Why Do You Need A Media Mix Model?

Why Do You Need A Media Mix Model?

Companies are swimming in a sea of marketing ROI analytics: Google Analytics, Facebook Analytics, Just-Trust-Me Analytics, etc. Seemingly they tell you all you need to know about how your marketing performs. But do they?

Simple case. A prospect sees a display ad that you placed through Liftoff. Then they do a search on Google and Bing. They see your ad on Facebook. Now they come to your website to do some research. Then your partner Criteo follows them around with more display ads. Finally, the prospect caves and buys when she accidentally sees your product at a discount offered by a third party reseller of yours. Why did she buy? How do you allocate this particular sale across all the touch points?

With the abundant ROI metrics provided to you by everybody this must be a piece of cake, right? Google will say it was because of them. Facebook will inform you the sale happened because of them. Bing, Criteo, Liftoff and a bunch of others will line up to take some (or rather ALL) credit for the sale. The third party reseller will take their share anyway and inform you that it was really the discount they offered that made this sale possible. The result is that with all the ROI tools and analytics you still have no idea what really happened. So what do you do?

Attribution modeling, they say. A little vague but most of the time this means that you need to roll up the analytics provided to you by everybody into one bottom-up attribution “model”. However, putting together several conflicting and mutually exclusive claims in the same spreadsheet or dashboard does not make them less conflicting and mutually exclusive. Now you have one place to get you confused instead of many.

I suggest a two step approach to get a real answer. First, build a top-down media/marketing mix model (MMM). Second, take the recommendations from the model and use rigorous experimental design tests to validate them.

Many marketers distrust MMM because it looks to them like a black box and for a non-statistician it actually is a black box. Are you going to feel better if I tell you that inside the box live strange creatures like GARCH, VAR, ARIMA and SARIMA? No? How about Durbin-Watson (no, not that Watson) and Augmented Dickey-Fuller? But wait, there is more…

The reality is that MMM is a very complex black box. Just like a computer is. Or like a BRM H16 sixteen-cylinder car engine that can deliver close to 600 hp. Do you understand even remotely how your phone works? Not really but you still use it anyway. I have to ask you to take the same approach to MMM. If it works, use it. And the only way to know if it works is to test the recommendations from the model.

So, what is a media mix model? click here

What skills are needed to build MMM? click here

Emil Mladenov

Senior executive and expert in strategy and marketing | Podcast host | FinTech | HealthcareTech | EdTech

6 年

I wish more companies understood the need for MMM. Unfortunately, most claim to be data driven but fall for the last-mile attribution trap all the time. I’ve had numerous futile conversations trying to explain that just because the final visit that resulted in the conversion came from paid search does not mean paid search is the best performing channel. This is a huge problem especially in the small business world where executives are averse to spending on sophisticated consultants and prefer to listen to the digital SEO/SEM “guru” who charges relatively small $ and offers “smoke and mirrors” analytics in Excel files.

要查看或添加评论,请登录

Nino Ninov的更多文章

  • The Very High Cost of Employee Turnover

    The Very High Cost of Employee Turnover

    We all suspect that the cost of employee turnover is high but how high? I recently conducted a survey sponsored by…

  • Branding Needs TV Advertising

    Branding Needs TV Advertising

    It is difficult to build a brand without TV advertising. Not that it is impossible; Google and Facebook were not built…

  • Price as Signal for Quality

    Price as Signal for Quality

    It is well known that price is, among other things, a signal for quality. The famous Van Westendorp Price Meter asks…

    2 条评论
  • Skills To Build Media Mix Models

    Skills To Build Media Mix Models

    Recently I was contacted by a recruiter who was scouring the web for people specializing in media mix modeling (MMM)…

  • What Is A Media Mix Model?

    What Is A Media Mix Model?

    Written in collaboration with Roumen Vesselinov ? Companies employ different media, different campaigns, and different…

  • Rumors of TV Advertising's Death are Premature

    Rumors of TV Advertising's Death are Premature

    The rumors about TV advertising's death are very premature. Prime time CPMs are through the roof compared to prior…

    1 条评论
  • Attribution Accelerator Conference: Summary of Presentations

    Attribution Accelerator Conference: Summary of Presentations

    On October 12 I attended a one-day conference in New York City dedicated to multi-touch attribution (MTA). The…

  • Tom Brady, Benoit Mandelbrot, and Statistical Probabilities

    Tom Brady, Benoit Mandelbrot, and Statistical Probabilities

    At some point during the LI Super Bowl somebody estimated that there was more than 90% probability of the Falcons…

    2 条评论
  • In Defense of Pollsters

    In Defense of Pollsters

    Pollsters have been blamed for predicting wrongly that Clinton will win the presidential elections. I find these…

  • Can Machines Have Feelings?

    Can Machines Have Feelings?

    Of course not is the intuitive and immediate answer that this question would most likely get. But let’s not be so fast.

    1 条评论

社区洞察

其他会员也浏览了