The Problem with “Best Practices”
In agency land, best practices have a way of turning into facts. You hear the same ones repeatedly: video always outperforms static, PMAX is a black box, this is the right way to structure a funnel.
They can be directionally helpful, but they are not universally true. In the current environment, that is more obvious than ever. With signal loss, increased automation, and constantly shifting consumer behavior, what worked for a restaurant brand last month can easily fall apart for a fintech app today. Even within the same brand, performance can shift from one quarter to the next.
At a certain point, relying on benchmarks alone stops being strategic and starts becoming guesswork.
The risk with testing a new tactic is not launching something that fails. It is launching something without a clear way to understand whether it worked.
It is common to see campaigns go live based on instinct, followed by weeks of trying to piece together a narrative from whatever metrics look acceptable. When success is defined after the fact, the problem is not the campaign itself. It is the lack of a testing framework going in.
A More Intentional Approach
If the goal is to get real answers, the approach has to be more deliberate.
To move from guesswork to growth, we use a structured, five-pillar approach that connects high-level strategy to how testing actually gets executed:
Causal Prioritization: We focus on the decisions that have real budget implications, specifically the questions that would change how you invest your next million dollars, not just what is easiest to test.
Causal Isolation: We design tests to limit variables so that any lift can be clearly attributed to the change being made, rather than a mix of competing factors.
Matched-Market Design: In a privacy-first environment, matched-market testing has become one of the most reliable ways to understand incremental impact. It helps answer the question of whether those conversions would have happened anyway.
Power Analysis: Test duration and scale are determined upfront to ensure results reach statistical significance. Ending a test early because performance looks strong often leads to false confidence.
Clear KPIs: Success metrics are defined before launch and tied to business outcomes. This avoids situations where an awareness campaign is later judged on immediate return, or where teams shift KPIs to fit the result.
What Changes When You Get This Right
We recently put this exact framework to work for a regional quick service restaurant client launching a new beverage line. Rather than leaning on industry benchmarks, we deployed a rigorous matched market design pairing specific test locations in diverse regional scales with tightly aligned control twins exhibiting nearly identical historical sales trends. By mapping out a precise five week flight, we established clear KPI thresholds upfront tied directly to our goal of driving sales. The result? The brand moved past post campaign guessing, walking away with definitive proof of how localized media scale directly converts digital awareness into physical sales.
When testing is built into the plan from the start, the conversation shifts. The focus moves away from defending performance and toward understanding what is actually driving growth.
Best practices still have a role, but they are a starting point, not a conclusion. In today’s environment, testing is what provides the evidence.