Aroa Reaffirmed Its FY27 Guidance Eight Days After Its Largest Customer Told the SEC There Is Substantial Doubt It Can Continue as a Going Concern

The claim#

Aroa Biosurgery manufactures surgical soft-tissue grafts from ovine forestomach in Auckland and sells them in the United States two ways: a direct sales force for its Myriad, Symphony and Endoform lines, and a single distributor — TELA Bio, Inc. of Malvern, Pennsylvania — for OviTex and OviTex PRS, its hernia and plastic-reconstruction products. The FY26 accounts, for the year ended 31 March 2026, put revenue from TELA at NZ$43.185m — 42% of group revenue of NZ$103.854m, down from 48% a year earlier as direct sales grew faster — and the accounts name it as revenue “derived from a single external customer”.

The market is being asked to pay for two things that depend on that customer.

The first is guidance. On 26 May 2026 Aroa guided FY27 to revenue of NZ$115–125m and normalised EBITDA of NZ$8–11m, on a stated assumption: “Sales of OviTex to TELA Bio are assumed to be flat.” The deck’s OviTex page puts it more fully: “Sales to TelaBio is conservatively expected to be flat due to hospital contracting headwinds.” The annual report’s CEO letter repeats it: “This guidance reflects continued growth in our direct business, an assumption of flat sales to TELA Bio, and planned investment of approximately $9 million …”. The same deck describes the TELA relationship as “recurring, high-margin revenue, not a dependency”, contributing “~75% Gross Margin, no sales expenses”.

The second is a balance-sheet item most summaries of Aroa do not mention. Aroa is paid by TELA on a formula — TELA’s own 10-K states it pays Aroa an amount that “results in an amount equal to 27% of the Company’s net sales” of OviTex — and Aroa’s accounting policy recognises that consideration early. In the accounts’ own words: “The Group receives consideration from TELA Bio comprising of a transfer price for the products shipped to TELA Bio, with the balance of the consideration received on a quarterly true-up to the agreed percentage based on TELA Bio’s net sales. Using the expected value method, the Group estimates the true-up on TELA Bio’s inventory at the reporting date considering the expected sale of those products by TELA Bio.” That estimate — the accrued top-up still to come on stock TELA holds but has not yet sold — sits on Aroa’s balance sheet as a contract asset. At 31 March 2026 it stood at NZ$21.442m:

• 21.6% of Aroa’s net assets of NZ$99.3m;
• 79.3% of its NZ$27.1m of cash and term deposits;
• 20.6% of FY26 group revenue;
• 2.26 times the NZ$9.5m midpoint of FY27 EBITDA guidance.

The accounts record that the Group assessed it for expected credit losses and “concluded that no material impairment provision was required due to the strong payment history and the financial position of the counterparty.” The accounts carrying that sentence were signed on 26 May 2026.

The constraint#

The counterparty files its own accounts, quarterly, with the US Securities and Exchange Commission, under its own Exchange Act obligations. Aroa does not control that record, and it is the record this paper tests the claim against.

On 11 August 2026 — 77 days after the accounts carrying that sentence were signed — TELA Bio filed its Form 10-Q for the quarter ended 30 June 2026. In its own words:

“These conditions and events raise substantial doubt about the Company’s ability to continue as a going concern within one year after the date these consolidated financial statements are issued.”

The conditions it describes are specific. TELA’s forecasts “indicate it is probable that the Company will not achieve the minimum revenue threshold required under the Credit Agreement” — the US$60m term loan from Perceptive Advisors it refinanced into in November 2025 — “during the twelve-month period following the issuance of these consolidated financial statements.” Management “is evaluating and pursuing various actions, including seeking a waiver of, or amendment to, the applicable covenant requirements” — seeking, not holding. And if the breach comes and the loan is called: “The Company does not expect to have sufficient liquidity to repay such obligations if repayment were accelerated.”

