Time series data is special because the order in which observations occur is meaningful. For example, looking at sunglasses sales at Sun-Glasses-For-All (an imaginary brand), we can see that knowing yesterday’s sales gives you information about today’s sales. And sales this time last week gives you information about today’s sales …
There’s a lot of confusion from marketers about the right way to use lift tests or incrementality experiments to calibrate the results of a media mix model (MMM). We see questions like: You need to be thoughtful when incorporating information from incrementality experiments into an MMM, so in this post …
Nearly every business has an annual cycle: seasons where the business waxes or wanes, or specific dates like Cyber Monday, Memorial Day, or Mother’s Day, which see enormous spikes in sales. Imagine you have a business that sells hot wings — when do you think the busiest time of the …
You’re probably modeling seasonality the wrong way Read More »
At Recast, we often speak with practitioners who are trying to build their own MMM in-house. While we think there are great reasons to use a tool like Recast instead of building in-house, many people want to try it themselves before committing to working with a partner. In this post, …
One of the core problems to solve when doing marketing mix modeling (MMM) is handling how the results from a previous period ‘carry over’ into the next period. In marketing we know that the full effect of advertising isn’t felt on the same day or week in which the campaign …
Adstocks: Accounting for the Lagged Effect of Advertising Read More »
You cannot double spend and expect sales to double. Advertising on auction-based platforms like Meta or Google exhibits diminishing returns, meaning as you increase spend, efficiency suffers. When you continue to increase your advertising spend you see increased customer acquisition cost (CAC) and lower return on investment (ROI). Essentially, there’s …
Diminishing Returns: Accounting for Channel Saturation Read More »
Disclaimer: This blog post was written by an external contributor about detecting and addressing heteroskedasticity in marketing mix models. The approach laid out in this document is not reflective of how Recast approaches heteroskedasticity and model validation. The actual model specification can be found in the technical model specification. Heteroskedasticity …
How to fix heteroskedasticity in your Marketing Mix Model Read More »
Multicollinearity is one of the hardest problems in marketing analytics. If you took a statistics course in undergrad, you might have covered it; it describes a phenomenon where two or more variables in a model are correlated. Let’s bring it to marketing to make it more tangible. Imagine we’re trying …
How D2C Brands Can Deal with Multicollinearity in their Marketing Mix Modeling Read More »
MMM is hard enough — but it’s just impossible if you don’t have the right data. Getting good, clean, and ideally automated data is critical to run a really good model because, without it, your model will take so much longer to set up and train – and it won’t …
What Data is Needed for Marketing Mix Modeling (MMM)? Read More »
Marketing Mix Models (MMM) are very complex, and the choice of the parameters your data scientists choose can really influence the results you will get. If the model allows for your analyst to guess, check, and set parameters until things look right to them, there’s a high chance the end …
How to reduce bias in your marketing mix modeling (MMM) with better parameter settings Read More »
