The purpose of Pareto Ads' automations is to increase your performance and productivity. For this reason, we work with machine learning / statistical algorithms and not just a rules system.

Thus, our algorithms depend on:

a) Minimal sample volume - The minimal sample guarantees that optimizations are statistically based, that is, they will really bring the expected performance. Algorithms with no minimal sample work by chance, generating a certain volume of data and conversions, but with little to no certainty about future results.

b) Statistical confidence ranging from 95% to 99% - The user's profile directly impacts the volume of one-clicks created. An algorithm can understand that one ad performs better than another with 88% statistical confidence and, even so, not generate a card. We prefer to wait longer, generate less quantity and more quality / certainty for each recommendation.

This system allows us today a high degree of account optimization, from small to large advertisers.

How does it look in my reality?

The tendency is that, the higher your investment, the more one-clicks will be created. More investment means more data. More data means more algorithms reaching the minimal sample. More minimal sample means more algorithmic definitions, generating MORE ONE-CLICKS.

If your investment is low, don't worry about card volume or if we're not generating all types of cards. The algorithm will suggest an optimization at the right moment. And this moment could be next month. Or today. Or it could have been yesterday. Pareto Ads will learn from each dataset, generating completely different patterns in each account.

But don't worry. It's more important to have high-quality optimizations. Not a big  quantity thereof.

Did this answer your question?