idea-010 · viability deck
Every slide states a condition that would have to be true, then reports where it stands using the candidate file's own claim and its own confidence tag. A deck never upgrades a tag, invents a number, or recommends anything, and there is no ask slide.
The chair: the investor, and the practice administrator who has to decide whether insourcing a lab test makes or loses money per patient.
How to read this: every slide is a condition, not a conclusion. A bracketed reference points into knowledge-base/candidates/idea-010.md — the number is the claim's position in its ## Claims list and the tag is copied from it, never adjusted here. [no claim] means nothing in the file speaks to the condition at all. No number appears here that is not in a claim, and there is no ask slide. See docs/deck-spec.md.
How to read the composite: 60/100 is computed over Factor 1 alone — 15% of rubric weight — and is the renormalisation the rubric prescribes for "factors with at least one verified claim". It is not a judgement that the candidate is 60% good. Every factor an investor reads first — reimbursement, TAM, FTO, competitive intensity — is blank, meaning no verified claim, not a zero. 1 of 9 factors is verified.
Would have to be true: Medicare Part B would have to pay the clinic for a potassium and a creatinine performed in the office, out of the Clinical Laboratory Fee Schedule, with the strip bought from practice operating expense.
Where it stands: The file's own load-bearing economic claim states the structure and points negative: these are laboratory codes paid from the CLFS rather than the Physician Fee Schedule, the allowable for a single electrolyte is low in absolute terms, and the strip's cost to the practice must sit far below it or the clinic loses money on every test it insources — with no dollar figure asserted on either side [claim 14: unconfirmed]. Billing at all requires the site to hold a CLIA Certificate of Waiver and to append the QW modifier, both unchecked [claim 29: unconfirmed], [claim 13: unconfirmed]. Two conditions that decide whether any revenue reaches the clinic are on record only through this deck: that insourcing moves the payment from the reference laboratory to the clinic [claim 37: unverified], and that two analytes off one strip are two payable tests rather than one [claim 38: unverified].
What would settle it: The CMS Clinical Laboratory Fee Schedule annual file — public, licence-free, and a human download into data/ — for the national limitation amount on the two codes. The CMS CLIA waived-test transmittal for the QW leg. Medicare billing rules and NCCI edits for who bills an insourced test and whether the two analytes bundle. Note payment reads a HCPCS/PFS cache that may not carry CLFS at all, so an unconfirmed there is a missing file, not a finding of low payment.
If it's false: Factor 2 has nothing under it in the direction that matters. If the clinic does not capture the payment, or the two analytes bundle to one, the per-visit revenue the entire consumable annuity rests on is smaller than the business case assumes — and Factor 3's price component [claim 17: unconfirmed] is bounded above by a smaller number than anyone has written down.
Would have to be true: A code has to exist and fit, a payment amount has to clear the cost of delivery, and no coverage policy may exclude the setting. All three, independently.
Where it stands:
connectors cpt returns unverifiable for by design**
[claim 11: unconfirmed], [claim 12: unconfirmed]. The HCPCS route cannot substitute: these are
Level I codes and the Level II cache is absent. The QW modifier rule is in the same position
[claim 13: unconfirmed].
payment lookup hit
a missing local cache — nothing was searched, which is not a finding of low payment
[claim 14: unconfirmed].
What would settle it: For the descriptors and the QW rule, a human attaching a licensed source — AMA CPT Professional 2026, or a licensed vendor feed — as a manual citation. Do not cite connectors cpt as a check; its unverifiable is by design and means nothing either way. For payment, the CLFS file above. For coverage, a CMS Medicare Coverage Database export plus a named-LCD search for point-of-care potassium or POC chemistry, read for stance — coverage verifies a policy's existence, not positive coverage.
If it's false: Code existence would not be a pathway even if all of it confirmed. The deciding question is margin, which is Slide 7 and the load-bearing condition — and it stays unanswerable while both sides of the inequality are blank.
Would have to be true: A countable number of clinics, running a countable number of titration tests per patient per year, at a price above cost, with a reachable share of them switching.
Where it stands:
What would settle it: procedures 84132 / 82565 once a human supplies the CMS Part B National Summary Data File or the Physician & Other Practitioners PUF into data/, filtered to office place-of-service and read as a floor. The clinic denominator from PECOS / the Provider Enrollment PUF by specialty, or a licensed POC-diagnostics report. The switching share needs a published preference study or a commissioned practice survey; market and adoption are unverifiable by design and must not be cited as having checked it.
