How to scale your Meta budget for Black Friday without breaking results
Learning phase, increase steps, campaign budget, duplicating vs raising, budget scheduling and bidding: the plan for your Meta budget up to 27 November 2026.
Vasile Dingă
.comparison-cards { display: flex; flex-direction: column; gap: 1.5rem; } .comparison-card { background: #fff; border-radius: 16px; padding: 20px; border: 2px solid #efefef; } .comparison-header { display: flex; justify-content: space-between; font-size: 16px; margin-bottom: 12px; } .comparison-row { display: flex; justify-content: space-between; padding: 8px 0; border-top: 1px solid #efefef; font-size: 14px; color: #1f2937; } .comparison-label { font-size: 14px; width: 30%; } .comparison-value { width: 65%; text-align: right; } @media (max-width: 600px) { .comparison-row { flex-direction: column; align-items: flex-start; } .comparison-value { text-align: left; margin-top: 4px; } } Black Friday falls on Friday, 27 November 2026 this year, and Cyber Monday on Monday, 30 November. The scenario I see often: Meta campaigns run well all November on a small budget, then on Thursday evening the budget is tripled in one go. On Friday morning the cost per order is higher than on any other day of the month, exactly when it matters most. Below I show how to raise the budget gradually, what the learning phase means, when it's worth duplicating a campaign and how to use budget scheduling for the peak days. At the end there's a planner that builds the day by day calendar for you. 1. Why a sudden increase can break results Meta delivers ads through a system that learns, for each ad set, whom to show them to in order to get the event you optimise for, usually a purchase. When you create a new ad set or make a significant edit, the ad set goes back into the learning phase. During this period, costs are usually less stable. In its developer documentation, Meta exposes each ad set's learning status and the date of the last significant edit, and states plainly that significant edits send the ad set back into learning. The well-known threshold for exiting learning is about 50 optimisation events within a week of the last significant edit. Check the current wording in the Meta Business…