Komplett's Gross Margin Is the Memory Cycle's Shock Absorber

Q3 earnings show a Nordic e-tailer squeezed by a semiconductor squeeze it can't price away — while leverage improves and management quietly widens its safety net

KOMPL.OL · Earnings Call · 2026-10-22

A difficult quarter, described in the mildest possible language

Komplett ASA — the Norwegian consumer-electronics e-tailer behind komplett.no, NetOnNet and Webhallen — reported a third quarter where the P&L did the talking and the tone did the apologising. Gross profit fell 9% in nominal terms, adjusted EBIT dropped NOK 12 million year over year, and yet the balance sheet improved: leverage came down to 2.4x from 3.0x a year earlier, liquidity stayed above NOK 1 billion, working capital tightened on disciplined inventory control. The headline tension is that this isn't a demand story, or a cost story, or an execution story. It's a supply-chain price shock arriving through the back door of the product mix, and it is arriving at a company whose category exposure is unusually concentrated in exactly the wrong place.

The memory tax, and why it lands harder on Komplett than on most

Management was unusually explicit about the mechanism. The memory price shock has been broadening — CFO Thomas Røkke described it as now visible "across all kind of core tech categories including components in gaming computers, and also telephone and mobiles" — and the damage is not linear. Rising memory prices crush volumes in adjacent components through cross-elasticity: motherboards, CPUs and similar parts don't see their own prices rise, so there's no price to pass through. Volume simply disappears.

You see memory prices going up, but that drops the volume of related PC components like motherboards, CPUs, and others, and drag them down in significant volumes, which cannot be compensated by price on those items.
Thomas Røkke, Chief Financial Officer · 2026-10-22

This is why the gaming segment and the B2C segment absorbed nearly all the damage while the B2B segment grew and distribution held flat with an uptick of NOK 2 million. The CFO was candid that margin delta across divisions is "basically proportional to the exposure to the gaming segment." Røkke put the exposed share of revenue at "north of 2/3 in terms of categories when it comes to memory-exposed products" — a higher proportion, he conceded, than most of the industry. Layer on seasonality and the picture darkens further: July saw declines of "up to 60%-70%" in some important summer categories, and Q3 is structurally a give-back quarter after an exceptionally strong Q2 product season. Sweden was more stable in real terms than Norway, where the NetOnNet turnaround deliberately removed unprofitable revenue — a choice that hurts the top line and helps the operating line. What did work was the cost base: adjusted OpEx fell NOK 36 million year over year, roughly half of it structural per management's own framing, aided by currency and restructuring. That's real — but it's roughly three quarters' worth of mitigation against a gross-profit hole that scales with a commodity price nobody at Komplett controls.

The part of the call that matters more than the numbers

Three evasions, taken together, tell you more than the reported figures. First, the company declined to guide Q4 at all — standard policy, but conspicuous in a quarter where the shock is broadening into phones and computing rather than narrowing. Asked how much of higher memory cost has been passed through, Røkke offered that "there is still upside in the prices" and then: "I would kindly ask to not have to answer that." Second, the decomposition of the 60-basis-point gross margin decline into memory versus mix was refused outright: “I don't think we would share that information even if we had it.” — Thomas Røkke, Chief Financial Officer · 2026-10-22 The explanation that followed was actually more useful than the non-answer — much of the mix effect lives inside categories, as consumers shift from low-end models (higher margin) to high-end models (lower margin). Third, and most tellingly, when pressed on incremental cost levers should the memory crisis persist, the answer was that it's "a bit too early to start discussing" them, with focus instead on delivering "a good peak." Then there is the financing footnote that deserves more weight than it got. Alongside extending revolving facilities two years, with visibility to 2030, Komplett raised leverage headroom into early next year — precautionary reasons, stressed twice, "not anticipatory." When a CFO pre-emptively widens covenant space and then insists it's unrelated to conditions, that is a hedging statement, not a reassurance. And the competitive backdrop is disorderly in a way that limits pricing power. Røkke described different inventory positions across different players, each with its own pricing policy, which "kind of also confuses the picture and makes it difficult to compensate volume drops entirely through price increases." In other words: the industry is holding mismatched inventories of depreciating-and-appreciating silicon, and nobody wants to be first to blink on price.

Komplett is short the memory supercycle — and the market knows it

Strikingly, this exact theme is showing up across Know Trend's global keyword set and across other reporters. memory pricing and memory cost are established market-level themes, and KOA.OL — another Nordic name, in auto parts — flagged the identical canonical "memory price" theme, tagged with the very same keyword id. This isn't a Komplett problem; it's a memory cycle problem landing on the consumer-electronics channel. The tape adds an important twist. high bandwidth memory is a 360-day leader among market movers, yet it sits in the 90-day decliner list, alongside high performance compute and data centre capacity. The AI-memory trade has cooled recently — which is precisely when the cost pass-through into consumer devices gets harder to sustain, because upstream pricing power and downstream inventory behaviour stop moving in the same direction. For Komplett, that's the crux. Its gross margin is effectively the inverse of the memory makers' pricing power. When DRAM and HBM repriced violently upward, memory producers captured value and a Nordic e-tailer with a 2/3 memory-exposed mix surrendered it. If the upstream trade has peaked, the pressure could ease — but the inventory overhang at competitors means relief arrives late, and through volume, not margin. Røkke's own framing — consumers "are not used to it," but if this sustains into 2027 and 2028 equipment will age out — is a bet on a replacement cycle that he himself called "extendable," and easier for consumers to defer than businesses.

The verdict

Komplett is a genuinely well-run balance sheet (leverage 2.4x, down from 3.0x; NOK 1bn+ liquidity; facilities to 2030) absorbing an industry shock it cannot control, with a mitigation toolkit — private label, category rebalancing outside traditional gaming related categories, refurbishment, cost discipline — that is real but finite. The stock case now rests not on Komplett's execution but on a memory price curve nobody on the call was willing to forecast. That's an uncomfortable place for a small-cap retailer to be, and the rhetoric — "a challenging quarter" softened by 2.4x leverage and facility extensions — is doing a lot of work to distract from the fact that the next two quarters are being outsourced to the semiconductor market.