If it's false: TAM below floor is a hard kill condition at Stage 5. It did not fire here — the gate recorded it as cannot be evaluated, because an undefined operand is not a low one. An investor should read that as the number being absent, not as the number being acceptable.
Would have to be true: The company would have to fund two predicate arguments in one submission, a CLIA waiver study programme at multiple untrained-operator sites, a novel three-channel strip, and a strip factory — before billing anyone.
Where it stands: The file records this as the candidate's honest principal weakness and asserts no dollar and no month figure, correctly, since either would be invented [claim 26: unconfirmed]. The costs stack rather than overlap, and two of the four legs are verified facts about what must be built: potassium and creatinine are separate generic types, Class II, under 862.1600 and 862.1225 respectively [claim 1: verified], [claim 2: verified], with distinct product codes [claim 3: verified], [claim 4: verified]. The waiver leg's dominant uncertainty — whether FDA's Dual 510(k) and CLIA Waiver by Application pathway is available as described — is structurally unconfirmable in this repo, because no connector reads FDA guidance [claim 8: unconfirmed]. And the calendar's largest single term, whether the potassium channel goes 510(k) or De Novo, is entirely unsearched: denovo has no local cache, so no De Novo absence may be inferred [claim 7: unconfirmed].
What would settle it: A human placing FDA's De Novo listing into data/ to close the 510(k)-versus-De Novo fork, and attaching the named FDA Dual 510(k)/CLIA Waiver guidance PDF. Then scope the evidence package off the special controls in the two regulations and the CLIA waiver criteria in 42 CFR 493.15 — noting the file's audit found that lookup did not reproduce today owing to an upstream eCFR outage, which is an availability failure and not a finding. The dollar figures are Tier 3 human estimates; no connector returns submission cost or review time.
If it's false: If the potassium channel needs a De Novo, Factor 7's calendar and Factor 1's route change together — the file says so directly — and this becomes a longer programme than anything else in the portfolio, against the same unquantified margin.
Would have to be true: The hemolysis detect-and-suppress logic, the potassium-selective membrane chemistry at whole-blood ionic strength, and the strip yield curve would have to be ownable — and someone would have to have checked that they are not already owned.
Where it stands: The moat is asserted as those three assets with the meter conceded as a commodity, on the Generator's judgement and not connector-checkable as written [claim 24: unconfirmed]. The IP field underneath it is unexamined, not clear: no patent search was performed, no patent number is named, and PATENTSVIEW_API_KEY is unset in this deployment [claim 25: unconfirmed]. An absence of found patents is not freedom to operate, and the named likely estates — Nova Biomedical, Abbott Point of Care, Roche, Siemens Healthineers, i-SENS — are precisely the parties who already own high-volume strip manufacturing and ISE chemistry [claim 27: unconfirmed], which means the yield-curve leg describes the incumbents' moat at least as well as this candidate's.
What would settle it: A real prior-art search by a human before any API key is set — Patents Public Search / Espacenet across CPC G01N27/333, G01N33/49, G01N33/72, G01N27/327 and C12Q1/00, with assignee sweeps on the five named firms. The file names the single highest-value query: art on hemoglobin or hemolysis detection used to flag or suppress a co-located electrolyte result. Each number found then through connectors patent. Definitive answer is a Tier 3 legal FTO opinion at Stage 7.
If it's false: FTO blocked is the rubric's only automatic kill regardless of other scores, and it would land on the one asset that differentiates this device. Stage 5 could not fire it here — an unexamined field is neither clear nor blocked — so this is unmeasured exposure, not cleared exposure.
Would have to be true: No incumbent one product-line extension away would have already done this — and the reason none has would have to be something other than "it does not work."