The balance sheet behind those sentences, from the same filing: cash of US$30.4m at 30 June, down from US$50.8m at 31 December 2025 — a burn of US$3.4m a month across the half. The credit agreement requires liquidity of US$5.0m at all times, so at that burn rate the covenant floor is about seven and a half months away. That figure is the desk’s linear arithmetic on the half just reported, not TELA’s own view: the same filing states that “[b]ased on our current business plan”, TELA believes its cash is “sufficient to meet our capital requirements and fund our operations for at least the next 12 months”, and that cash used in operations in each remaining 2026 quarter “is expected to be lower” than the first half’s. The substantial doubt in the filing is driven by the covenant test, not by a forecast that the cash runs out. Total liabilities of US$75.6m exceed total assets of US$59.7m by US$15.9m; the accumulated deficit is US$421.1m. Most of those liabilities is the US$60m Perceptive term loan, which matures in November 2030, and TELA reports working capital of US$36.3m — the deficit is an accounting position, not a statement that TELA cannot pay its debts as they fall due. The revenue that must clear the covenant threshold went backwards in the latest quarter: US$19.3m against US$20.2m a year earlier, −4.5% — and the OviTex and OviTex PRS lines, the base Aroa’s 27% actually applies to, fell faster, −5.3%.

The covenant itself is public: the credit agreement is a filed exhibit, and its section 10.02 sets a rising schedule of minimum trailing-twelve-month revenue. On the desk’s arithmetic from TELA’s filed quarters, TELA passed the 30 June 2026 test — US$79.9m required — with trailing revenue of US$79.908m: a margin of roughly US$8,000, or one hundredth of one per cent. (The arithmetic is the desk’s; the covenant’s definition of revenue excludes one-off licence and milestone receipts and so may not equal the GAAP figure exactly, and the fourth-quarter 2025 figure is derived, both as set out in the methodology.) The schedule then climbs: US$80.8m at 30 September, US$86.3m at 30 December — a threshold 8% above where trailing revenue now stands, for a company whose revenue is falling and which, on 10 August, withdrew its own guidance: “we are withdrawing our full year 2026 revenue guidance”, its new chief executive said, in her seventh day in the role. The guidance withdrawn had been “at least 8% growth over full year 2025” — almost exactly the growth the December covenant requires.

Two smaller clocks are also running in the public record. TELA received a Nasdaq minimum-bid-price deficiency notice on 17 March 2026 — thirty consecutive business days of closing bid prices below US$1.00 — with the initial cure period expiring 14 September 2026; the same notice records that a second 180-day period may be available if TELA transfers to the Nasdaq Capital Market and meets its standards. And its US$50m at-the-market equity programme, in place since November 2023, sat entirely unused at the latest filing (“[n]o sales have been made under the Equity Agreement since its inception”), with the share price around sixty-five cents.

Around the filing sit two more dated facts from TELA’s own record. On 3 August 2026, eight days before the 10-Q, its board determined that chief executive Antony Koblish would no longer serve, and he resigned from the board the same day; Heather Getz was appointed chief executive. The 8-K records the departure as “a termination without cause” and states that it “did not result from any disagreement with the Company on any matters relating to the Company’s operations, policies or practices”. The amended 8-K provides that his performance stock units “shall remain outstanding and eligible to vest upon the achievement of certain annual revenue and gross margin goals to the extent achieved on or prior to December 31, 2026” — the awards’ existing performance-period end — and “shall vest at target” if a change in control occurs by that same date, a standard equity-plan protection. Under the Perceptive agreement, “events constituting a change of control” are an event of default, so any sale runs through the lender. As at 19 August 2026 — the latest filing on TELA’s EDGAR record when this paper was prepared — no waiver, amendment, forbearance or new financing had been announced.

The gap#

The gap is between what Aroa’s forward numbers assume and what its counterparty’s own filings describe — and it shows up in three places.

