How PrizePicks achieved a 71% lower CAC on Apple Ads ahead of Super Bowl LX

For reporting and analysis purposes, states are segmented into tiers (e.g., Tier A, B, C) that reflect differences in regulatory environments, product availability, and business performance.

Challenge

For fantasy sports and prediction markets apps, the App Store is where high-intent users become customers. Someone searching for a fantasy sports or prediction markets app during football season isn’t just browsing: they’re ready to pick a lineup. Apple Ads is how PrizePicks reaches that audience.

In PrizePicks’ most important markets, customer acquisition cost (CAC)—the cost of acquiring a user who takes a meaningful action after installing—was running at more than twice the team’s internal targets for non-brand keywords. Spend was high, but installs from those terms weren’t following at the rate they needed.

PrizePicks’ previous approach was highly complex, requiring the team to manage multiple campaigns and ad groups, with all budget allocations and bid adjustments handled manually. As the number of campaigns and ad groups continued to grow, this approach became harder to sustain as a long-term strategy.

The year before, one of PrizePicks’ main competitors had increased bids on their brand keywords as search demand spiked around Super Bowl LIX. With Super Bowl LX six weeks out, the team couldn’t afford to enter that window with non-brand underperforming—not when they needed to perform across brand, discovery, and competitor-related search terms.

Solution

PrizePicks partnered with AppTweak to rebuild their Apple Ads bidding strategy across priority markets.

Using Campaign Manager, AppTweak’s consultants built custom automations—rules that automatically adjust keyword bids based on performance.

Rather than applying one set of automation rules across all campaign types, the team built separate logic for Brand and Non-Brand. Brand campaigns ran on tighter CAC thresholds, while Non-Brand ran on flexible rules with more room to pursue new users, even at a higher initial cost. Each set of rules had different CAC targets depending on the market.

Day-parting—scheduling bid adjustments to run at specific times—was added for peak hours before Saturday games, when users were most likely to download.

As the weeks progressed, automation rules were refined. Conditions that had been filtering out viable brand keywords were loosened; rules for cutting bids were tightened so budget could shift faster toward more efficient terms. This strategy also improved budget pacing, enabling the team to distribute spend more evenly throughout the day and avoid exhausting budgets too early.

In parallel, the team used AppTweak’s Ad Intelligence to analyze estimated competitor activity and keyword trends from the previous year’s Super Bowl period. Those insights allowed PrizePicks to set bid ceilings and allocate budget in advance, rather than reacting to competitive pressure after the fact.

In the final days before kickoff, CAC ceilings were adjusted and budgets scaled +140% on Brand and +218% on Non-Brand.

Results

PrizePicks started seeing results within the first week of activating the automation rules.

Customer acquisition cost (CAC)

Within the first seven days, non-brand CAC had already fallen 32% in Tier A markets and 42% in Tier B markets.

By game week, non-brand CAC was down 71% from baseline in Tier A, and non-brand CAC was down 78% in Tier B. Brand CAC dropped 87%.

Conversion rate (CVR)

At Super Bowl peak, non-brand ad campaigns were converting 155% more users in Tier A markets. For brand campaigns, conversion rates increased 53% in Tier B markets.

Scale and efficiency

With non-brand budgets up 218% and brand up 140% for the event window, CAC on both stayed below the targets set in advance. As the automations kept running after the game, brand CAC in Tier A markets continued to fall.

Over six weeks, AppTweak ran 4,452 optimizations for PrizePicks—93% of them automated—allowing the team to consistently improve performance against their original CAC targets.