The more experiments you conduct

the more you’ll learn about how your audience perceives your brand. Not knowing for certain what is working with your ads will be a thing of the past.


Introducing the Creative Experiment

Testing Framework

A Creative Experiment is a testing framework used to explore a hypothesis with the goal of uncovering insights that can inform a creative, messaging and media strategy.

Data Insights

We use ads, behavioral, preference and perception data, to build our predictive models. We run the experiment until we reach at least a few hundred submissions, which helps your Creative Experiment’s model reach results that bear statistical significance.

Experiment Variants

In a Creative Experiment, there is always a control variant and at least two experimental variants that respond to a specific hypothesis.

Getting Started by Asking the
Right Questions

We facilitate three categories of hypothesis: Creative, Audience, or Inventory (Publishers).

Creative Based Hypothesis

  • TV Campaign Ads vs. Digital Campaign Ads
  • Custom Creative vs. Generic Creative
  • Narrative A vs. Narrative B

Audience Based Hypothesis

  • Google Audience vs. Facebook Audience
  • Demo Audience vs. Contextual Audience
  • Impact on Gender, Age, Geo and Income

Channel Based Hypothesis

  • Category Publishers A vs. Category Publishers B
  • Open Exchange vs. Private Exchange
  • Direct vs. Programmatic Direct
  • Exchange vs. Direct

Designing a Creative

We’ll work closely with you to design the experiment, promotion, and customize the design of the end user experience.


Launching a Creative

We don’t stop working, even after your Creative Experiment is designed and launched. With our intuitive promotional white-labeled and web-based application, your target audiences are naturally incentivized from the time they see the promotion's ad on Google or Facebook, through to their submission on your Creative Experiment.

Our Ads Management team helps set up the ad campaigns that will promote your Creative Experiment among the target audiences in a way that eliminates any cross-pollination of users from one audience to another making for cleaner analysis.

During the time your ads are live, we’ll fine-tune their performance to help you get the cheapest possible submissions to your experiment’s promotion. Our data insights team will then work with you to unpack the findings and apply them to your broader campaign strategy.

Unlocking More
Predictive Models

As your promotion receives more submissions from your Target Audience, the platform is able to unlock more advanced and predictive models. Each model is built from scratch using the data collected from your Creative Experiment.

Reaching statistical significance is great, but getting even more concrete, (and often surprising!) insights sets your team up to conduct increasingly more valuable Creative Experiments.

Studies show that, on average, media placements only account for about 30% of a brand campaign's success while the creative drives 70%, according to Dynamic Logic.