First, the guidance. FY27 assumes OviTex sales to TELA of NZ$42.8m — flat on FY26. TELA’s sales are not flat: −4.5% in the June quarter, −0.9% across the half, and the covenant arithmetic above gives the company itself a probable-breach horizon inside the guidance year. The desk’s sensitivity is simple and stated: hold Aroa’s guided cost base (the NZ$9m of new investment is, in the company’s word, “planned”, and the deck allocates it to named programmes), apply the ~75% OviTex gross margin Aroa’s own presentation states, and flex only the OviTex line. Every dollar of OviTex shortfall then takes roughly 75 cents off normalised EBITDA:

A 24.9% OviTex decline erases the bottom of the guidance range; 29.6% erases the midpoint. Those are not small declines — but the scenario the guidance must survive is not a soft quarter. It is a distributor that is publicly seeking covenant relief, burning US$3.4m a month against a US$5.0m liquidity floor, and whose most economical way to conserve cash is to slow its buying of inventory from its supplier. The filed agreement’s minimum-purchase protection is time-limited and appears exhausted on its face: section 8.6 sets “Minimum Amounts” for North America of US$2.0m across the first and second Contract Years in aggregate, US$2.0m for the third, US$3.0m for the fourth and US$4.0m for the fifth, enforced by a section 8.7 make-whole remedy — and the schedule ends there, with a sixth-year US$5.0m minimum reviving only “[u]pon any Change of Control of TELA Bio”. Beyond the binding near-term portion of TELA’s rolling forecast and the purchase orders that must match it, no continuing minimum-purchase obligation appears in the filed agreement or in either company’s description of it. Aroa’s revenue from TELA is booked on shipment plus accrued true-up: sell-in stops before sell-through does. And the exposure is to TELA’s dollars, not its units: TELA’s June quarter sold 12% more core OviTex units for 4% less revenue — on what its release calls “price and mix headwinds in hernia procedures” and a decline in OviTex PRS unit volumes — and Aroa’s 27% is a share of net sales, so price and mix pass straight through. One more definitional point: Aroa states its FY27 guidance on a constant-currency basis at NZ$/US$ 0.60, against an FY26 average of 0.59, so “flat” means flat in US dollars before translation — roughly NZ$42.1m of reported OviTex revenue on Aroa’s own convention, not FY26’s NZ$42.8m.

Second, the carrying value. The NZ$21.4m contract asset is, by Aroa’s own policy, an estimate of the true-up on TELA’s unsold stock. Set the accrual against the stock it is measured on:

Over FY26 alone — 31 March to 31 March — the accrual rose 14.6% while TELA’s inventory fell 9.8%. (The fuller quarterly series is choppier: TELA’s inventory has been broadly flat at US$11–12m since mid-2025, and rose 10.9% between December 2025 and March 2026; the March balance dates are the ones on which Aroa measures the accrual.) Converted at the NZ$/US$ rate implied by Aroa’s own disclosure of its TELA shareholding in both currencies, the accrual moved from 0.79× TELA’s total inventory at March 2025 to 1.01× at March 2026. The level of that ratio is not itself the finding: TELA carries the inventory at the transfer price it has already paid, Aroa accrues the true-up still to come, and because the transfer price is set at 200% of cost of goods while Aroa’s stated OviTex gross margin is ~75%, the two amounts are mechanically of similar size — a ratio near 1.0× is roughly what the arrangement produces. The observation is the direction of travel: the accrual per dollar of the stock it is measured on has risen by a quarter in a year. What is not inference: Aroa cut the sales-growth assumptions inside the accrual this year — OviTex from 19% to 2.7%, OviTex PRS from 37% to 15.8% — and states the change reduced accrued revenue by just NZ$471k, while the asset still grew NZ$2.7m. Even the cut rates remain positive. They measure a different quantity from the guidance assumption — the accrual rates project TELA’s own sell-through of stock it already holds, while the flat assumption governs Aroa’s sell-in to TELA, and a distributor can sell through while it stops buying — but both projections were set before the counterparty publicly doubted its own continuity, and neither has been revisited since.

