Devsisters Posts Q2 Loss: Can Cookie Run Bounce Back?

The headline number

Devsisters, the Korean mobile studio best known for the Cookie Run franchise, swung to an operating loss of 16 billion won (roughly $12 million) in the second quarter of 2026. For a company that has spent the past several years talking publicly about cost discipline, slipping back into the red raises an obvious question: is this a one-quarter timing issue, or the start of a structural problem?

Management's explanation is straightforward. Marketing and early live-operations costs for a newly launched title landed almost entirely within the quarter, while leading indicators — active users, retention, and paying-user conversion — improved. That pattern is familiar to anyone who reads game-company earnings: in a launch quarter, spending arrives first and revenue arrives later. The explanation is credible. Whether the revenue actually follows is the part worth watching.

Where the money went

Devsisters concentrated its user-acquisition spending on Korea, the US, and Taiwan — the markets where it says the underlying metrics justify the investment. Rather than spreading a global budget thinly, it funneled it into territories with proven retention and monetization.

Strategically, that is the more efficient approach. Accounting-wise, it is brutal. Marketing costs hit the income statement immediately, while the lifetime value of the players those campaigns bring in trickles in over the following quarters. In that sense, a meaningful chunk of the Q2 loss can be read as pre-purchased Q3 and Q4 revenue — but only if churn among those new players stays contained.

Three pillars of the second-half plan

Keeping Cookie Run: Crumble alive

The new title, Cookie Run: Crumble, is slated for a content update cadence of once every two weeks following its first patch. For a casual collection-style game, a fortnightly cycle is aggressive. The intent is clearly to defend the first three months post-launch — the window where casual titles bleed players fastest — with sheer content volume.

The open question is sustainability. Shipping content every two weeks requires a development team locked in place, which sits awkwardly against a cost-reduction narrative. The metric to watch in the second half is not gross revenue but the ratio of marketing spend to revenue.

The global Cookie Run Classic relaunch

Cookie Run Classic went into global service on June 25, meaning its results land squarely in Q3. Nostalgia-driven relaunches of an original title carry a structural advantage: acquisition costs tend to be lower, because the game is recalling players who already know the brand rather than introducing it cold.

The catch is that such titles typically spike hard and decay fast. When Q3 figures arrive, the number that matters is not the launch peak but the floor the game settles onto a few months later.

Beyond games: the IP business

Devsisters' non-game licensing arm — trading card games (TCG) and character merchandise — grew again this quarter. It is not yet large enough to offset game revenue, but it matters for a different reason: revenue uncorrelated with launch cycles.

Physical TCG in particular is a direct test of whether character recognition converts into sales outside of an app store. If the long-term ambition resembles the Sanrio or Pokémon model — cycling character equity through games, merchandise, and media — then steady, unspectacular growth here is a reasonable directional signal, even if it does not move this quarter's bottom line.

What this actually tells us

This loss looks less like a company failing to earn and more like a company choosing to spend first. That is not the same as being safe. Launch-quarter losses are ordinary; losses that are not recouped in the following quarter get reclassified, fairly or not, as a growth problem.

So the next earnings report comes down to three checkpoints: Crumble's three-month retention, the post-peak floor for Cookie Run Classic, and the share of total revenue coming from IP licensing. If all three improve together, this quarter reads cleanly as an investment. If they do not, the conversation shifts from timing to cost structure — and that is a harder one for management to control.