Compared: Last-click attribution and Media Mix Modeling (MMM)
May 30, 2023 | Paul Arpikari, Konsta Mustasilta, Carmen Bozga
In the realm of marketing, measurement methodologies have evolved considerably. In this digital age, two prominent approaches stand out - last click attribution and Media Mix Modeling (MMM).
Last-click attribution, commonly associated with Google Analytics, is popular due to its accessibility and the allure of free basic features. However, its focus on the final customer touchpoint may overlook other significant interactions in the buyer's journey. While convenient, this approach may provide a somewhat limited view of marketing effectiveness.
Contrastingly, MMM offers a comprehensive perspective, accounting for all touchpoints across the customer journey, whether the touch point is clicked or not. Despite traditionally being seen as a complex and costly method for marketing measurement its holistic analysis and capability to manage large-scale campaigns make it a compelling alternative to last-click attribution. By offering a more rounded view of marketing impact, MMM allows for more informed decision-making and effective resource allocation in the ever-evolving field of marketing.
Last-click vs MMM through a practical example
Imagine a scenario where an individual sees a sneaker ad on television. The next day, the same ad surfaces on their social media feed, but they choose not to engage with it. Intrigued by the stylish sneakers, they search for the brand on Google and click on the branded search ad to visit the brand's website, yet refrain from making a purchase. After a couple of weeks, they receive a newsletter from the brand, engage with it, and are enticed to buy the sneakers due to a substantial discount being offered.
How will different last-click attribution reporting and MMM reporting look like?
What last-click attribution will report: The email campaign would receive full credit, 100 %, for the sale since it was the last touchpoint before the purchase.
What MMM will report: After the model has been trained with historical data, it would understand that there are multiple drivers for that sale so it would attribute that sales for example 32 % to TV, 11 % to Social media, 6 % to Google, 15 % to Email campaign and 36 % to promotion.
Comparison of Last click attribution and MMM
Comparing the two marketing measuring methods can be a difficult task considering their inherent process differences. These models provide different perspectives and can be utilized based on the specific requirements of a business. However, how do they stack up against each other? Let’s dive in.
By comparing these two models, it becomes clear that each has its strengths and weaknesses, and the choice between the two depends largely on a business's specific needs and resources. Last Click Attribution can be a great choice for businesses seeking a simple, easy-to-implement model that provides immediate insights, while MMM could be a better fit for businesses looking for a more comprehensive analysis of their marketing effectiveness, and are willing to invest more time and resources into the process.
Nevertheless, both come with a few shortcomings.
Problems of last-click attribution
Despite its popularity and widespread adoption, last-click attribution is not without its drawbacks. The first, and perhaps the most significant, is its narrow focus. Last-click attribution, as the name suggests, solely takes into account the final click or interaction that leads to a sale or conversion. On the surface, this might seem an effective, straightforward way of attributing sales to specific marketing efforts. However, it fundamentally overlooks the complexity of modern customer journeys.
In today's multi-channel, multi-device digital landscape, customers rarely convert after a single interaction. Instead, they engage with brands across numerous touchpoints or see multiple different ads from various channels before finally deciding to make a purchase. Last-click attribution, unfortunately, fails to consider these interactions, leading to a skewed understanding of what's truly driving conversions.
Secondly, this narrow focus results in a significant disregard for the holistic marketing funnel. By attributing the entire value of a conversion to the final click, last-click attribution negates the potential influence of other marketing touchpoints a customer may have interacted with throughout their journey. This includes everything from initial brand awareness campaigns to mid-funnel retargeting efforts, all of which play a crucial role in nudging the customer towards conversion.
No incrementally report
Finally, another inherent problem of last-click attribution is its methodology's focus on reporting traffic rather than incrementality. This leads to a skewed understanding of marketing performance, often resulting in misguided strategies and misallocated resources. For instance, it tends to attribute a disproportionately large share of sales to branded search, while underestimating the value of other channels like online video. The lack of incrementality understanding can result in an overinvestment in channels that appear successful according to last-click attribution, while potentially undervaluing other impactful channels.
Advantages of Last-click attribution
One of the most significant advantages of last-click attribution is its simplicity. The concept is straightforward: the last click that leads to conversion gets all the credit. There is no complex math or detailed modeling involved. This simplicity makes it an ideal choice for small businesses or those just starting with digital marketing analytics. It’s easy to understand, implement, and interpret, which is particularly beneficial for those with limited resources or expertise in the field.
