Q4 2025 ops, what we learned.
Our busiest quarter ever. 47 percent more GMV than Q4 2024. The campaigns that worked, the ones that did not, and the three process changes we rolled out in January.
Q4 2025 was our highest volume quarter by a meaningful margin. Total GMV up 47 percent against Q4 2024. Three reasons matter: Black Friday was bigger than expected for D2C customers, crypto launches concentrated into November as token market heat cycles aligned, and a cohort of creator customers crossed Premium eligibility thresholds they had been working toward since summer.
What worked
Managed Campaigns scaled cleanly. We ran 18 Managed campaigns across the quarter, up from 11 in Q3, and the win rate (defined as meeting the stated campaign objective) was 94 percent, which is the highest we have tracked. The team composition stayed stable despite volume, no burnout events, and the strategist coverage held on every launch day war room.
Bookmark fulfillment volume was our fastest growing SKU, up 180 percent quarter over quarter. This tracks the late November algorithm re-weighting. Customers who paid attention to the change and layered bookmark signal saw compounding results through December.
What broke
Drip scheduler latency under peak load. On two separate days in late November, the scheduler ran 10 to 15 minutes behind target during US peak hours. No SLA misses because shipments still completed within their tier windows, but the tight tolerance we usually run was not tight. Fix shipped mid December with thread pool auto scaling and validated under synthetic Q1 load.
One enterprise engagement had an embargo communication failure on our side where we asked a customer to reschedule an NDA launch by 72 hours for internal ops capacity reasons, which was a reasonable ask handled poorly. Customer accepted the reschedule but the intake call documentation had not flagged the capacity concern up front. New intake questionnaire now captures launch window rigidity explicitly.
Three process changes
First, dedicated scheduler capacity windows for Managed tier shipments so they do not compete with direct order drip at peak. Second, tighter intake documentation with a dedicated launch window rigidity field. Third, moving the retention monitor poll interval from hourly to every six hours to reduce API load, which freed capacity and had zero effect on warranty fidelity because drops cross the threshold slowly.