The asset’s history points one way. It has risen at every balance date on record — NZ$4.8m (March 2022), 11.1m, 15.1m, 18.7m, 21.4m. The true-up itself settles quarterly under the agreement, so the balance turns over; what the record shows is that new accrual has outrun settlement in every year — the cash-flow statement’s combined movement in prepayments and contract assets has been an outflow in each of the last four. Its growth assumptions were, in Aroa’s own words, set “[h]aving considered TELA Bio’s revenue guidance” — and TELA withdrew its guidance on 10 August. And the underlying agreement, a filed exhibit to TELA’s SEC registration statement, gives the estimate hard edges: on the general product lines the annual true-up runs both ways — “Aroa shall pay such difference to TELA Bio” if the year’s 27% entitlement falls short of what TELA has already paid — and TELA holds a one-way right to “offset payments owed to it by Aroa … against any Transfer Price amounts that are or may become due to Aroa”. TELA’s own balance sheet at 31 March 2026 records US$2.8m due to its contract manufacturer, inside accrued expenses, against the US$12.3m that Aroa’s NZ$21.4m accrual converts to at the same date’s rate — but neither company maps that balance onto Aroa’s accrual, and on the disclosed mechanic it should not: the accrual is a claim that crystallises only as TELA sells the stock, against a counterparty whose lender holds a first-lien security interest over substantially all of its assets, for a loan whose US$60.0m principal already exceeds them.

Third, the dates. The sequence matters more than any single number:

• 25 November 2025 — Aroa’s interim presentation devotes a slide to TELA’s October capital raise and refinancing, headed “More confidence in TELA Bio’s financial position” and “Future revenues are much more secure strengthening AROA’s position”, tabulating TELA’s pro-forma cash of US$60m and a “~24 months” cash runway. This is the only Aroa document the desk found that engages with TELA’s financial condition — and it engages with it as reassurance.
• 26 May 2026 — Aroa signs accounts carrying the NZ$21.4m asset and the Group’s recorded conclusion that no impairment provision is required, citing “the financial position of the counterparty”; guides FY27 on flat TELA sales. The subsequent-events note: “There have been no significant events subsequent to reporting date.”
• 3 August 2026 — TELA’s board determines its CEO will no longer serve; the 8-K records a termination without cause and no disagreement, and a new CEO is appointed the same day.
• 10 August — TELA reports its second quarter after the US close and withdraws its full-year revenue guidance.
• 11 August — TELA files the going-concern 10-Q.
• 19 August — Aroa holds its AGM. The lodged Chairman’s address lists “[c]ontinuing cash flows from the Ovitex products via TELA Bio …” among the drivers of the FY26 result it reports; separately calls TELA “an important partner”; and — of the model in which TELA cash flows funded the build-out of Aroa’s own sales force — says “This model continues.” The lodged CEO presentation restates FY27 guidance in full, with “Sales of OviTex to TELA Bio are assumed to be flat” and “conservatively expected to be flat due to hospital contracting headwinds”, on slides reproduced from the 26 May results deck. Neither lodged document, nor the results of meeting, contains the words “going concern”, “substantial doubt”, “covenant”, “waiver” or “impairment”. (This paper is scoped to the lodged record: what was said orally in the room, including in Q&A, is not public, and a field assignment below is directed at exactly that.)

No ARX announcement between 11 and 20 August addresses TELA’s filing: the lodgements in that window are two securities notices on 12 August and the three AGM documents of 19 August. The one public market that has repriced is TELA’s own:

Corroboration#

These are secondary signals. None of them is the thesis; each is consistent with it.

The concentration is audited as a risk. BDO’s audit report on the FY26 accounts treats the TELA Bio revenue share — the accrual above — as a Key Audit Matter, recording that the consideration “is variable since the quantum of TELA Bio’s inventory that is sold and the price that it is sold at, are uncertain”, and that the auditor obtained confirmation from TELA of its stock holding and sales history. The facts in this paper are in the audited record; the question the paper raises is what has happened to them since the record was signed.

