What Are Benchmarks?
In a Brand Lift Study, benchmarks are aggregated industry standards used to measure how your campaign performs compared to the average in your specific sector (e.g., Automotive, CPG, Fashion). Without benchmarks, a 5% lift in Brand Awareness is just a number; with them, you can determine if that 5% is "Above Average" or "Leading" for your market.
How We Use Benchmarks
Aryel compares your campaign’s results against a database of historical performance, focusing on:
Industry Average: Performance standards for your specific vertical.
Platform Average: Comparison against all studies conducted on Aryel.
Historical Performance: Comparison against your brand’s own baseline from previous studies.
How Benchmark Deltas Are Calculated
The way a benchmark delta is calculated depends on the type of question being measured:
Ad Recall and Creative Diagnosis questions
For these questions, the delta is calculated by comparing the value recorded by exposed ad recallers directly against the benchmark (BM) value:
Delta = Exposed Ad Recallers value − BM value
Ad Recall Impact questions
These are the questions whose result is an uplift — that is, questions where the metric itself is the comparison between the exposed and control groups. For these, the delta is calculated by comparing the campaign's own uplift (exposed minus control group) against the benchmark uplift:
Delta = (Exposed value − Control Group value) − BM value
In other words, the benchmark for Ad Recall Impact questions is itself expressed as an uplift, not as a flat value — so the comparison is uplift vs. uplift, not value vs. value.
Custom Questions & Benchmarks: The Role of AI
A common question is: “If I ask a unique, custom question, how can there be a benchmark for it?” The answer lies in our AI-driven categorization engine. Even when questions are tailored to your specific brand goals, we maintain benchmarking accuracy through:
1. AI-Powered Semantic Mapping
We don't rely on manual tagging. Our Natural Language Processing (NLP) algorithms analyze the semantic intent behind your custom questions.
Example: If you ask "How familiar are you with brand X?" or "Which of these best describes your knowledge of Brand X?", both map to Brand Familiarity.
The AI then automatically maps this to the relevant benchmark cluster, ensuring your "unique" question is measured against the most accurate industry data points.
2. Automated Intent Clustering
The algorithm groups custom queries into standardized KPI buckets (Brand image, Consideration, Interest, Attribution). This allows us to provide a Normalized Lift Score, showing how your specific creative execution performed compared to thousands of other interactions within the same category.
3. Vertical-Based Baselines
The AI applies industry-specific "elasticity" models. Since driving a lift in "Purchase Intent" for a luxury car is statistically different than for a soft drink, the algorithm adjusts the benchmark comparison to reflect the behavioral patterns of your specific audience vertical.
Why Choose Custom Questions?
Despite the "standardized" nature of benchmarks, custom questions are highly recommended when you need to:
Test specific Creative Attributes (e.g., "Did you find the 3D interaction helpful?").
Measure Message Association (e.g., "Which brand do you associate with sustainability?").
In these cases, our AI provides the raw lift data and instantly bridges the gap between your custom query and the most relevant strategic context.
Benchmarks aren't about matching your question word-for-word; they are about measuring your campaign’s effectiveness through intelligent, algorithmic association.