Allan Stone, CEO at Intelitics, talks about the rise of Media Mix Marketing and why campaign managers are failing to unlock its true capabilities simply because they are not also using last-touch attribution.
Media Mix Modelling (MMM) is not a magic bullet.
I’ve now seen plenty of operators throwing hundreds of thousands of dollars at it, but it won’t save them, especially when Meta kills their campaign at 3pm on a Thursday and they don’t know why.
The problem is that despite the big money being put behind MMM campaigns, marketing managers can’t determine what creative or channel is ultimately driving deposits.
And this means money is being burned.
Before I get into what’s actually happening with MMM, let me first explain what we mean by Media Mix Modelling, or Marketing Mix Modelling as it’s also known.
MMM – a quick definition:
MMM is a statistical analysis technique used to understand the impact of different marketing creatives, campaigns and channels on the overall business outcome.
It helps organisations to get a clear understanding of how each marketing channel, such as SEO, paid media, affiliate, TV, radio, etc, contributes to KPIs such as conversions, deposits or revenues.
By analysing current and historical data, MMM should be able to reveal the effectiveness of each channel and therefore guide future marketing spend.
It goes beyond basic tracking of individual channels to assess how different marketing efforts work together and how they are affected by external factors like major events or competitor campaigns.
It is complex and involves statistical techniques like regression analysis to model the relationship between marketing activity and business outcomes.
But when done right, it allows organisations and their marketing teams to optimise future marketing spend, allocate budgets more effectively and ultimately boost the ROI from all marketing activity.
So, what’s going wrong with MMM?
Recently, brands have gone all in on Media Mix Modelling, but with a top-down view, with fancy charts and quarterly reports that keep the CFO happy.
What they don’t have is last-touch attribution – in most cases, this is completely abandoned. But to crack the code with MMM, you need both. Period. Why? Let me use an analogy to explain.
MMM gives you the weather forecast, and last-touch tells you if it’s raining right now. Here’s a quick breakdown of what you should use each for.
MMM:
- Quarterly campaign planning
- Granular budget allocation
- Decisions over the mix of channels to be used
Last-touch attribution:
- Which campaign to kill today
- Which creative is actually working
- Where your actual players are coming from
For me, MMM is marketing level 400, but a lot of operators and marketers have yet to pass marketing level 101. Leapfrogging the basics is always bad business, and we are seeing that in hundreds of thousands of dollars being burned on adopting MMM without understanding it.
My advice. Use MMM to win the quarter and last-touch attribution to win the day.
But how do marketers apply last-touch attribution to their creatives and campaigns? Let me explain.
Last-touch attribution explained:
This is a marketing model that gives all credit for a conversion (this can be any conversion set by the marketer, such as registration, FTD, etc) to the final interaction the customer has with the brand before they take the desired action.
While MMM can be complex, last-touch attribution is a fairly simple method that focuses on the last touchpoint in the customer journey. Let me give a quick example.
Online sportsbook A is running a display ad for its latest sign-up promotion. If the player clicks on that ad and immediately signs up, the display ad would get 100% of the conversion credit, even if they had previously interacted with the brand on social media or subscribed to the newsletter.
Of course, this does ignore the influence of other touchpoints in terms of pushing the customer to the point of conversion. For example, they might not have clicked on the display ad in the first place if they were not already a newsletter subscriber.
Last-touch attribution can also lead to a bit of a skewed view, as it favours last-touch channels such as paid media while undervaluing other important marketing efforts. In short, it might not reflect the full customer journey or the full effectiveness of different marketing campaigns and channels.
But this is why it should be combined with Media Mix Modelling, as by doing so, marketers get a full picture of what is and is not working across their entire marketing activity.
Here’s a quick side by side comparison for what each should be used for.
Last touch focuses on individual touchpoints while MMM gives the big picture; last touch uses granular, user-level data while MMM uses aggregated data; last-touch is for immediate optimisation while MMM is for strategic planning; last touch gives insights into specific actions while MMM provides a strategic view.
Get your head around that, and you can use both to hit the target with your marketing activity at scale and well within the budgets set.
