How to Win at Paid Social
Paid Social is a competitive arena where advertisers vie for the attention of a limited pool of users. The most crucial lever for outcompeting in ad auctions is creative. This document outlines my approach to success, divided into two parts: 1) Creative Testing and 2) Scaling Creatives.
Creative Testing
A Paid Social campaign is like an engine—fuel it with high-quality creatives, and it will perform well; feed it low-quality creatives, and it will break down.
A Meta campaign with mediocre ads is like a car on fire.
My creative testing methodology is designed working backwards from the goal of uncovering the top 1% of ads to feed a campaign:
Ads follow the Pareto Principle: A small percentage of concepts drive the majority of results.
Finding these rare top-performing ads unconditionally requires extensive testing.
To enable high testing volume, you need an affordable, upper funnel testing KPI.
Find an Early Funnel Testing KPI
Some marketers object to relying on upper-funnel KPIs for creative testing, arguing that mid- or lower-funnel KPIs offer more consistent predictions of creative winners. While this may be true, the tradeoff is too great. The higher costs associated with mid- or lower-funnel KPIs inherently limit the number of tests you can run, slowing down learning cycles. Furthermore, the reduced data volume from high cost per events often results in unreliable or misleading findings. And, even for those creatives that initially outperform in testing, it is not a guarantee for success as most will fail in the scaling campaigns anyway. Additionally, mid- or lower-funnel KPIs make it difficult to compare creative learnings over time, as product changes can have significant impacts to conversion rates.
The goal of creative testing isn’t to identify outright winners (which isn’t possible)—it’s to filter out obvious losers before they reach the scaling phase. By analyzing the upper-funnel KPI range for historical winners, you can determine an aggressive cost per event that sufficiently predicts creative success. More importantly, using upper-funnel KPIs will enable significantly higher testing volume, which is the cornerstone of success in Paid Social.
Increase Testing Volume
Most User Acquisition teams produce a fraction of the ad volume they actually need. The importance of testing volume cannot be overstated. Doubling the number of tests doesn’t just double the winners–it triples them, because learnings compound over time. By looking back and identifying patterns among winners and losers, a UA team becomes more adept at making successful creatives.
Here are some ways to increase testing output:
Establish a Weekly Testing Cadence: A habitual routine for designers ensures a steady flow of new ideas.
Test Multiple Hooks Per Concept: Multiply your initial ad by x4-5 by testing different copywriting, visual, and edited hooks.
Use Agencies on a Pay-Per-Spend model: Agencies are a great way to source UGC content. A top-tier pay-per-spend agency is in tune with organic trends, has large teams for storyboarding, acting, and editing, and are mutually incentivized to find winners.
Make High-Quality Creative
Successful creatives share universal elements. I outline my key pillars here:
Center on Key Value Proposition: What are the top one or two things that users care about? Focus on these. Disregard the rest.
Engage Immediately: Social media users have small attention spans. You have one second to captivate users before they scroll away.
Keep it Simple: Complexity hinders communication. Don’t overwork a user’s brain. Keep things simple.
Make Native Ads: People dislike out-of-touch and boring ads. Make organic looking content.
Scaling Creatives
The second phase of Paid Social focuses on effectively managing scaling campaigns.
Scaled Spend Matters Most
If the KPI for creative testing is an upper-funnel metric, the KPI for scaling campaigns is spend volume.
Why not use CAC or ROAS as the scaling KPI? Simply put, an ad’s value isn’t determined solely by its CAC but also by how much spend and revenue it can generate. An ad with excellent metrics is meaningless if it fails to scale effectively. On the other hand, ads that scale do so because they meet your bid targets in the long run.
Many marketers make the mistake of interfering unnecessarily with Meta’s self-regulating ad optimization process, often doing more harm than good. In today’s era, Meta’s machine-learning algorithms are far better than humans at identifying and scaling winning ads. As long as the appropriate optimization event—such as Purchase or ROAS—is selected, Meta will make the right allocation decisions over time.
The cost of prematurely removing top-performing ads can be substantial due to lost opportunity cost. The best ads can run for months, often accounting for a disproportionate share of total spend. In my experience, I’ve seen how only 1% of ads can drive 60% of overall spend.
When to Pause Ads
There are several reasons to avoid pausing ads solely based on cost per event.
Rapid scaling: Meta may scale an ad quickly within a short timeframe, leading to an initially high cost per event that often normalizes over time.
Apples-to-oranges comparisons: Older ads have had more time to convert than recent ones, making direct comparisons of cost per event misleading.
Down-funnel performance: Some ads with higher cost per events may deliver stronger conversion rates further down the funnel or yield higher revenue per purchase than average.
While cost per event can serve as a supplementary metric, the primary factor for pausing decisions should be the ad’s trending spend percentage over time. In 95% of cases, ads should only be paused if they fail to scale after sufficient time or show a meaningful decline in share of spend.
Automate to Streamline
Automating redundant or repetitive tasks is essential for streamlining campaign management.
Cloud providers should update event data daily into preconfigured Google Sheets dashboards, enabling real-time insights. Prebuilt pause rules should automatically identify underperforming ads for removal, while other tools highlight the best new ads to refresh campaigns. Additionally, copyable JSON strings should be prepared for use in the Meta Web API Interface, allowing for the bulk removal of up to 40 ads at a time.
By automating these processes, you can save dozens of hours each week, reallocating that time to higher-value tasks. This approach allowed me, at Hopper, to manage a multi-million-dollar User Acquisition program across 3-4 channels with just one other marketer.
Conclusion
Success in Paid Social requires a balance of rigorous creative testing, effective scaling strategies, and smart automation. By adopting these practices, advertisers can uncover top-performing ads, maximize spend efficiency, and maintain a competitive edge in an ever-changing marketplace.
Hope you enjoyed the read! If you’re a B2C business looking to achieve higher ROI on your ad spend, I’m currently taking on a limited number of new clients. The best way to get in contact is to click the button down below—I’d love to help drive meaningful growth for your business.