TELA’s own market has voted on the disclosures. A 30.1% fall in seven sessions, from the last close before the 10 August after-market release — on a stock that had risen 23% on the day the CEO transition was announced the week before. The window prices three things the desk cannot separate: the second-quarter miss, the guidance withdrawal the same evening, and the going-concern note filed the next day; the single largest session, −15.1% on 11 August, followed the release and preceded the filing.

No rescue is on the record — and neither is a crisis footing. No strategic review or sale process appears in the TELA filings reviewed; the 10-Q names three actions, on a list it introduces with “including” — seeking a waiver or amendment, revenue initiatives, and expense reduction — and no waiver, amendment or financing had been announced as at 19 August. What the record shows in the other direction is small but dated: TELA’s incoming chief executive bought stock on market in the week after the filing — 15,000 shares at US$0.6978 on 14 August and 12,500 at US$0.6649 on 17 August, roughly US$19,000 in total, disclosed on Forms 4 filed 17 and 19 August. The desk draws no conclusion about what is under way in private; the point is that nothing resolving the doubt is yet on the record in either direction.

The register shows no positioning for it. ASIC’s published short-position data shows ARX short interest never exceeded 0.17% of issued capital in 2026 and had fallen to effectively zero — 440 shares — by 10 August; the latest published file (14 August; the series lags about four trading days) shows no reported ARX short position at all. The only two substantial-holder notices for ARX in 2026 were both increases, the more recent disclosing purchases of roughly A$1.5m of stock on 5 August, six days before TELA’s filing. Those are dated facts from the register, not a judgment on anyone’s portfolio; they say only that whatever this paper describes, the market has not traded on it.

Aroa’s framing is unchanged. The FY26 deck characterises TELA revenue as “recurring, high-margin revenue, not a dependency” — on a slide that itself prints “Partner sales: 41%”, so the concentration is disclosed on its face; the desk’s point is about the characterisation, not the disclosure. The same characterisation, and the chairman’s “This model continues”, were repeated at the AGM eight days after the model’s other half disclosed substantial doubt about its ability to continue.

What would explain this instead#

The flat assumption already is the conservative case. Aroa’s own words describe the flat assumption as conservative, adopted for “hospital contracting headwinds” — the company may regard TELA-side weakness as already inside guidance, with direct-sales growth (Myriad grew 54% in FY26) doing the work. That is a genuine defence of the P&L: Aroa is debt-free, held NZ$27.1m of cash and term deposits at March, generated a NZ$10.5m operating cash inflow in FY26, and 58% of revenue sits outside the TELA line — the chairman put direct sales at “close to 60% of our revenues”. What it does not answer is the balance sheet: “conservative” was written 77 days before the counterparty’s going-concern filing, and no Aroa document since has said whether the assumption — or the impairment conclusion — survives it. The H1 FY27 accounts in November will answer this directly; if they show the flat assumption holding and the asset converting to cash, this paper’s forward pillar is wrong.

TELA is rescued, and the asset converts in full. TELA is seeking a waiver; Perceptive may grant one — lenders rarely accelerate a borrower whose collateral is worth less than the loan. TELA also holds an entirely unused US$50m at-the-market equity programme, and its new chief executive bought stock on market in the week after the filing. A recapitalisation, or an acquisition by a solvent buyer (the agreement’s change-of-control provisions cut both ways), would leave the stock selling through and the accrual converting to cash on schedule. TELA has, on Aroa’s own account, always paid — “strong payment history” is accurate. On this reading the going-concern note is the audit-standard consequence of a probable covenant test failure, not a prediction of collapse, and the correct response to it is patience. The evidence that would settle it: a waiver or financing 8-K, or the next quarter’s covenant disclosure.

The credit conclusion was right when signed, and remains defensible. On 26 May 2026 TELA had just refinanced into a five-year facility and reported US$39.5m of cash at its most recent balance date; nothing in the public record then carried going-concern language. An expected-credit-loss assessment is made at the reporting date on the facts at the reporting date. The staleness this paper describes is real but has a routine remedy: the next reporting date is 30 September 2026, and NZ IFRS requires the assessment to be remade then on the new facts. On this reading the absence of an intra-period announcement reflects a judgment that the exposure is not yet material enough to require one — a judgment that is the company’s and its advisers’ to make, not the desk’s — and the November accounts will either provide against the asset or explain why not.

