibility, record quality, and high-risk patient signals flagged before they're missed.
• Expanding next (mapped, marked future): Inbox Management, Care Coordination, and Operations. They're on the journey map and grayed until they're real ... we show the whole map and our position on it honestly.

Set per provider, not just per clinic

The three settings are three named solutions ... Pulse, Assist, and Live ... detailed in §11. They aren't three different products; they're three points on one toggle bar, on one platform, so a clinic can start on the left and move right without switching tools. And the setting is chosen per provider, not just per clinic. A practice can run its highest-volume providers on Live and the rest on Assist, all under one agreement ... a level of fit no single fixed tool offers.

The mix ... any number of providers, at any setting

A buyer never has to pick a lane. The organization distributes its whole roster across the bar ... Pulse for some, Assist for most, Live for the few who need it ... and changes any provider's setting as needs change. Jason's framing, from the July 16 working session: a clinic could pick Pulse for 10, Assist for 20, and Live for 5. That's the model. Meet every provider where they are, and the whole roster lives on one platform ... which is what turns Scribe-X from a one-off solution vendor into the platform that penetrates the whole account.

One discipline rule rides with the mix, always: results attach to each provider's solution, never averaged across the platform. A clinic evaluating outcomes sees what Assist delivered for the providers on Assist and what Live delivered for the providers on Live ... no blended claims.

The canonical mix visual. On the homepage it appears twice: whispered in the hero (small provider facepiles on each toggle-bar card) and taught in the solutions section (the interactive "Build your mix" board, where a buyer toggles provider counts up and down per solution and watches the roster take shape). Every provider figure carries the medical cross ... these are clinicians, not seats.

Why a competitor can't copy it

Most competing tools are a single fixed setting: AI only, no human, the note and not much past it. This is the only platform that runs the full range, and the only one with the people and the process built in to run the right side of the bar. A tool designed around the algorithm can't bolt a human onto it later and call it the same thing ... the human structure is the product, and it has to be built in from the start. That's the moat. It's not a feature they're missing. It's a different model.

---

That's the mechanism. Next, the three solutions it produces, in detail.

---

Three solutions on one platform. A clinic doesn't assemble features one by one ... each provider is set to a solution, and what changes across the three is how much human intelligence stands in the work and the service level behind it. And the counts are the clinic's to set: any number of providers at each solution, mixed freely under one agreement ... 2 on Pulse, 6 on Assist, 2 on Live, or Jason's working example, 10-20-5 ... and moved between solutions as needs change, with no migration (see The Mix, §10).

The intelligence layers the solutions package

The layers run in journey order, and each carries its honest status ... live, packageable, or coming. Each solution packages them differently ... this is what "configuration" actually configures, and showing the whole map with our real position on it is the brand.

• Pre-Visit Intelligence (packageable add-on · built into Live). The provider walks in prepared. Chart prep, prior visit summary, problem-list and medication review, and open orders and labs surfaced before the visit ... with quality-gap, preventive, and HCC prompts coming behind them.
• Encounter Intelligence (live today · the core). More of the visit with the patient, less time managing the record. SOAP note generation, after-visit summary, mobile capture, provider style and specialty customization, EHR workflow fit, multilingual support, and assisted EHR delivery ... with dictation capture and order staging scoped next.
• Revenue Intelligence (scoped add-on · visibility first, advanced coming). The practice captures the value already created in care. Today: coding completeness support, documentation specificity prompts, HCC visibility, level-of-service support, and charge capture visibility. Coming: ICD-10/CPT suggestions, charge validation, denial-risk flags, prior-auth support, and revenue impact analysis.
• Quality Intelligence (scoped add-on · visibility first, advanced coming). The gaps that affect patient care and quality programs, caught before they're missed. Today: care gap visibility, preventive reminders, problem and medication list accuracy, and UDS/HEDIS measure visibility. Coming: record quality scoring, high-risk patient dashboards, registry worklists, and SDOH visibility.
• Side layers (future · scaled programs first): Inbox Management ... the next likely focus ... then Care Coordination (referral support, follow-up tasks, recall workflows) and Operations (productivity reporting, SLA monitoring, capacity visibility). Mapped on the journey, grayed until they're real.

How the solutions are sold

• Packaged, not à la carte. A clinic buys a configuration, not a menu of features. The support process behind each solution is what makes it work, and cherry-picking features breaks that.
• The line between solutions is service level, not just feature count. What a clinic really buys as it moves right is how much human accountability stands behind the work.
• Set per provider, under one agreement. A practice can run a few high producers on Live and the rest on Assist, all on one contract. Providers mix solutions; the agreement stays one agreement.
• Pricing is configuration-based. No fixed pricing appears in any asset ... the public answer is "Contact Scribe-X," and the price conversation happens on the call.

---

The buyer isn't choosing between Scribe-X and nothing. They're weighing a short list of real options. Name them honestly, and the gap shows itself.

What a clinic is choosing among

• Do nothing. Keep the manual process and the pajama time. It costs nothing today, and the cost compounds quarter after quarter (see §9).
• An ambient AI scribe. The point solutions ... Abridge, Nuance, and the like. They draft the note, and they do it well. They also stop at the note ... chart prep, coding, orders, referrals, the inbox, and the follow-ups all stay manual. And it's one more vendor to integrate.
• The EHR's own built-in AI. Epic turning on ambient AI as part of the EHR deal ... the Abridge and Nuance partnerships, with Epic's own tools coming. This is the real competition, because clinics want fewer vendors and the EHR already sits in the workflow. It's the same AI-only note, bundled. It often looks free, and the ROI math assumes the provider sees more patients per day to pay for it ... which takes time away instead of giving it back.
• Offshore or virtual human scribes. Cheaper hands on the documentation. But the patient's record leaves the country, which is a compliance exposure some states already restrict, and there's turnover and no AI carrying the volume.
• Hire more in-house staff. More medical assistants, scribes, and coordinators. Expensive, and it deepens the very problem ... a churning, low-paid layer around an expensive clinician, with the clinician still stuck in the middle.

