When ChatGPT descended on us late last year, no one imagined its overnight success. Today, you can’t browse a news website or social media platform like LinkedIn without an overdose of opinions and cheat sheets on integrating Generative AI into your daily tasks – I almost believe AI could help me brush my teeth and tie my shoelaces, too!
And with this new evolution of AI, doomsday predictions for market research and “traditional tracking” have also been flooding my timeline. Having been around brand tracking for 25+ years, I think I can also say a thing or two about it.
Market research, and specifically brand tracking, has sped up and undergone significant changes to keep pace with the evolution of our core stakeholders - marketers! The way we track brands today is different from what was available five years ago. Not every piece of software coming out of the tech hot shops is fit for purpose. At Kantar, we test and adapt our offerings to match specific consumer needs and use cases. Technology allowed the scaled use of neuroscience-based or system 1 methods to understand a brand’s imprint within tracking. Let’s delve into the interplay between AI and traditional methods in brand tracking.
The power of lag-free, real-time data
We’ve adopted existing science to build our own proprietary AI tool that removes noise from survey data so clients don’t get jumpy or squiggly trendlines. Unlike traditional methods that accumulate data over several days, our tool does not waste samples or report old data. This is crucial because survey data is often plagued by sampling and non-sampling errors that generate noise. With Kantar, clients can stay ahead of the curve with daily trends and insights that are both accurate and real-time. Our AI tool has also been trained to provide short-term forecasts on KPIs, allowing clients to make informed decisions and outpace their competitors. Our latest solution, Kantar BrandDynamics, incorporates automation in tracking to provide even greater value to our clients. It's worth noting that our solution does not rely solely on predictive AI technology but rather on a variety of building blocks to ensure the best possible results for our clients.
Obsession with evidence-based brand insights
At Kantar, we have a longstanding tradition of analysing successful and valuable brands in order to uncover the secrets to their success. Just so we can unravel the magic formula that other brands can use to achieve the same level of success. Kantar BrandZ is now in its 18th edition and uses our Meaningful Different Salient (MDS) framework for brand equity to understand the drivers that enable some brands to consistently outperform in the long term. The strongest brands create value in more ways than one. They capture 9X more volume share, can often charge 2X more, and are 4X as likely to grow future value share.
Our market-leading MDS framework has been consistently validated with new data and is globally accredited to deliver commercial and business outcomes by the Marketing Accountability Standards Board. The MDS framework is at the heart of our brand thinking and the golden thread embedded across all solutions and tech-enabled brand products. With BrandDynamics, you get a balanced account of daily measures like brand and advertising salience, along with long-term measures of brand equity. Our clients deeply appreciate our expertise in all things brand, which comes from over 50 years of pioneering experience.
Gold standard data quality
I am reminded of a comment from a client many years ago. After a presentation that went particularly well, she said, “You do such good analysis but is the data quality reliable…?” In the context of AI, the data set the AI is trained on is often overlooked. You might already be noticing the many biases that AI-based tools are reinforcing, if not intensifying. The output of an artificial intelligence tool is only as useful and meaningful as the data set it has been trained on; it does not know the world outside of this knowledge pool. The same holds for AI applied to brand tracking.
We are very conscious of the GIGO (garbage-in-garbage-out) rule therefore getting the best quality survey data was of the utmost priority while setting up BrandDynamics. Our Profiles Audience Network uses market-leading fraud and bot detection AI to keep the panels clean and human, providing high-quality, daily sample. We do 4X more fraud prevention than anyone else. Our Qubed anti-fraud software uses state-of-the-art machine learning and 3 deep neural networks to process over 300 features for each survey session. This allows us to detect and eliminate within seconds the fraudulent activity that may go unnoticed by humans! The application of advanced intelligence to high-quality data ensures that our insights are reliable and can lead to robust decision-making.
On the surface, a traditional brand tracking programme might look like it always has: a duck peacefully floating, oblivious to the world around it. But beneath the surface, there is furious paddling as we add new skills and processes to deliver an enhanced yet seamless experience on brand tracking.
Ready to discover how to make informed marketing decisions with continuous brand tracking? Keep a watchful eye on your brand's performance compared to your competitors and track essential performance indicators for growth with BrandDynamics on Kantar Marketplace.
Book a demo today to learn more.