Black Friday Mobile Game CPMs: What to Budget For Now

Black Friday mobile game CPMs surprise nobody. Every buyer who has pushed spend through late November knows the curve by heart: prices climb from the Thanksgiving weekend through Cyber Monday as retail advertisers pile into the same inventory, they ease back through early December, and then they spike again for the final holiday push. Forecasting is rarely what separates the team that handles this well from the team that gets mauled by it. The difference is structural, and it comes down to whether the budget had room for a spike weeks before the spike landed.

Why Black Friday mobile game CPMs move before the holiday itself

Most teams treat Black Friday as one expensive day. In practice the crowding starts building a week or more ahead of it, as retail advertisers ramp toward their own promotional calendars with travel and finance close behind, and the pressure does not fully clear until well into the following week. A campaign budgeted around a single elevated Friday ends up paying inflated prices for four or five days nobody planned for. A campaign that writes off the whole week as compromised territory usually comes out ahead, even when its point estimate of the peak turns out to be wrong.

The annual MMP forecasts that land ahead of Black Friday and Cyber Monday, the ones AppsFlyer and Liftoff typically publish in the run-up to the holiday, are worth reading for shape rather than precision. Their exact percentage increases vary by vertical and by region, and mobile games do not always track the same curve as e-commerce categories. The directional signal is what holds up. Spend concentrates hard around the last week of November, and any buyer who waits for the numbers to show up on their own dashboard before reacting is already behind the curve those reports described weeks earlier.

What the forecasts actually tell you

Treat a BFCM forecast as an input to a budgeting decision rather than something to plan a campaign against literally. The useful questions are narrower than the headline numbers suggest: is the CPM increase concentrated in specific placements or spread evenly across a network, does the forecast separate install volume from CPI rather than CPM, and does it say anything about the shape of the recovery afterward? A report that only says prices will rise is not actionable. A report that says where the increase concentrates lets you move budget toward the parts of your media mix least exposed to it, and that is the actual decision a buyer faces in the fortnight before the spike.

A worked example for setting the week's ceiling

Suppose a UA team runs a blended CPI target of 2.50 dollars across its top three networks in a normal November week, on a weekly budget of 140,000 dollars. Rather than carrying that target flat into Black Friday week, it models three scenarios: a 15 percent CPM increase concentrated in its largest network, a 25 percent increase spread evenly across all three, and a 40 percent spike isolated to one high-performing placement. Under the first, it pre-commits to shifting 20 percent of that network's budget into the second-largest for the week and accepts a slightly worse blended CPI in exchange for volume stability. Under the third, it decides in advance to pause that placement entirely for three days rather than pay a 40 percent premium to keep it live, and redirects the saved budget into creative refresh testing that runs cheaper once the network's inventory normalises. None of those numbers has to be right.

What matters is that the decision rule exists before the week starts, because a team improvising its response to a CPM spike in real time, during the highest-pressure week of the quarter, will make worse calls than one working from a plan it agreed calmly a fortnight earlier.

Protecting the campaigns that cannot flex

Not every campaign survives a pause or a budget shift. A soft launch mid-test, a campaign tied to a fixed calendar event inside the game, a cohort-based experiment that needs continuous volume to stay statistically valid: these want ring-fencing before the spike, with an explicit decision to protect the budget even where that means a worse CPI elsewhere. This desk's Q4 UA Budget Planning Checklist Before the Rush covers the quarter-level version of the exercise. Black Friday week is the sharpest single test of whether that planning held, because it compresses a quarter of pricing pressure into roughly seven days.

Worth being specific about what ring-fenced means, because the phrase gets used loosely. It means the team excludes that campaign's budget from every automatic reallocation rule it sets for the week, agrees the spend cap in writing before Monday, and tells whoever owns the weekly budget review not to touch it without a separate conversation. A ring-fenced campaign that quietly gets swept into a general Black Friday cut, because nobody flagged it during a rushed Wednesday check-in, was never protected in the first place; the team simply forgot about it in a more official-sounding way.

What tends to go wrong anyway

Even a well-modelled plan meets one surprise.

A network's CPM behaves differently from what the forecast implied, or a competitor's spend lands in a placement the team assumed was safe, or a creative that had been working loses efficiency under the changed auction dynamics. Teams that recover fastest build a review checkpoint into the plan itself, usually 48 hours after the peak, instead of waiting for the standard weekly report. A short deliberate check at that point, comparing actual CPM movement against the scenarios modelled in advance, catches a bad decision before it compounds through the rest of the week. It costs far less than finding the same problem in a monthly retrospective, long after the holiday money has gone.

Related archive reading

These articles provide related context and remain subject to their stated review status.

• A Q4 UA Budget Planning Checklist Before the Rush
• Chart Watch: Mobile Game Charts Enter the Holiday Season