What only Scribe-X does

• Covers the whole journey, not just the note. Chart prep before the visit, the note during, and the orders, coding, referrals, inbox, and follow-ups after. This is the clearest single differentiator ... and the free trial is where a clinic feels it first.
• Pairs AI with human intelligence. AI carries the volume; a trained, US-based Remote Healthcare Assistant owns the judgment. We don't win by being more AI than the AI companies ... we win by completing what AI alone leaves undone.
• Configurable on the toggle bar, per provider. A complement to what providers already use, set to each one's risk tolerance, with no rip-and-replace. No competitor offers the full range.
• US-based Remote Healthcare Assistants and onshore data. Scribe-X owns the accuracy of the work product, catches the misses, and keeps the patient's record in the country ... reducing the risk the clinic always carries instead of adding to it.

On the EHR head-on

The EHR's built-in AI is real, convenient, and getting better, and it's smart to expect every EHR to keep building. We don't compete by trying to out-AI it. We complete the work it leaves undone. The note was never the hard part ... everything around it is, and that's the ground the bundled AI doesn't cover. Our own AI-only solution, Pulse (coming), is by design comparable to what's already in the market ... we put it on the menu so the buyer can see the floor for themselves, then choose how far past it Assist and Live go.

---

What Scribe-X delivers

The proof that matters most is our own. Across Scribe-X accounts, the whole-journey model shows up as more patients seen, better-captured revenue, and providers who are measurably happier.

• 1.9 more patients per day, on average ... and up to 11 more per provider per day in some settings. That's time handed back, spent on patients.
• A 30-point jump in provider satisfaction ... a 30 percentage-point increase against pre-Scribe-X and non-user surveys. This is the retention story in a single number.
• 10% more level-of-service coding ... visits coded to the level actually delivered, instead of leaking reimbursement.
• 24% better HCC scoring ... risk capture that funds the care these panels need.
• 45% faster time to close ... the note and the work around it finished in nearly half the time, so charts stop following the provider home.

What the old way costs

The burden these numbers relieve is well documented, and it isn't easing on its own.

• Two hours of documentation for every hour with patients. The paperwork is now the larger half of the job.
• 45.2% of physicians reported a burnout symptom in 2023 ... down from the 62.8% pandemic peak in 2021, but still above every reading taken before 2020. This is structural, not a blip.
• 77% of physicians report that a significant part of their time goes to non-reimbursable tasks ... documentation, prior authorizations, denials, and payer communications. The work doesn't disappear when the visit ends ... it follows the clinician home.

Where AI alone stops

Physicians want AI, and the independent evidence is that AI by itself moves the needle only modestly ... and only when people actually keep using it. This is Position 1 with the numbers behind it.

• The demand is real. 81% of physicians report using AI professionally, 76% say it can help patient care, and seven in ten see it as a way to automate the tasks that drive burnout.
• The gains from AI-only scribes are modest. A multisite JAMA study of more than 1,800 AI-scribe users (against 6,770 controls) found about 16 fewer minutes of documentation a day and 13 fewer minutes in the EHR ... a 10% cut in documentation time, 3% of total EHR time, and just 0.5 more visits a week. Revenue moved a real but nominal $167 per clinician per month.
• And the gains depend on adoption that mostly doesn't happen. Only 32% of clinicians used the AI scribe in more than half their visits. Frequent users saw two to three times the benefit ... which is the whole point: the tool only pays off when a person keeps using it, and most don't.

The market we're moving in

• Ambient AI is already everywhere. An estimated 80-85% of providers now use some ambient note solution, depending on specialty and setting. The question in the room is no longer "AI or not" ... it's whether the AI they turned on is actually enough.
• The room to grow inside our own accounts is large. About 15% of providers in current Scribe-X accounts use live scribes, which means roughly 85% are on other ambient tools or nothing at all ... the clearest expansion path we have.

---

A buyer who understands the offer still has real questions. Here are the ones that come up, and how we answer them straight.

• "Isn't this just another AI scribe?" No. AI carries the volume; the product is the human intelligence and judgment on top of it, across the whole journey. The note is the smallest part of it. If a note is all a clinic needs, the market already sells that ... and so will we, as Pulse, the AI-only floor.
• "Our EHR already gives us ambient AI, close to free." It does, and it's convenient. It's also the AI-only note, and the ROI math behind it assumes your providers see more patients to pay for it ... so it takes time rather than gives it back. We finish the work it leaves undone: chart prep, coding, orders, referrals, the inbox, the follow-ups.
• "We tried AI before and it didn't stick. Why is this different?" Because what didn't stick was AI alone, turned on and handed to the provider to figure out. Here a trained Remote Healthcare Assistant carries the work and owns the accuracy, so there's no rollout for a busy provider to get right. The thing that failed is the thing we replace.
• "It costs more than the tool we're comparing it to." On the sticker, it does. Set against what standing still costs ... the providers who burn out and leave, the revenue that leaks, the follow-ups that get dropped ... it's the cheaper choice (see §9). And the toggle bar lets a clinic take only the level it needs, per provider, so it never pays for more human than it wants.
• "Will our providers actually use it?" Adoption is our job, not theirs. Because a Remote Healthcare Assistant carries the work, there's nothing for a busy provider to use wrong or quietly abandon. That's the difference between a tool a clinic has to adopt and a service that simply runs.
• "Is our patient data safe?" Our Remote Healthcare Assistants are US-based and the record stays onshore. That's deliberate. In a climate where offshore handling is getting restricted and the legal liability is landing on the provider who used the AI, it reduces the risk a clinic carries instead of adding to it.

---

A buyer who's been burned by a stalled rollout needs the path to feel safe, not heavy. It's three steps, and we carry the load on every one.

Why the plan removes the risk

The failure a clinic fears ... another tool that's switched on and quietly dies ... can't play out the same way here. A Remote Healthcare Assistant carries adoption, so there's nothing for a busy provider to get right. The pilot proves it in a real clinic before anyone scales. And the clinic grows only as fast as it works, one provider and one toggle setting at a time. The decision is reversible at every step, which is exactly why it's safe to start.

---

That closes Phase 3 (Solution & Edge), with §13 Proof parked for data. Now Phase 4: the rebellion and the words to carry it. (The full Language Library now lives in the Appendix at the back.)

