Episode #9 | How to Use Marketing Mix Modeling to Increase ROI

🎙️ Unlock the secrets of Marketing Mix Modeling (MMM) to optimize your marketing campaigns and maximize ROI. In this episode, Ievgen, Head of Marketing at OWOX, teams up with Analytics Solution Architect Ilya to break down the essentials of MMM.

🚀 Discover how MMM can help you identify which marketing channels are driving the best results and learn how to allocate your budget smarter. From understanding the 4Ps to leveraging regression models for actionable insights, this episode simplifies the complexities of modern marketing analytics.

📈 Ready to transform your marketing strategies? Learn how to use MMM to analyze performance, predict future outcomes, and refine your marketing plans based on real data. Start applying these techniques to see the impact on your bottom line.

👉 Take the first step by exploring the resources and tools at OWOX BI.

💡 Subscribe to The Data Crunch Podcast for more expert insights into leveraging data for smarter business decisions.

Ievgen:
Hey everyone! Welcome to the 9th Episode of The Data Crunch Podcast!
My name is Ievgen, I am the head of Marketing here at OWOX.
If you’ve watched our previous episodes, you know, I typically act like a speaker, but today is different as now I want to really dive deeper than my analytics knowledge is,
into something that might sound like a fancy term but is super useful for anyone in marketing—Marketing Mix Modeling, or MMM…
So, if you’re wondering which of your marketing channels is really bringing in sales or if you’re just throwing money at ads, then this episode is for you.
So to unpack this, I’ve invited our Analytics Solution Architect, Ilya, to help us break it all down for you.
Also, please don’t forget to subscribe so you don’t miss out the next episode on reporting. With this being said - let’s get started.
Ilya, are you ready to make Marketing Mix Modeling easy for everyone?

Ilya:
Absolutely, Ievgen! MMM might sound complex, but it’s a desired topic for many marketing teams. Just imagine, how easier your life could be, if you had a report that points you exactly what to do with your campaigns. We’ll keep it straightforward and dig into how the modeling helps you get more out of your marketing budget.

Ievgen:
Great! So, for those new to this concept, what exactly is Marketing Mix Modeling?

Ilya:
Alright, let’s keep it simple. Marketing Mix Modeling is like a tool that shows you which of your marketing activities—things like ads, emails, social media—are actually helping to drive sales. Think of it as a bird’s-eye view of all your marketing efforts, showing you what’s really working. On top of that, the word “modeling” comes here for the action scenarios. In simple terms, what actions you should take with your marketing and how you will be rewarded for that. 

Ievgen:
That sounds super helpful, especially with so many channels available now. You’ve got Facebook ads, Google Ads, email marketing, SEO, affiliates, and so much more. It can get overwhelming!

Ilya:
Exactly! The more channels you add, the harder it is to figure out which ones are worth the budget. That’s where MMM shines. It gives you clarity on which channels are making a difference, so you’re not just guessing.

Ievgen:
Alright, so Marketing Mix modeling helps to see the big picture. I know there’s something called the 4Ps in MMM. Can you explain what those are without getting too technical?

Ilya:
Definitely. The 4Ps are the basement of your marketing mix: Product, Price, Place, and Promotion.

• Product: What does your audience need?

• Price: How much should it cost?

• Place: Where should you sell it?

• Promotion: How are you spreading the word?

These four things have been around since the 1960s, and they’re still the foundation. But in today’s world, we sometimes add a few more elements like People and Processes to cover all bases.

Ievgen:
Got it. So now, what are some real-world ways companies can actually use Marketing Mix modeling?

Ilya:
Sure! There are three main components:

#1. Performance Analysis: This helps you see which channels—like Google Ads or Meta—are really worth the money. You’re not just looking at high-level numbers but actually digging into which channels are bringing in sales. It always works better when the business owns their data, not using a 3rd party black box tools. 

Ievgen: Like Google Analytics, right?

Ilya: Yes, exactly. The second component is Predictive Modeling: This is like using data and signals to make educated guesses. The model looks at past trends to predict what could happen in the future. For instance, if spending more on social media ads has led to more sales in past, the model might navigate you to further budget increase. In the advanced models, we’re adding one more input: the audience capacity that a specific campaign allows you to buy for a reasonable price. That helps to draw the marketing mix curve.
Ievgen: amazing. You’ve mentioned there is the third one?

