Unified Marketing Measurement and Optimization, LIFT ROI may sound like the latest electric vehicle to drive around the block, but it’s actually a significant new development in analytics that can help auto manufacturers predict and improve how media and marketing spend will impact sales. The automotive sector has been – and is – going through an unprecedented period of change and challenge: the pandemic which on one hand, limited commuting but on the other hand, replaced air travel with family trips to personal vehicles, severe supply chain shortages, the push to more sustainable cars, oil prices, and looming recession. Dramatic structural dynamics like this demand a new approach to analytics that answer the age-old marketing question: “what works?”, but now needed more agile, granular manner to help marketers respond quickly to consumer mindset and behavior shifts.
Why is LIFT ROI a powerful fuel for the auto industry?
LIFT ROI provides unified media and marketing optimization, and it harnesses the innovations of AI and machine learning – along with all your media and marketing data, in a new revolution that goes way beyond traditional media mix modelling. The benefit is the ability to identify placement-level insights on what is working to enable continuous optimization decisions on your marketing tactics.
I know you’re thinking we have been down this road before: pricey hierarchical data models that are updated once a year and grow stale before you can actually use them. Kantar’s LIFT ROI capability, offers significant advantages. Auto manufacturers need insights on a daily rather than yearly basis to compete in this changing-by-the-minute market.
Here’s why you need to pay attention to LIFT ROI:
With privacy dictates and Walled Gardens, MTA is just not working!
GDPR and CCPA, the European and Californian privacy laws, forever changed the data we have access to, moving away from user-level cookie-based approaches. MTA is built on having access to an endless amount of user-level data.
Google, Amazon, and Facebook are the biggest holders of data globally and they are just not sharing. They have presented ideas for so-called “clean rooms” where data is stripped of personally identifiable information and shared in some sort of hermetically sealed environment. While there is skepticism about the practically and reality of this approach, there is acknowledgement that at least 30% of an advertiser’s data is just not available anymore. That brings up the question of how can we build models that still provides granularity of measurement to feed tactical optimization, like MTA models did? Perhaps, go a step further and tackle some of the disadvantages of MTA in not accounting for offline media or marketing influences.
Bayesian is the way to go!
You have likely heard the term “Bayesian probability” and have not given it much thought. Now, with AI-based models this approach, which assigns probabilities to marketing mix levers, will give you more reliable and complex insights. Yes, Bayesian was a theory developed in 1896, but Kantar’s marketing attribution model applies it to your data. It integrates domain knowledge – the beliefs that you already have about how your marketing works and new information to form a more accurate attribution and prediction of business outcomes. This contrasts with old models which focused purely on a stated hierarchy of events: if population A was exposed to Y at X time period, Z will happen. That just isn’t useful given the extraordinary dynamics and complexity of macroeconomics and human behavior.
Hierarchical models, are still used with LIFT ROI to determine which parameters, such as specific marketing and media channel investments cause a “trickle-down” effect to other touchpoints and how they might be interdependent. With LIFT ROI we combine both approaches and do it with great granularity based on insertion level data, i.e., the data we gather from a specific media placement with any given media publisher. We also have developed synergy models that allow us to extract insights about what indirectly drives customer behavior and conversion from one touchpoint to another. Additionally, we tease out the effects of uncontrollable factors such as the pandemic’s corollary effects on inventory levels, and how that impacts marketing effects.
Modelling needs to incorporate Automotive marketing’s complex layers to address the long shopper journey.
Given the long purchase cycle, a critical challenge for automotive brands is to ensure that marketing addresses consumers long before they enter the shopping cycle to make sure that they are intrigued and knowledgeable enough about the brand to put it into the consideration set. This means there is considerable focus on demand creation and a key question is how to allocate budget to brand building initiatives versus demand capture initiatives, and to specific touchpoints and experiences within those two buckets. Kantar’s modelling framework behind LIFT ROI is designed to provide answers here. Even when it comes to measuring demand capture levers.
Here’s what we did with a major automotive manufacturer.
A global automotive manufacturer wanted to understand their marketing effectiveness over time in a specific market and be able to predict the total number of cars sold. They needed to better forecast production needs, look at how their media spend was performing and understand how they could better optimize marketing spending.
They had traditionally found it difficult to link marketing activities to the number of sold cars. They had a marketing measurement model through their media agency, but it wasn’t precise enough and was a so-called “black box” analytics mystery. The data pipeline wasn’t fast enough, and the data needed more cleaning before it was usable. Campaign performance was measured on website traffic rather than actual sales.
What We Accomplished
Using Hamilton AI, which learns from each model update, the manufacturer was able to refit marketing on an ongoing basis. They were able to see what was happening last year, why it was happening and predict what will happen in the future- Kantar:
- Established the link between marketing activities and the number of sold cars.
- Defined the variables that affected sales and determined to what extent.
- Increased business transparency, implementing better insights into dynamics at work
- Offered a better sense of the validity of forecasts and established more security in the models succeeding.
- Made it easier to test hypotheses
- Presented the client with a dynamic platform that is easy to use and can be used to quickly survey media efficiency and predict how new marketing campaigns will fare.
- Increased the granularity of the output, making it possible to look at several different variables. The assessment included non-media variables and external factors such as holidays, Covid-19, macroeconomics and competitor incentives.
- Developed an understanding of the synergy effects of various marketing and media, creating a hierarchy. The client now understands which variables affect each other and to what extent.
- Measured media performance by each auto model.
- Meshed the various national branding and regional tactical media campaigns to understand the halo effects of the brand campaign on the actual number of cars sold in each region.
Takeaways
Kantar understands how challenging but also fundamental advanced analytics and optimization can be. With the expertise of Kantar we help marketers globally get to the level of granularity they need in real time in an ever-changing media and economic landscape. Fast. Accurate. Science based. LIFT ROI is a tool that will revolutionize your marketing.