---

A rebellion line does one of two jobs. It either names the promised land (the new world, the way the game is now played) or it's a battle cry (verb-first, the call to arms). A brand can run both ... the promised land where it lives, the battle cry where it rallies. Above both sits the credo: the value everything rolls up under.

Naming the Promised Land ... the new world, how the game is played

• AI + Expertise  #AIPlusExpertise
  The new standard, and the punchy public form of the locked banner "AI + Human Intelligence." Implies AI alone is the lesser thing.
• Clinical Intelligence  #ClinicalIntelligence
  The world and the product share a name (the Clinical Intelligence Platform). The most ownable of the three.
• Co-intelligent Healthcare  #CoIntelligentHealthcare
  The category and thought-leadership banner for the movement ... blogs, keynotes, the POV ... not the homepage tagline. Names the AI-and-human era for a business-literate audience.

Battle Cry ... verb-first, the call to arms

• End Pajama Time.  #EndPajamaTime
  Kills the villain every clinician feels ... the 10pm and weekend charting.
• Free Up Your Providers.  #FreeUpYourProviders
  The relief, made personal to the practice and its leaders.
• Practice Medicine, Not Paperwork.  #PracticeMedicineNotPaperwork
  The provider's calling restored; rhymes with the credo for a deliberate anti-paperwork drumbeat.

---

That's the rebellion. Now the story made sayable.

---

The elevator pitch

Thirty seconds: Most AI scribes stop at the note, which leaves everything else around the visit sitting on your providers ... the chart prep, the coding, the orders, the referrals, the inbox, the follow-ups. That's why they're still charting at 10pm, why your best people burn out, why earned revenue leaks through missed codes, and why dropped follow-ups start putting patients at risk.

Here's what we do, with you: we complete the whole patient journey, before and after the visit, not just the note. AI carries the volume, a trained, US-based Remote Healthcare Assistant owns the judgment, and you decide how much human each provider needs. Your providers get their day back, and the risk you carry goes down instead of up.

How are you using AI in your practice?

Ten seconds (lead with the question): How are you using AI in your practice? Most stop at the note, which leaves your providers doing the chart prep, the coding, and the follow-ups ... and it's burning them out. We finish the whole patient journey with you: AI for the volume, a Remote Healthcare Assistant for the judgment, configured around each provider.

Five seconds (lead with the question): How are you using AI in your practice today? Most AI scribes stop at the note ... we finish the whole patient journey, AI plus a Remote Healthcare Assistant.

Opening hooks for outreach

Most of these are built to draw a "No," not a "Yes." Pushing someone toward "Yes" puts them on guard; a "No" feels safe and in their control, and the honest "No" makes them name the situation they can't keep tolerating. That admission is the starting point for action. Then an open "what / how" question widens it, getting them to say the stakes out loud, in their own words.

• Pajama time: "Have your providers stopped finishing notes at night?" No ... the 10pm charting is still here.
• The whole journey: "Did your AI scribe take the chart prep, the coding, the orders, and the follow-ups off their plate too?" No ... it only handled the note.
• Retention: "Are you confident you won't lose another good provider to burnout this year?" No ... and that's the one loss they can't afford.
• Revenue: "Are you certain you're capturing every code you've earned?" No ... reimbursement is leaking every month.
• Productivity: "Are your providers able to see as many patients as you'd like?" No ... they're struggling to keep up with what they have.
• Liability (for the CIO): "Have you settled who's accountable when the AI gets a note wrong?" No ... and the liability is shifting onto the provider who used it.
• The cost of standing still (the open follow-up): "What are the ripple effects, for you and for the practice, if this stays the way it is?" An open question that makes them name the stakes themselves ... burnout, turnover, leaked revenue, dropped follow-ups (ties to §9).

Calls to action

Low-friction and value-first. Lead with the scorecard or the free trial, not a demo.

• See your gaps: "Run the Practice Health Scorecard ... your gaps and what they cost, in minutes." (the front door)
• Feel the win: "Start a free trial ... a few providers, in your real workflow. You tell us what changed."
• See it in action: "Let's get you into a pilot so you can see the whole journey in action ... a few providers, clear KPIs."
• Map the journey: "Book a call and we'll map your patient journey together." (for larger, multi-site organizations)

---

That's the language made sayable. Now the category we plant a flag in, and how we get found.

---

The category we lead

"AI scribe" is a crowded category, and a losing one to fight in. It's a race to the cheapest, the EHRs are bundling it for free, and it stops at the note. Competing there means being compared on price for a commodity. That's why we don't fight there. We define and lead a bigger category.

That category is the Clinical Intelligence Platform (CIP). It's bigger than scribing: the whole patient journey, AI for the volume, a Remote Healthcare Assistant for the judgment, set on a toggle bar. The old category, the AI scribe, becomes the floor we offer (Pulse, coming) ... the Clinical Intelligence Platform is the category we want to be known for and measured by.

SEO & AEO terms

Scribe-X already has real search equity in the medical-scribe family ... the site ranks for terms like "medical scribe services" and "remote medical scribe" (currently around page two on the core term). That equity is an asset, not a legacy to abandon. Medical scribing is one capability under the Clinical Intelligence Platform, so the play is to build on what's ranking and bridge it up to the category, not start over.

Four buckets, from proven to aspirational:

• What's already ranking (protect and grow): "medical scribe services" · "remote medical scribe" · "real-time remote scribe" · "hybrid AI scribe" · "medical scribe company / partner." Scribe-X.com's proven equity ... keep these strong, and frame the scribe as one capability of the platform.
• Broader existing demand (capture): "AI medical scribe" · "ambient AI scribe" · "AI scribe for FQHC / community health" · "AI medical coding" · "chart prep automation" · "physician documentation burden." Where the wider volume is, even though it's the old category.
• The category we're seeding (own the new term): "Clinical Intelligence Platform" · "human-led AI scribe" · "AI plus human intelligence (healthcare)" · "human-at-the-center medical AI" · "whole patient journey documentation."
• High-intent pain searches (long-tail, ready to act): "providers charting at night" · "AI scribe didn't work / didn't stick" · "missed medical codes lost revenue" · "FQHC provider retention / burnout" · "who is liable for AI medical notes."