Where it stands: This is the slide the file's own frontmatter gets wrong, and the file now says so against its own interest. The frontmatter still reads "no specific predicate identified for the potassium channel"; the Stage 3 audit enumerated the potassium generic type completely across both its product codes — CEM and MZV, the second missed by the first pass — read 34 clearances over 30 years, and found that predicate candidates demonstrably exist: K200865 (Piccolo Potassium Test System, heparinized whole blood at a point-of-care location) and K042270 (StatChem Stat K, explicitly "may be used ... at point of care sites"). What does not exist among the 34 is one in this form factor — a single-use strip read on a handheld meter from a fingerstick drop [claim 7: unconfirmed]. That enumeration is a Stage 0 ScanResult carrying no confidence tag and unable to acquire one, so it bounds how the gap may be described and is not itself evidence in either direction. A CLIA-waived potassium test on the Piccolo platform is claimed but the waived half was never checked; a 510(k) IFU saying "point-of-care location" is not a CLIA waiver [claim 10: unconfirmed]. The incumbent roster is a named-firms judgement with no clearance scan behind it [claim 27: unconfirmed], and the i-STAT CHEM8+ leg did not establish its potassium half at all [claim 9: unconfirmed]. The file's own reading of the empty form factor is that the likelier explanation is an unsolved physical problem rather than an unnoticed opportunity [claim 22: unconfirmed].
What would settle it: A human's substantial-equivalence judgement on K200865 and K042270 — that is an FDA judgement, not a database fact. FDA's De Novo listing into data/, since De Novo remains unsearched. The CMS CLIA waived-test transmittal, which is the cheapest document here and either confirms or refutes the Piccolo waived premise on its own. And a human full-text read of the published point-of-care potassium literature: the Stage 0 scan logged to knowledge-base/signals/log.md on 2026-08-25 names records in this exact form factor and carries no confidence tag, so it is a pointer for a human to pull, never a finding, and publication is neither clearance nor a US regulatory record.
If it's false: If the form-factor gap is empty because the physics does not work, Factor 8 is not white space and Factor 6 is the whole story. If it is empty because nobody bothered, the incumbents named can close it the moment it is shown closable — and the moat on Slide 5 is the only thing standing between this and a fast follower with an existing strip line.
Would have to be true: One public document plus one internal cost estimate would have to be enough to determine the sign of the margin — whether the payment for an insourced potassium and creatinine exceeds the fully-loaded cost of doing it in the office.
Where it stands: Both sides are blank. The payment side is a licence-free public CMS file nobody has downloaded [claim 14: unconfirmed]; the cost side is a Tier 3 human number that no connector returns [claim 17: unconfirmed]. The full-cost form of the inequality — strip plus meter amortisation plus QC material plus operator time, at one clinic's test volume — is on record only through this deck [claim 36: unverified], as are the two conditions that determine what lands on the payment side at all [claim 37: unverified], [claim 38: unverified].
What would settle it: The CMS Clinical Laboratory Fee Schedule annual file for the national limitation amount, downloaded into data/ by a human — free, public, and hours of work. Against it, a bill-of-materials estimate for a three-channel electrochemical strip at volume, which is Tier 3 engineering judgement. The file's own gap analysis calls this the number the whole business case hangs on.
If it's false: If the allowable sits below the fully-loaded cost, the clinic loses money on every test it insources and no amount of clinical value fixes it — the product would need a different payer, a different setting, or the home-use route carried in device_class:, which is a second submission with a lay-user evidence package on top [claim 8: unconfirmed].
Every condition above with nothing verified behind it — which is all seven. Slide 4 is the only partial exception and it does not change the count: four verified claims establish that potassium and creatinine are separate Class II generic types, which describes part of what must be built, while the condition that slide actually states — that the sequence is fundable on a knowable calendar — has nothing verified behind it at all. No verified claim in this file touches payment, market, IP or competition. Factors 2, 3, 4, 5, 7, 8 and 9 are all blank in the Verified Composite.
data/, then a commissioned survey for the switch.
One asymmetry worth carrying to Stage 6: the two audit findings that moved this file both moved it against its own pitch — the whitespace claim is weaker than the frontmatter says [claim 7: unconfirmed], and the one verified fact that bears on the predicate argument names a sample matrix this device does not use [claim 1: verified]. A file that corrects itself downward is more trustworthy than one that does not, and it is also a file whose headline is worse than it looks.
If only one thing from this chair could be checked: whether the CLFS allowable for potassium and creatinine exceeds the fully-loaded per-test cost of running the strip in an outpatient clinic [claim 14: unconfirmed], [claim 36: unverified].
It outranks TAM here, and deliberately. Market size determines how big the business could be; this determines whether there is a business at all, because a negative per-test margin makes every volume assumption on Slide 3 an amplifier of losses rather than of revenue. It is also the cheapest unresolved question in the whole file — one free public CMS download and one internal bill-of-materials estimate — and, uniquely among the conditions here, it can be answered before any strip exists, any clinician is interviewed, or any patent is searched.
Naming it is not a recommendation, a gate, or a kill.