We transform business problems into business solutions using data, automation, and math.
We have built a technology team that combines decades of experience in data infrastructure, engineering, data science, and machine learning in marketing applications. Over the years, we’ve seen these projects go well and add large amounts of lasting value -- but we’ve seen them go wrong too. From these successes and failures, we’ve learned a few things.
Choose Projects Carefully
We pick carefully where we deploy automation or Machine Learning -- a simple regression or spreadsheet-based model can often get you 80% of the way there. Sometimes it can't.
Aim For Simplicity
Simplicity and Interpretability often lead to longevity.
But Don’t Fear Deep Learning Methods
There’s a reason why flexible albeit opaque Machine Learning techniques like neural networks are so ubiquitous in e-Commerce for forecasting, prediction, and many other applications – it’s because they work really well.
Foster A Culture of Experimentation
Constant tweaking and fast iteration, evaluated by pertinent business metrics, are the gateways to progress and success.
The Next Frontier