---

That closes Phase 4 (Language). Now Phase 5: the go-to-market idea banks, starting with the sales playbook.

---

The motion

• Open with value, not a demo. The Practice Health Scorecard and the free trial are the front doors.
• Run the plan: Discover → Pilot → Expand (see §15). The pilot is where they feel it.
• Multithread the buying group: the CMO champions it, the CMIO usually supports, the CIO clears the risk, the CFO is the cost-first wall you get past through the champion armed with numbers, and the CEO is won separately on strategic and community grounds (see §6).

Plays by persona

Discovery questions that move a deal

Open with a no-oriented question (§17), then size the pain and the stakes.

• "How are you using AI in your practice today?" (the signature opener)
• "What happens to the orders, the coding, and the follow-ups your AI scribe doesn't touch?"
• "What did your last AI rollout actually change?"
• "Who owns the accuracy when the AI gets a note wrong?"
• "If you could set how much human each provider gets, who'd need the most?"
• "What are the ripple effects, for you and the practice, if this stays the way it is?"

---

Next: §20, Lead Magnets & Scorecards.

---

Lead magnet ideas

Gated, value-first, each tied to a pain we already named. Every one captures an email and feeds the pipeline.

• The Practice Health Scorecard (flagship): a few questions in, a score out ... where the workflow breaks, what it's costing, and the next steps. Email-gated, results to the CRM. The hero's secondary CTA.
• "Cost of doing nothing" calculator: estimate the annual cost of burnout, turnover, and leaked revenue (ties to §9).
• Whole-journey gap assessment: where work goes undone before, during, and after the visit.
• FQHC benchmark report: how community health centers are pairing AI with human intelligence (industry data, feeds the AEO play and the webinar).
• Compliance check (for the CIO): "Is your AI documentation a liability?" (ties to the liability research).
• Pajama-time self-check: a quick, provider-facing read on after-hours charting.
• The "5 ways" guide: the webinar's content as a download (ties to §21).

The flagship: the Practice Health Scorecard

• What it does: a short interactive assessment, then a score plus the gaps, what they cost, and recommendations with next steps.
• Where it lives: the secondary CTA on the homepage hero (the primary CTA is the pilot / book a call).
• What it is: the Discover step productized (§15), and Deliverable 5 of this engagement.
• Status: parked. The scoring logic, the questions, and the wire-up (capture email, deliver recommendations, push to the CRM) get built when we render the homepage.

Naming the scorecard

All of these stay live possibilities ... we keep the menu and start with one for the homepage. Grouped by angle:

• Ties to the category: Clinical Intelligence Scorecard · AI + Human Readiness Scorecard.
• Ties to the practice: Practice Health Scorecard · Practice Efficiency Scorecard.
• Ties to the journey: Patient Journey Scorecard · Care Journey Scorecard.
• Ties to readiness: FQHC Readiness Scorecard · Documentation Readiness Scorecard.
• Button copy (can differ from the name): "Score Your Practice" · "Get Your Score" · "See Your Gaps."

---

Next: §21, Webinar & Event Topic Ideas.

---

The anchor: Episode 1 (live)

The pattern (so the team can keep generating)

A provocative truth about the risk or the problem, then a twist line. Underneath, "5 ways [this audience] pair AI with human intelligence," ending in the payoff that audience cares about. Same spine every time; swap the audience and the payoff.

More episodes, by audience

---

Next: §22, the Blog Engine.

---

What the engine is for

• One post a day, in Scribe-X's voice ... it inherits the Language Library and the whole MMF.
• Two goals: seed thought leadership in the FQHC and AI-documentation space, and the bigger one, be the source answer engines (Claude, ChatGPT, Gemini, Perplexity) pull from when an ideal buyer asks about AI scribes, documentation, FQHC operations, or liability.

Why structure matters (the citability thesis)

Answer engines preferentially quote content that's well-structured, answers the question directly, and carries original facts and figures. Every post is engineered to be quoted, not just read. Data isn't decoration here ... it's the thing that gets cited.

Anatomy of a citable post (the agent's output skeleton)

• TLDR up top: one or two lines that answer the post's question outright ... the quotable line.
• Key takeaways: three to five bullets right under the TLDR, each self-contained and grab-able.
• The direct answer first: the core question answered in the opening, before any wind-up.
• Original facts and figures: freshly researched for that post, each with a citation. Every claim that can carry a number, does.
• Clean structure: question-shaped H2s and H3s, one idea per section, scannable.
• Concrete and grounded: examples from the FQHC and ICP world, never generic.
• A short FAQ block at the end: question-shaped, matching how people actually ask an engine.
• A soft close: a relevant lead magnet (the Practice Health Scorecard) or next step.

Operating rules

• Research fresh every post. Pull current data and cite it; don't recycle stale stats.
• Map every post to an ICP pain, a belief, or an objection. No adjacent fluff.
• Inherit voice and positioning from the MMF: human at the center, the whole journey, no AI-Parmesan, the Language Library rules.
• Never fabricate data, never overclaim on liability (follow the Regulatory & Liability Context appendix), and never overclaim scope ... pair "whole journey" with the live-vs-coming split (see the scope-honesty guardrail).

---

AEO, not just SEO

SEO is about ranking in the links. AEO is about being the answer the engine gives, and cites, when a buyer asks "the best AI scribe for community health" or "an AI scribe with human review." Buyers increasingly ask the engine, not the search box. We engineer to be the answer it returns.

The pages to publish (about 10 to 20)

• A deep FAQ page: the canonical, question-shaped answers to every buyer question. The single most answer-engine-friendly asset.
• Comparison posts: "Best AI Scribes of 2026" and a "Top 5 AI Medical Scribes" piece. Comparison data is exactly what an engine grabs when asked to compare ... so we seed it with accurate, structured, fair tables.
• Definitional pages: what is an AI medical scribe, what is a Clinical Intelligence Platform, what is hybrid AI scribing.
• Use-case pages: the AI scribe for FQHCs, for specific specialties, for multi-site groups.
• The category page: the Clinical Intelligence Platform, defined and owned.