The exposure is bounded and the arithmetic overstates it. TELA’s inventory was US$12.2m at transfer-price cost at 31 March and US$11.1m at 30 June; even a full destock is a bounded event, and some of the accrual has already been invoiced in the ordinary quarterly true-up cycle since March. If TELA fails entirely, Aroa’s agreement gives it a fallback the desk cannot price: the products are Aroa’s, the term runs to 2031, and a successor arrangement — direct sale or a new distributor — would recover some of the economics. A change of control at TELA would even cut in Aroa’s favour on one axis: the filed agreement revives a sixth-year North American minimum-purchase amount of US$5.0m “[u]pon any Change of Control of TELA Bio”. And NZ IFRS 15 itself constrains the accrual — variable consideration is recognised only to the extent a significant reversal is not highly probable, a constraint Aroa’s auditor tests annually. A full write-off of the contract asset is not the base case here and this paper does not model one; the finding is about the asset’s carrying basis and the guidance assumption, not a prediction of total loss.

Kill criteria#

This thesis is wrong, and this paper will say so at the top of the page, if any of the following appears on the record:

1. A waiver, amendment, forbearance or new financing that resolves the going-concern doubt, announced by TELA — or an acquisition of TELA by a buyer that affirms the Aroa agreements.
2. TELA’s net sales returning to growth for two consecutive quarters, or its Q3 2026 filing dropping the going-concern language.
3. Aroa’s H1 FY27 accounts showing the contract asset stable or reduced through collection (not impairment), an ECL note that engages with TELA’s condition on its face, and OviTex revenue at or above the flat line — the combination would show the accrual self-liquidating on schedule and the flat assumption holding.
4. Evidence that the true-up cycle is materially shorter than modelled — e.g. disclosure that most of the 31 March balance had been invoiced and collected by September — which would make the accrual’s exposure window a quarter, not a stock of risk.
5. A record of the AGM showing management addressed TELA’s condition orally and qualified the flat assumption — which would kill the disclosure-silence observation specifically (the guidance and carrying-value arithmetic would stand or fall on 1–4).
6. A disclosed reconciliation of the contract asset’s conversion cycle — for example, disclosure of how much of each year’s opening balance was invoiced and collected within the following year — showing the accrual turning over faster than the balance-date series implies. That would convert the carrying-value pillar from a stock of risk into a quarter’s timing difference, and this paper would say so.

Right of reply#

The company was not contacted before publication; non-contact before publication is this desk’s stated policy, recorded on the methodology page. Every claim above is drawn from the company’s own ASX-lodged documents, annual reports and investor presentations, from TELA Bio’s own SEC filings, and from dated price records, each linked in the source list with access dates. Where the company has addressed a point, its position is quoted in its own words and engaged with rather than dismissed — including its description of the flat-TELA assumption as conservative and its reasons for it, its “not a dependency” characterisation and the direct-sales growth that supports it, its expected-credit-loss conclusion and the payment history behind it, and its AGM account of the TELA relationship. Nothing in this paper alleges any breach of law, of the ASX Listing Rules, of the NZX or NZ Companies Act requirements, or of any regulator’s rules, by Aroa Biosurgery, by any of its directors or officers, or by TELA Bio or any of its officers; nothing in it alleges that any statement by either company was made otherwise than in good faith; nothing in it alleges any failing by BDO, by Perceptive Credit Holdings V, LP, or by any holder of Aroa securities named in the register findings; and nothing in it speaks to the clinical merit of either company’s products, which the desk has no basis to assess. Whether TELA’s August filing required any Aroa disclosure, and when any re-assessment of the contract asset should occur, are judgments for the company and its advisers under the continuous-disclosure and accounting frameworks; this paper records only what the public documents say and the dates they say it. Should the company respond, its response will be published here in full.

Methodology#