Co-founder/CEO of Vujà Dé Digital , on a mission to mindful capitalism and reinvent the media agency model.
As marketers have tried to solve the problem of third-party cookie leaks in recent years, new, smarter ways to spend media dollars have emerged. With the advancement of cookies, it will become more important than ever for triggers to track desirable marketing behavior.
How do you know if you are reaching the right consumers and if your media and marketing mix is working? Simply waiting for the cash register to chime isn't enough, and mixed media models (MMM) or multisensory attribution (MTA) models are being replaced by dinosaurs. What does it replace? I think it's a big data media mashup model.
The ultimate solution to media purchasing and performance analysis can be found smarter through big data media mix modeling. It's an increasingly complex way to track and predict future marketing performance and manage your media investments to increase your return on investment and return on investment.
The democratization of machine learning, the power of big data processing in virtual machines, and access to big data analytics from Google and Amazon have changed the rules for strategic media investments and forecasting business performance.
Where does the big data media mix model come from? Direct Response Television (DRTV) was the prototype for BDMMM. This marked the beginning of combining investments in linear media with digital campaigns.
So what makes big data media mix modeling so powerful compared to legacy campaign analytics and media planning options? Media mix modeling based on big data provides a unified view of marketing. View your media spend and campaign performance by bringing together all your marketing, economic, political, seasonal and even retail data into one view.
The beauty of combining these is that you get a true picture of what happens over time for each possible variable. Big data media mix modeling allows marketers to see a direct correlation in sales by adjusting channel spend based on media types, seasons, and spending levels. Ideas can also get smarter over time by incorporating machine learning.
AI compares spending levels to show you where you are overinvesting so you can cut back or reinvest in more efficient channels for higher ROI. For example, instead of using intuition and a few data points to increase your Instagram spend, you can tell TikTok to do better by directly linking to other media elements.
A big data media mix model can tell you the recommended media mix and where to increase or decrease costs based on actual performance and the interaction of all revenue and sales variables. Its goal is to provide less guesswork and more informed decisions in the real world based on statistical facts rather than guesswork.
What's wrong with multisensory referral (MTA)?
Naming problems caused the MTA to crash. This tactic always fails because of branding issues which become problematic for all types of media. You need a tag for every type of media, including trademarked and unregistered, and with the walled gardens of Google and Facebook, you'll never have a full 360-degree view of your campaign performance across all types and levels of media. Marketers are forced to make decisions based on incomplete or incorrect data, which turns continuous improvement into nonsense.
What about media mix modeling (MMM)?
While it is not as cumbersome and expensive as modeling a large mix of data, it is not suitable for in-depth analysis. Big data media blend modeling brings data up to computational speed in less time and in more increments with an extremely high level of confidence. You can shape it according to your KPIs. No MMM or MTA can boast of such a feat.
So, one might ask, why don't marketers take this smarter approach to media planning and purchasing? Many things can make it difficult to implement, although it is by far the most sophisticated marketing planning and analysis tools and strategies available to marketers today.
Cost, the media investment required to calibrate the model, and the skills required to implement it are the three most common barriers to deploying and operating big data media blending models. You need access to a data scientist or data engineer, hundreds of thousands of dollars to access cloud data management and analytics services from Google or Amazon, and a large advertising budget for at least three months.
Another potential issue could be accessing your retail data if you are a FMCG brand. This can be overcome by paying for the data to get a complete overview of the performance. The truth is that BDMMM will not be effective if you lack a large percentage of your retail sales data.
The key to successful big data media modeling is access to data that is as accurate and complete as possible. Data or microdata located in different locations that are not easily collected over time can lead to an inaccurate model.
The power and accuracy of big data media mix modeling will pay off again and again. As the saying goes, there is no gain without pain.
The Forbes Agency Council is a gated community of successful PR, media strategist, creative and advertising agencies. Am I eligible?