Citability principles (same DNA as the blog)

• Direct answers, question-shaped headings, structured data (tables and lists), facts and figures with sources, schema markup, and freshness.
• Own the category term (Clinical Intelligence Platform) and capture the existing-demand terms (the medical-scribe family, see §18).
• Be the most accurate, most complete, most structured answer on each question. That is what gets cited.

---

Next: §24, the FAQ.

---

What is the Clinical Intelligence Platform?
One platform that pairs AI for the volume with a trained, US-based Remote Healthcare Assistant for the judgment, across the whole patient journey ... not just the note. How much human each provider needs is set on a toggle bar, per provider.

How is this different from an AI scribe?
An AI scribe stops at the note. The Clinical Intelligence Platform completes the whole patient journey, with a Remote Healthcare Assistant owning the judgment and the work around the visit.

How is the CIP different from native AI documentation tools?
Native AI generates a note. The Clinical Intelligence Platform supports the full visit ... chart prep before, human quality review during, and follow-up after ... tailored to how each provider works. That's where documentation problems actually live.

We already use our EHR's built-in AI. Why add Scribe-X?
Built-in tools handle the encounter for providers who use them. They don't prep the chart, adapt to individual workflows, or support providers who won't drive the technology themselves. Scribe-X covers what happens around the note.

What role does AI play in the CIP?
AI accelerates documentation across the encounter. Scribe-X leads with human expertise, so every provider gets support tailored to how they work, not just a tool set to a default. The result is documentation that's accurate, complete, and built around the outcomes that matter to your organization.

What's the difference between Live, Assist, and Pulse?
They're the three solutions a provider can be set to on the toggle bar, based on how much human support their work needs. Live puts a remote medical scribe on the encounter in real time; Assist pairs AI with a Remote Healthcare Assistant reviewing the output; Pulse is autonomous AI with minimal review, and it's a future option, not active today.

Do all our providers have to use the same solution?
No. Providers are set to Live, Assist, or (later) Pulse individually, based on what each one needs, all under one agreement. Your providers don't all have to work the same way.

Are your Remote Healthcare Assistants US-based?
Yes. They're US-based and the patient record stays onshore, which reduces the risk a clinic carries rather than adding to it.

We tried an AI scribe and it didn't stick. Why is this different?
What didn't stick was AI alone, handed to the provider to run without support. Here a trained Remote Healthcare Assistant carries the work and owns the accuracy, so there's nothing left for a busy provider to adopt wrong.

What EHR systems does Scribe-X support?
Scribe-X works with all major EHR platforms, including Epic, Athena, NextGen, and eClinicalWorks ... over 20 EHRs. Scribe-X learns your specific platform and workflows rather than forcing a generic setup, and any edge cases are confirmed during discovery before anything is signed.

How long does implementation take?
Most organizations are live within a few weeks of contract signing. Your onboarding contact walks through the process before anything is signed, so timeline and go-live specifics are set upfront and there are no surprises on go-live day.

How much IT involvement is required?
Most implementations require standard EHR access provisioning and a BAA (Business Associate Agreement) review. Scribe-X handles the setup process end to end ... IT gets looped in where needed, not asked to carry the project.

Is the CIP HIPAA-compliant?
Yes. Scribe-X executes a BAA before any provider encounter, and the platform runs on HIPAA-compliant controls: encrypted in transit and at rest, with role-based access and audit logging. Detailed security documentation is available on request.

How does pricing work?
Pricing depends on configuration ... which providers are on which solution, the scale of the program, and any added intelligence layers. Contact Scribe-X to scope the right fit; the price conversation happens on the call.

Who is liable if the AI gets a note wrong?
The clinic always owns the clinical and legal risk. Scribe-X reduces that risk by putting an accountable, US-based Remote Healthcare Assistant at the center of the work ... the better question isn't the model's accuracy percentage, it's who's accountable when it's wrong.

What specialties do you support?
27 medical specialties, across FQHCs and for-profit healthcare centers.

How do we start?
Get your Practice Health Score, start a free trial on a few providers, or book a structured pilot with defined KPIs ... your contact scopes the right entry point during discovery.

What happens during a trial?
A small number of providers run on Scribe-X in their real workflow for a defined period, with success measured against KPIs you set going in. You see the actual output before committing to anything broader, and the account team walks the results with you at the end.

How fast can we get a program live?
Faster than hiring in-house. Scribe-X handles the recruiting, training, and EHR onboarding, so a program goes live in weeks, not the months an internal build takes.

What does ongoing support look like after we go live?
Your account team stays active after go-live, monitoring performance, supporting slower-to-adapt providers, and flagging issues early. We start from your definition of success and work backwards, and if something isn't working you have a defined contact path, not a ticket queue.

Can a Remote Healthcare Assistant handle orders, meds, and instructions?
Yes, and that's the point. The platform goes past the note: the Remote Healthcare Assistant enters the orders, medications, referrals, and after-visit instructions the provider directs, so the provider isn't stuck on data entry. The provider always reviews and signs.

What if a Remote Healthcare Assistant is out or unavailable?
Coverage is part of the service model, since you're working with a service, not a single hire ... a trained backup steps in, and scaled Live programs carry a 100% coverage guarantee. Your providers don't lose a day to it.

Can you help with value-based care?
Yes. The whole-journey work the platform carries ... closed loops, captured codes, caught care gaps ... is exactly what value-based and quality programs reward. A Remote Healthcare Assistant makes sure the documentation and follow-through actually support the measures you're accountable for.

Is this just transcription, or an AI-only tool?
Neither. Transcription turns speech into text and stops. An AI-only tool drafts the note and stops. The Clinical Intelligence Platform completes the whole patient journey, with a Remote Healthcare Assistant owning the judgment and the work around the visit.

We're already fairly happy with our current setup. Why look at this now?
Most organizations that look into this have an AI tool handling the note but not much else ... the chart prep, the coding review, the follow-ups still sit with the provider. If that isn't costing you anything visible yet, it likely will as volume grows or turnover hits ... worth a conversation before it does.

How do we build the internal business case?
Scribe-X can build a tailored ROI analysis around your provider mix and payer profile, so the program is revenue-generating from the start, not a cost center you have to defend later. Your champion also gets the cost-of-doing-nothing math (§9) to set the price against.