Last-click attribution is incredibly action-oriented. It clearly identifies the final touchpoint in the customer journey, which is often the most persuasive or influential in driving conversion. This clear line of sight allows marketers to directly see which channels or campaigns are driving conversions, facilitating quick, data-driven decisions.
In the digital age, speed is everything. Marketers need to know what's working (and what isn't) as quickly as possible. Last-click attribution shines in this respect. Because it focuses on the final interaction before conversion, it can provide real-time insights into which touchpoints are most effective. This speed allows for rapid adjustments and real-time optimization of marketing strategies.
Last-click attribution is a cost-effective solution for businesses of all sizes. Tools like Google Analytics offer free last-click attribution reporting, making it accessible even to those with tight budgets. The affordability coupled with its ease of use makes it an attractive choice for many marketers, especially those in startups or small to medium-sized businesses.
Problems of Traditional Media Mix Modeling projects
For a long time, MMM has been considered a luxury product due to its price tag and implementation procedure. While MMM's rich analysis and the comprehensive view it offers are certainly commendable, it isn't without its fair share of challenges. The first obstacle that often springs to mind is its slow nature. Traditional MMM isn't built for speed, and that's an issue in the high-speed world of digital marketing. It's akin to planning a road trip with a map that only updates every few weeks or months – you might find yourself facing unexpected roadblocks or missing out on newly opened routes simply because your map isn't up-to-date.
Another significant issue is the expertise required to effectively use MMM. This isn't a tool you can simply plug data into and expect results. Implementing and interpreting MMM requires a deep understanding of statistical modeling and econometrics. This complexity can be intimidating for many businesses. Imagine being asked to pilot a plane without any prior training – it's a daunting prospect and one that's ripe for potential errors.
Let's not forget the cost factor either. The implementation of a robust MMM solution is often associated with a considerable financial investment. This doesn't just encompass the cost of the software or service fees, but also the expense of recruiting and training skilled analysts. This high cost barrier further cements the perception of MMM as a 'luxury' product, seemingly out of reach for businesses with smaller budgets.
Although traditional MMM presents these challenges, it's important to note that the landscape of marketing measurement is rapidly changing. Thanks to technological advancements and software-as-a-service (SaaS) solutions, modern MMM has become almost immediate, keeping pace with the lightning-fast demands of digital marketing.
These SaaS MMM solutions bring in a significant shift in how MMM operates. Remember the traditional map we were talking about earlier, the one that updates every few weeks or months? Well, it's like we've swapped that out for a GPS that updates in real-time. Now, businesses have access to dynamic, frequently updated data that accurately reflects the rapidly changing marketing landscape. No more unexpected roadblocks or missed opportunities.
Advantages of Media Mix Modeling
Perhaps the greatest strength of MMM lies in its ability to provide a holistic view of marketing performance. Rather than focusing on individual touchpoints, MMM considers the entire marketing mix – paid advertising, social media, content marketing, SEO, and more. It recognizes that consumers interact with brands across a multitude of touchpoints, and that these interactions collectively influence their decision to purchase.
Improved ROI and optimized budget allocation
MMM doesn’t just provide data; it offers strategic insights. By identifying trends and patterns in historical data, MMM can help marketers predict the potential impact of different marketing strategies. These forecasts can inform budget allocation decisions, helping marketers optimize their marketing mix for maximum return on investment.
Unlike models like last-click attribution, which attribute all credit to the final touchpoint, MMM aims to distribute credit fairly among all marketing channels. It acknowledges that a consumer's journey is a culmination of various touchpoints, and each touchpoint plays a role in guiding the consumer towards a conversion. By attributing credit to each channel based on its influence, MMM prevents the overvaluation or undervaluation of any single channel.
MMM is highly scalable, making it well-suited for large-scale, multi-channel campaigns. It can handle vast volumes of data from multiple sources, providing comprehensive insights irrespective of the size or complexity of the marketing landscape.
To sum it up, really the best way to describe the differences between these two methodologies is to read the picture above carefully. It condenses quite well the article and most likely all eCommerce and omnichannel retailers can relate to the following conclusion: if there are 0 % of sales attributed to discounts and you are claiming that almost two-thirds of the sales come because of the media channels thenit’s really hard to trust on those numbers.
Curious to learn more? Book a demo.
Related articlesRead more posts
No items found!