Ilya: So our component number 3 is Marketing Strategy
Here, you’re pointing the direction, that your campaigns show follow. For example, one of the most popular strategies is to get more Net Revenue meeting the better ROI down that road. So, you’re adjusting your direction applying the business inputs you have in the team. 

Ievgen:
I love the idea of using data as a “marketing GPS.” Now, with all the changes in 2024, why is MMM especially important?

Ilya:
Good question. A big reason is privacy. Tracking users isn’t as easy with all the privacy laws and limits on cookies. However, if “consent” users data is enough, MMM can give you insights without needing to rely on tracking every click. So it’s almost perfect for today’s privacy-first landscape.

Ievgen:
Makes sense! So, if someone wants to start using MMM, what’s the basic process?

Ilya:
There are three main steps:
Step #1: Planning: Look at your past data—sales, ad spend, etc.—to get a baseline of where you’re at.
Step #2: Model Educating: Define variables to make short-term predictions.
Step #3: Results Tracking: This is the ongoing part. You’re adjusting your direction knowing how the campaigns are actually performing. 

Ievgen:
So, Marketing Mix modeling isn’t just a one-time thing; it’s something you keep working with, right?

Ilya:
Exactly. It’s flexible. As the market changes, Marketing Mix modeling helps you adapt, so you’re always optimizing. Basically, those models are not working until you build the right “updating” process. 

Ievgen:
Alright, you mentioned something called regression models before. Can you break that down in simple terms?

Ilya:
Sure! A regression model is just a tool that shows how changes in one area affect another. For example, if you increase your ad budget on Facebook ads, it shows you how much it might impact sales. It’s all about finding those connections.

Ievgen:
So, a regression model shows how one thing, one component, like increasing your ad budget on Facebook, affects another, like sales. But how does it actually work step by steps?

Ilya:
Good question! Let’s break it down.
The first step is figuring out the right variables—all the factors that could influence your sales. Think about things like your ad spend on different channels, maybe seasonality, pricing, and competitor activities. These are all pieces of the puzzle that might impact your results.

Ievgen:
Got it. So, those are like the ingredients in a recipe?

Ilya:
Exactly! And then, the model actually measures the impact of each ingredient. It’s like the model is saying, “If you add a dollar to Facebook ads, here’s how much it might boost your sales.” It gives you a sense of which factors are really driving your results.

Ievgen:
Oh, so it’s not just about looking at one ad—it’s about finding the real connection between what you’re spending and what you’re getting back.

Ilya:
Exactly! That’s the power of it. The model helps you see the connections between actions and results. For instance, it might show that Facebook ads have a stronger impact on sales than, say, a black friday email campaign. So, you can decide where to put more of your budget based on those connections.

Ievgen:
Makes sense. And once you set it up, are you done? Or can you update it?

Ilya:
Oh, you can absolutely update it. The model is dynamic, meaning you can keep updating the date ranges or re-educate model with new variables as you run more campaigns. As the main controller is budget, and it is something you adjust manually, it’s better to keep the eye on the reports at least once a week. 

Ievgen:
That’s pretty cool. So, in a nutshell, a regression model helps you pinpoint which channels or actions are giving you the best return and lets you adjust as you go.

Ilya:
Exactly! It’s all about making data-backed decisions and finding those winning strategies that keep driving your business forward. 

Ievgen:
Alright, I think I’m getting it now. 

Thanks, Ilya! And for everyone listening, if you’re ready to take your marketing to the next level with Marketing Mix modeling, check out our website owox dot com for more resources or to get expert help setting it all up, as well as other valuable and fully automated reports for your business.

And don’t forget to subscribe so you don’t miss out the next episode.

Ilya:
Thanks for the fantastic conversation, Ievgen! And thanks, everyone, for joining us. Remember, a well-designed Marketing Mix modeling isn’t just about crunching numbers—it’s about getting the most out of your marketing. See you next time!