---

Next: §25, Use Cases.

---

• The FQHC losing providers to burnout. Pajama time and the unfinished work around the visit are pushing good clinicians out. The platform gives the day back, and the providers stay.
• The practice leaking revenue through missed codes. Reimbursement earned and not captured, month after month. A specialist owns the coding so the revenue actually lands.
• The clinic burned by an AI-only rollout. The tool was turned on and nobody used it. A human carries the work here, so it sticks where the last one stalled.
• The providers struggling to see more patients. Admin work and the endless checkboxes are still a burden despite AI alone. Provider support expands patient access and captures the revenue that comes with it.
• The multi-site group that needs flexibility. Different providers need different levels of help. The toggle bar sets each provider's solution under one agreement.
• The practice drowning in chart prep. Providers walking into visits cold, prepping their own charts at night. Chart prep done 48 hours ahead changes the whole day.
• The organization building next year's board case. Leadership needs to show the board a technology investment tied to measurable outcomes, not another line item. A tailored ROI analysis ties the program to retention, revenue cycle, and capacity in numbers a board can act on.
• The multi-site group standardizing quality reporting. Different sites document differently, and leadership-level reporting doesn't roll up cleanly. Consistent, leadership-level reporting gives one view across sites instead of a spreadsheet stitched together by hand.
• The organization deciding whether to build in-house or partner. Hiring, training, and managing an internal documentation team takes months and carries its own turnover risk. Scribe-X gets a program live in weeks, with recruiting, training, and coverage already built in.
• The FQHC preparing for a compliance or security review. The CIO needs a clear answer on where data lives and who's accountable before a reviewer asks. US-based Remote Healthcare Assistants and onshore data handling give a documented answer instead of a vendor promise.

---

Next: §26, Social Proof.

---

Provider quotes

• Capacity: "We are now able to see 4 more patients per day. We couldn't have done it without the scribes." ... Dr. Anna Lundeen
• Staying in medicine: "Adding a medical scribe to my team made the difference between closing my doors and practicing for five more years." ... Dr. Murlan G.

More on the site: Dr. MacDonald ("I walk out of here and my charts are done") and Greg Parker (multi-year client, praised COVID-era responsiveness). Pick by message: capacity, retention, or pajama time.

Named customers (verifiable, public case studies)

• Hometown Health Center ... rural FQHC, Newport, ME (case study).
• Foresight Health ... featured partner.
• Reach: FQHCs and for-profit centers across 27 specialties and 23 states, on over 20 EHR systems.

---

That completes the planned framework (§13 Proof parked for data). The Appendix below always stays last.

---

Language is where a strong narrative gets kept or lost. These are the words that carry the positioning, the ones that quietly undercut it, and the rules for how we write.

The words we use

• The model: "AI + Human Intelligence" · "human intelligence" · "a human at the center, not at the edge."
• The platform: "the Clinical Intelligence Platform (CIP)" · "the toggle bar" · the three solutions: "Pulse," "Assist," "Live" (Assist and Live available now; Pulse future, shown grayed). Solutions are positions on the toggle bar, set per provider.
• The people: "Remote Healthcare Assistants (RHAs)" ... trained, US-based, accountable. This is both the formal role title and the positioning term (adopted July 6, 2026, replacing "clinical specialists"). Spell it out on first use in any asset; "RHA" after that.
• The scope: "the whole patient journey" · "before, during, and after the visit" · "chart prep" · "close the loops" · "the work around the visit."
• The problem: "pajama time" · the "good enough" trap · "the work around the visit stays undone."
• The villain: "Set-and-Forget AI" (the idea we fight) · "AI alone isn't enough" · "AI-only" (the neutral name for the approach).
• The risk: "reduce the risk you carry" · "own the work product" · "US-based" · "onshore."

The words we avoid, and what to say instead

• "Human in the loop" / HITL → "a human at the center," or "a Remote Healthcare Assistant at the center." HITL implies the AI starts the work and the human reacts; we mean the reverse ... the RHA directs, the AI assists.
• "Humans" (for our people) → "Remote Healthcare Assistants." They're trained, experienced professionals, not a generic species.
• "Scribe" / "just a scribe" (for the offer) → "the Clinical Intelligence Platform," or "Remote Healthcare Assistants." "Scribe" alone undersells the whole-journey scope and sounds like the cheap offshore option we're not. (Exception: "remote medical scribe" stays correct inside Live. "Live Scribe" is retired ... the solution is just "Live.")
• "Essentials," "Professional," "Enterprise" (as tier names) → RETIRED July 6, 2026 (Jason + Greg). Say the solution name (Pulse, Assist, Live) or "configuration." The commercial commitment behind the work is scoped per agreement, never named as a tier in any asset.
• "Dial" / "slider" → "the toggle bar."
• "AI-powered," "AI-driven" → name what it actually does. Sprinkling "AI" on the message is the parmesan that hides a weak point.
• "Automate the provider" → "complete the work around the visit." We never automate the clinician; we finish everything around them.
• Anti-AI framing → "AI alone isn't enough." We fight Set-and-Forget AI (the idea), never AI itself.
• "No-risk trial" → "free trial." We lead with the free trial (flipped June 11, Jason's call); the pilot that follows is framed by its KPIs.
• "Cheaper, faster" (a tech-only pitch) → the value of human judgment where a miss costs the most. We don't win going tech-only.
• "Ambient AI" / "autonomous AI" (as our label) → that's the competitor category. We're the platform with a Remote Healthcare Assistant at the center. (Describing Pulse's mechanics as "autonomous AI drafting" is fine; branding Scribe-X with the villain's category name is not.)

How we write

• No em dashes. Use "..." or single hyphens.
• No antithetical parallelism. Avoid "it's not X, it's Y." State it in positive terms.
• No anaphora. Don't open three sentences in a row the same way.
• No manufactured rule-of-three. Group three things only when there genuinely are three.
• Concrete over metaphor. Name the real consequence ... lost revenue, fewer patients seen, a dropped follow-up ... never "wobble" or "rising tide."
• Plain and direct. Contractions always. Short, punchy lines mixed with longer teaching lines. Write the way a clinical leader actually talks.

