AI is here to stay, but that does not signal doom and gloom.
The role of AI-generated (synthetic) data versus human input has recently come to light due to its application on brand tracking and market research. Synthetic data mimics real data, using machine learning on a database of human-like responses.
This made me think of the quote by Ralph Waldo Emerson:
“The man who knows HOW will always have a job. The man who knows WHY will always be his boss.”
Humans are inherently curious creatures. As we age, we are forced to conform to societal standards that accelerate our lives – often forgetting to stop and understand the ‘why’ behind what we do. It is in this acceleration, that technology has afforded us new ways to be more efficient and productive with our time.
So, why is AI any different?
If AI is trained effectively, with the right inputs and parameters, it will work favourably for the user. As a strategist I’m all for making life easier by finding ways to fast-track laborious tasks like insight gathering and extensive market research, normally taking up weeks of resource, time, and money.
After all, time is money.
There is no silver bullet for accuracy.
There have been recent developments made in the space of brand tracking using GenAI to provide speedy insights for marketers. One recent study has looked into the application of LLMs on market research and it has revealed some extraordinary results; proving synthetic data to be around 90% accurate to human collected sources.
To achieve 100% accuracy in research is virtually impossible. The sample size will always be a representation and the model will always attach a bias. The good news, there is a way to bridge the ‘accuracy gap’ and synthetic data may be the solution. Gartner even estimates that 60% of the data used in AI and analytics projects will be synthetically generated by 2024.
This is music to my ears. If we can get hold of this data in half the amount of time and a third of the cost, why would clients not consider this? By doing so, it provides the freedom for marketers to focus on the ‘why’ by being more innovative, effective, and quicker at tackling broader business challenges.
Perfection Paralysis.
We are paralysed by perfection. This often results in cumbersome, outdated models that hinder progress. Having recently moved from a Holding Group to an Indie agency made me re-evaluate new ways to problem solve. Indie agencies act like start-ups and embrace prototype thinking. We prioritise innovation and iteration, offering faster solutions that are aligned with client needs. Our clients also emulate this behaviour, seeking out ways to challenge the marketing “playbook” through cost-effective solutions. Embracing AI offers the ability to identify faster insights and foster innovation, despite its imperfections.
If you are still not convinced, AI has already made some fascinating inroads into the world of food and beverage. Mondelez successfully used AI to help predict popular taste profiles and found the predictions for a new product closely match those of their human “supertasters” dataset. This insight led them to not only change the formulation but assisted in outselling the old recipe and even surpassing their competition in market!
AI is a tool; it is not the solution.
AI is only one piece of the puzzle and to rely solely on this would be naïve. It requires “experienced eyes” to use its outputs in broader context, applying industry experience, socio-economic trends, and category headwinds to make an informed decision for your brand.
AI is not magic; it is maths. We still require a hypothesis to the challenge we seek an answer to. What used to take weeks to prove, now take seconds to validate. Yes, it can compute large amounts of data and detect patterns at speed, but it is the creative application of human analysis that shows our true value in the equation. Pair our human ability to think critically with AI and we build a winning combination – demonstrating how both can work in harmony.
Let us embrace AI for what it is. A tool to help compute facts quicker. We will continue to be AI’s boss if we do not expect answers to existential problems like “The Ultimate Question of Life, the Universe, and Everything” and focus on WHY the answer is 42.
Rachel Teh, National Head of Strategy, Speed