Why Custom Algorithms?

Adam Heimlich
2 min readJun 17, 2020
This won’t help you compete with Facebook

In cars and clothes, “custom” implies a level of fanciness not everyone needs. In media-buying, though, custom is the next wave of innovation, naturally following from broad adoption of search, social and programmatic. Custom algorithms are how leading brands are earning back their edge in media.

Leading brands differentiate with bold strategies and unique creative. Those are not well served by generic bid rules and unrestricted data-sharing. Shared algorithms produce cookie-cutter media buys. The platform giants will never deliver custom: they prioritize goals of their own when researching and developing models for predicting who everyone’s next target should be.

Over the last two years, upon the insistence of a few firms with extensive technology and analytics resources, all the major DSPs introduced the ability for advertisers to “Bring Your Own” predictive model to power their bidders. When they do, everything a brand knows about its audience and message can be leveraged in a bespoke algorithm, and everything a brand learns through iterative testing can be captured.

It’s not fancy: it’s an order of magnitude more powerful, whether for acquisition or influence.

The founders of Chalice have been helping clients benefit from the latest in marketing technology and data science for as long as digital advertising has existed. Of all the new waves to break upon the industry, custom algorithms are the most exciting. It’s with this transition that any brand can become a platform, and position to profit from new advances in machine learning.

For companies too independent to outsource strategy, custom algorithms are what search, social and programmatic were leading up to all along.



Adam Heimlich

CEO of Chalice Custom Algorithms, an optimization platform for ad delivery.