---

1. Position statements & thought leadership

Voice: analytical, third-person, executive register. Long-form and unhurried ... this is the register for a reader who has ten minutes and wants to be convinced by argument, not moved by a slogan.

Structure: Position (the thesis, stated in bold up top) → Problem (grounded in outside data, not just our claims) → the Model (multiple parts, each introduced by a bolded topic sentence) → Outcomes (quantified, bulleted, and only at the end).

Signature moves: cites outside research with real, named sources to ground claims rather than asserting them; uses precise operational and revenue-cycle vocabulary (days to close, level-of-service coding, HCC scoring, chart preparation) instead of generic software language; every section opens with a bolded declarative sentence that is that section's whole thesis; closes with one synthesizing paragraph, not a punchy tagline.

House reference: the HITL Position Statement. Natural variation in long-form phrasing is welcome ... for UNLOCKED phrasing only. Locked terms (toggle bar, a human at the center, CIP, the solution names, Remote Healthcare Assistants) always appear in their one locked form; variation never touches them.

Do not: open with a stat or a question hook (that's sell-sheet or blog register); use bullets before the final outcomes section; drop a battle-cry tagline into the middle of the argument.

2. Sell sheets, conference collateral & flyers

Voice: fast, numbers-forward, visual-first. The thirty-second register ... a reader standing at a booth or skimming before a session starts.

Structure: mirrored Challenge and Solution pages, where every cost on the challenge side is answered by a return on the solution side in the same order; a roadmap or timeline visual; a closing page with exactly one call to action and contact information.

Signature moves: every claim carries a number next to it, not a paragraph; section headers follow the "[Topic] | [Subtitle]" pattern; trust markers appear as icon-plus-word combinations (HIPAA COMPLIANT, FQHC-TRUSTED); a visual side-by-side makes the AI-alone gap visible at a glance instead of arguing it in prose.

House reference: the Fiscal Health conference flyer (P&I Conference 2026) ... the standard for this format's STYLE only. Its stats and claims are outdated; reuse the structure, pacing, and visual-numeric density, never the content. Every number must clear the §13 / §26 proof guardrails, and the copy still passes the Anti-AI Writing Standard and the banned constructions ... stat-forward is a layout choice, not a license.

Do not: write a full sentence where a stat callout will do; bury the call to action; explain the mechanism in prose when a two-column comparison lands in half the time; put em dashes or "Coming Soon" feature grids in customer-facing pages.

3. FAQ

Voice: direct, warm, unhedged. Two sentences per answer on average, three at the most. The register for a reader who wants the specific answer, not the narrative around it.

Structure: the question phrased exactly as a buyer would ask it; the answer leads with the direct fact or contrast, no throat-clearing; many answers close with a concrete process detail or reassurance rather than a slogan.

Signature moves: addresses the organization directly ("your organization," "your onboarding contact"); frames outcomes in business terms; never stacks bullets inside an answer. Functional descriptions of the human element ("human quality review") may vary the phrasing ... but the locked people term, Remote Healthcare Assistant, anchors every answer where the people are named.

House reference: the CIP FAQ v2 document; §24 is the applied result.

Do not: hedge more than once per answer; open with filler; let an answer run past three sentences.

4. Blog posts & AEO pages

Citable, structured, direct-answer-first. Fully specified in §22 (the Blog Engine) and §23 (AEO) ... this entry cross-references those rather than duplicating them. The short version: TLDR up top, the direct answer before any wind-up, original facts with citations, question-shaped headers, and a short FAQ block at the end.

5. Case studies

Voice: narrative and concrete ... close to the position-statement register, but anchored in one real organization's specific numbers and quotes rather than a category-level argument.

Structure: Before (the specific problem, in their language) → What Changed (the solution, described operationally) → After (their numbers, their quote) → What It Means (the takeaway for organizations like them).

Signature moves: uses the organization's own words wherever possible, sourced and rights-cleared per §26; leads the headline with the outcome, not the mechanism; keeps the platform explanation brief ... the customer's proof does the persuading.

Do not: re-explain the full Mechanism section; use a customer name or quote without verified usage rights (§26 guardrail).

6. Sales playbook & internal enablement

Plain and tactical, with no persuasion built in ... written for the seller, not the buyer. Governed by §19; cross-reference rather than duplicate.

Choosing a register when a request doesn't name the format

If a request doesn't specify the output type, ask which one before drafting ... the same fact set produces very different copy depending on the register: a retention statistic becomes a bolded thesis sentence in a position statement, a stat callout in a sell sheet, a two-sentence direct answer in a FAQ, or a customer quote in a case study. The fact doesn't change. The register does.

---

Our buyers read B2B copy and LinkedIn all day, and they've learned to feel when a machine wrote something, even if they can't name why. The moment a sentence reads as generated, the trust we just built with the message drops. That's expensive for us, because the whole AEO play depends on being the source an answer engine cites and a clinical leader believes. Every piece runs through this pass first.

Vary the rhythm (the staccato rule)

A run of short, clipped sentences is one of the loudest AI tells. A single short line for emphasis is good and on-brand. A sequence of fragments is not. Merge them into fuller sentences using the connective tissue a person actually uses: and, but, because, which, so.

• AI-sounding: "End of clinic. Already running on empty. That's when the notes pile up."
• Human-sounding: "By the end of clinic they're already running on empty, and that's exactly when the notes pile up."

This is how to read the Language Library's "short punchy lines mixed with longer teaching lines": punchy single lines are welcome, but never three or four fragments stacked in a row.

Words and phrases we cut

Each of these shows up constantly in generated copy. Cut it, or replace it with the specific thing you actually mean:

• "delve into" → "get into," "look at," "dig into"
• "it's worth noting" / "in today's world" → just say the thing
• "transformative" / "game-changer" → name what changed
• "leverage" (the verb) → "use," "apply," "rely on"
• "robust" / "holistic" → be specific
• "seamlessly" · "at the end of the day" · "the bottom line is" · "it's no secret that" · "in conclusion" · "this is a reminder that" · "let's unpack" → cut them

Tells in the tone

Three tone patterns read as machine-written even when the words are clean:

• The balanced pivot. Acknowledging a thing and then turning on "but" in a tidy little seesaw: "Documentation is hard. But it doesn't have to be." Replace with something more specific and less symmetrical.
• The appended rhetorical-question closer. A perfectly formed question bolted onto the end to feel thoughtful. Carve-out: the no-oriented power questions in §17 ("How are you using AI in your practice?") are a deliberate sales device, earned and specific ... those stay. The lazy appended one goes.
• The motivational kicker. A last line that could live on a poster. If the closing sentence sounds like a LinkedIn inspirational post, rewrite it.

