Dynavox Grew EBIT 87% — and Quietly Told You Where It Came From

A punchy margin beat, a three-quarter-long US Medicaid dodge, and a new CFO who spent his debut deferring questions

DYVOX.ST · Earnings Call · 2026-10-21

Dynavox: The Beat Is Real, The Increment Is Borrowed

Dynavox Group sold SEK 684 million of assistive communication hardware and software in Q3 2026 — up 13% reported, 15% in constant currency, 13% organic. Currency-adjusted EBIT “grew 94% year over year to SEK 119 million” — Jacob Thordenberg, CFO · 2026-10-21, a 6.8 percentage point expansion to a 17.4% margin. On a company whose stated profitability target is "reach and exceed 15%," that is a headline that writes itself. But the composition is the story. Inside that SEK 119 million sat a SEK 16 million reversal of contingent earn-out liabilities — the "majority" of SEK 19 million of other operating gains, per the CFO — plus SEK 9 million of lower restructuring cost, SEK 9 million of lower ERP cost as the rollout completed, and a net SEK 19 million R&D benefit from higher capitalization and lower amortization. That single earn-out reversal is roughly 13% of quarterly EBIT. Strip the tailwinds and the operating improvement is real but far smaller than 87%. To management's credit, the mechanics were volunteered rather than buried: “I am not completely sure which acquisition it relates to” — Jacob Thordenberg, CFO · 2026-10-21, the new CFO said, before explaining the accounting cleanly. The genuine operating story is still good — organic OpEx rose 7% against 13% organic revenue, cash flow after continuous investment more than tripled to SEK 62 million, and leverage fell to 1.3x EBITDA from 2.0x. The structural logic holds; on the Q1 call management noted fixed costs are only ~20% of COGS, so volume drops through. The question is what the run-rate margin is once the earn-out, the restructuring and the ERP savings have all annualized out.

The Reimbursement Question, Answered Identically Three Times

Here is the arc. Q1 2026: winter storms deferred shipments. Q2 2026: insurance-provider churn stretched reimbursement processes and prescription lead times. Q3 2026: Medicaid and Medicaid managed-care organisations got more combative. Different label, same underlying hole in the US growth rate, and the CEO's answer this quarter was that nothing has moved:

Today, the Medicaids and the managed care organization represents some 40% of our current revenue, so it is significant... Each state in the U.S. is required to have a budget in balance. So they are looking to do all kinds of cost-cutting... this is a whack-a-mole game.
Fredrik Ruben, CEO · 2026-10-21

Forty percent of revenue routed through state budgets that must balance, in a business already ~75% North America. That is not a weather event; it is a structural counterparty. Management insists it is plumbing, not demand — “the ratio between cases being rejected and the ratio of cases coming out of our pipeline has not changed” — Fredrik Ruben, CEO · 2026-10-21 — and the claim is consistent with July's insistence that denials were not rising. But "deferred, not lost" was also the July promise: “This should be caught up eventually” — Fredrik Ruben, Chief Executive Officer · 2026-07-22. Three quarters later, it still hasn't been. The backlog keeps rolling forward into a quarter that never arrives. The new geography is where the growth actually came from. Tender markets, a brand-new top-five keyword, is the tell: Europe grew 16% FX-adjusted but only 3% organically, with Germany and Australia carrying it and France soft on public financing. Tender-driven markets are lumpy by construction, which is exactly why a quarter can look scattered while the underlying demand is fine.

Europe, IFRS, and the Apples-to-Apples Problem

Analysts would not let go of the organic-versus-FX-adjusted gap, and the CEO's answer was unusually candid: “This is where me, as the engineer, can debate whether IFRS rules and reporting really make sense” — Fredrik Ruben, CEO · 2026-10-21. His point is legitimate — Dynavox buys its own resellers, so acquiring a distributor with 100 of turnover adds only the margin differential, not 100 of revenue, and anniversary timing distorts the comparison. But the effect is that neither number is honest on its own: organic understates, FX-adjusted overstates. Apples-to-apples entered the company's top-ten keywords for the first time this quarter — a linguistic fingerprint of disclosure being contested rather than settled. When pressed on whether the methodology changed, the CFO was flat: no. So the ambiguity is structural, and it will keep generating this argument every ninety days.

The Target That Keeps Receding

The most quietly telling thread runs across three calls. In April: “We believe that we will meet our financial targets on a full year basis, and that is 20% in local currencies” — Fredrik Ruben, CEO · 2026-04-24. In July, that became an average: “read it exactly how it's spelled out, that it's long-term and that it's on average per year” — Fredrik Ruben, Chief Executive Officer · 2026-07-22. In October, after a 15% quarter and a weaker H1, the ambition survives but the timeframe has evaporated entirely — “there will be fluctuations over weeks, months, quarters, and years” — Fredrik Ruben, CEO · 2026-10-21. That is a defensible framing for a 25%-CAGR business selling into a market where 50 million people need a solution and ~2% are served. It is also the third consecutive softening of the same sentence. The new CFO, fourteen days into a September 14 start, absorbed the cost of that ambiguity. On Europe's organic trajectory he offered to “owe you a better answer going back” — Fredrik Ruben, CEO · 2026-10-21; on Q4 currency guidance he promised to revert. First calls are always like this — but it left the CFO seat unable to reinforce the CEO on exactly the questions where the story is softest.

The One Global Thread: Memory and Component Inflation

Almost everything Dynavox discusses is idiosyncratic to assistive communication — the global keyword tape is running on AI data centres, tariffs and memory, and none of it touches this company except at the component level. Gross margin held at 69%, slightly down year on year on higher component costs, offset by cheaper freight. In July the CFO quantified that headwind at “slightly below one percentage point” — Linda Tybring, Chief Financial Officer · 2026-07-22 over a rolling twelve months. This quarter the quantification disappeared even as the pressure persisted — component cost has now ranked high for two straight quarters, and the hardware is iPad-based. That ties Dynavox to the market's broader memory cost cycle, and it is the one input management does not control: unlike freight, which they optimised away, and unlike the earn-out, which reverses in their favour, memory pricing cuts the wrong way. Restructuring cost is now literally a tailwind keyword — it ranks second this quarter precisely because it is vanishing.

What To Watch

Dynavox is not a demand problem; it is a throughput problem wearing a demand problem's clothes. The bet management is making — that 40% Medicaid exposure is an administrative scrap it wins case by case — is plausible and, so far, consistent with flat denial rates and rising US volumes at identical ASPs. But the Q4 setup is awkward: the earn-out, restructuring and ERP tailwinds all fade together, memory costs persist, and the comp against an "exceptional" 2025 gets no easier. The quarter that answers the question is the one where nothing helps and the margin has to stand on its own.