What human copy looks like (the positive test)

• Sentences vary in length and lean a little longer.
• Transitions feel natural, and starting a sentence with "and" or "but" is fine. (Not "So" ... sentence-start "So" stays banned by the locked voice rules.)
• The writer has a point of view, not just a neutral observation.
• Abstract claims are grounded in a concrete, FQHC-world detail.
• The voice is consistent without being perfectly calibrated.

The carve-out: slogans and credos are exempt

The no-antithetical-parallelism and no-anaphora rules govern body copy, not the rebellion lines. A locked slogan earns its contrast; a paragraph doesn't. These stay as written, and authors should not "fix" them:

• "Patients, Not Paperwork" (the credo)
• "Practice Medicine, Not Paperwork"
• "Most AI Scribes Stop at the Note. The Risk Doesn't."

If a line is a tagline, headline, or credo, the contrast is the point. Everywhere else, state it positively.

The detox pass (run before anything publishes)

• Read the full piece first.
• Flag every em dash, every run of fragments, every tidy three, every blocklist word, every balanced pivot, and every poster-line closer.
• Rewrite the flagged parts using the rules above.
• Read it aloud in your head. If it sounds like a podcast ad or a motivational speaker, revise.
• Confirm the voice is consistent and human start to finish, and that the message and the locked language didn't drift while you were cleaning it up.

By channel

• Personal LinkedIn (SDR, founder, individual). Lean more casual. Contractions throughout. First-person grounding helps: "Something I keep hearing from FQHC leaders..."
• Company page and brand copy. Same rules, more polish. Less "I," more "we" or third person.
• Email. Shorter reads more human here, but still no staccato. Vary sentence length inside the paragraph.

What not to do

• Don't overcorrect into run-ons.
• Don't bolt on slang or filler to fake casualness.
• Don't strip all structure. We use structure too, just not perfectly symmetric structure.
• Don't change the meaning or the locked message while humanizing the voice.

---

The legal trend: a licensed human must review the AI

• Texas SB 1188 (effective Sept 1, 2025): a practitioner may use AI for diagnosis or treatment only if they personally review all AI-generated content before the clinical decision, plus patient disclosure. Penalties run $5,000 to $250,000 per violation.
• California AB 3030 (effective Jan 1, 2025): generative-AI patient clinical communications need a disclaimer ... but communications a licensed provider reads and reviews are exempt. A direct legal reward for a human at the center.
• California SB 1120, the Physicians Make Decisions Act (effective Jan 1, 2025): when a health plan uses AI in utilization review, a licensed physician must make the medical-necessity decision, based on the patient's own clinical record.
• The scale: law firms tracking the space report roughly 40-plus bills across about 25 states in 2026, mostly requiring clinical oversight of AI and patient disclosure.

Offshore PHI: the US-based advantage

• Florida SB 264 (effective July 1, 2023): patient EHR data must be physically maintained in the continental US, its territories, or Canada ... including third-party and cloud vendors ... with a compliance affidavit signed under penalty of perjury. Stricter than HIPAA, which sets no geographic limit.
• HHS OCR warning: storing or processing ePHI overseas can raise HIPAA risk that the US provider cannot offload, because OCR has limited reach over foreign vendors.

Colorado (correct the common misread)

Colorado is not anti-AI-in-healthcare. Its AI Act (SB 24-205) was an anti-discrimination law that carved out provider-action healthcare AI, was repeatedly delayed, and was repealed and replaced in May 2026 with a narrower, more HIPAA-friendly version (effective Jan 1, 2027). Colorado's 2026 healthcare bills (HB 26-1139, HB 26-1195) mandate human review, which favors our model.

The litigation reality

• Patient-side AI malpractice is still forming: no landmark case yet holds a clinic liable for an AI-documentation error. The emerging theory is over-reliance and automation bias.
• Billing enforcement is already live and expensive: UCHealth paid $23M (False Claims Act, 2024) over an automated coding rule that upcoded ER visits; Aetna paid $117.7M over inaccurate Medicare Advantage diagnoses.

How it maps to our messaging

The law is mandating the model we sell: a US-based human reviews and owns the AI output. Pure-autonomous or offshore AI leaves the provider as the unprotected "liability sink." The sharpest reframe: turn "accuracy" into "accountability." The question isn't the model's accuracy percentage ... it's who's accountable when it's wrong. A trained, US-based Remote Healthcare Assistant is the documented answer.

---

Terms that need judgment

Locked terms (one right form only)

• "The toggle bar," never "the dial" or "the slider." Three positions: Pulse (left), Assist (middle), Live (right).
• "A human at the center" / "a Remote Healthcare Assistant at the center," never "human in the loop" or "HITL."
• "The Clinical Intelligence Platform (CIP)," the platform's name. Spell it out on first use, then "the CIP" or "the platform."
• "AI + Human Intelligence," the banner for the new way. "AI only" is the old way (a legitimate approach, just not enough), never named derisively.
• "Free trial," the entry point ... flipped June 11 (Jason's call). Don't revive "no-risk trial."
• Solution names are proper nouns: Pulse, Assist, Live (capitalized). "Scribe-X Assist," etc. The retired tier names (Essentials, Professional, Enterprise) never appear in any asset.

How we write (quick rules)

• No em dashes ... use "..." or single hyphens. No antithetical parallelism ("it's not X, it's Y"). No anaphora. No manufactured rule-of-three. No "AI-powered."
• Concrete over metaphor. Name the real consequence (lost revenue, fewer patients seen, a dropped follow-up), never "wobble" or "rising tide."
• Contractions always. Plain and direct. Full detail in the Language Library appendix above.