Here we go ag – AI – n...

As 2024 kicks off, the trade press and LinkedIn have been flooded with the usual swathe of predictions for the year ahead. Predictably, artificial intelligence (AI) crops up a lot, as it has done for many years now – even more so since the launch of ChatGPT. But this year’s musings about the subject are different to what’s been touted around before, they’re more relevant, more accessible, and importantly, more realistic.

Now, I am by no means an expert in this field, so ahead of writing this I’ve been doing some research and talking to experts to get a better understanding of AI and its potential applications in marketing. I’m still not an expert, and I never will be, but I feel it’s important that business leaders from all industries, as well as AI gurus feed into the conversation. So here we go...

The most useful definition of AI that I have found is “intelligent machines that mimic human-like cognitive abilities”. Within this, there are, apparently, many sub-categories, including but not limited to, algorithms, machine learning (ML), natural language processing, generative AI, problem-solving, pattern recognition, planning, and decision-making.

From this list, ML is perhaps the most interesting to marketers as it enables them to gain insights from data that is now generated at a volume and complexity that humans simply cannot manage. Until now, ML has largely been used as a stand-alone solution to create predictive models, used to observe, and classify patterns in data or other types content. Adding generative AI capabilities to ML is new and implications that we are only beginning to comprehend.

Generative AI is an advanced branch of AI that uses ML techniques to generate new, original content - such as images, text, audio, and video - based on the patterns and information present in the vast amounts of training data available for input in specific algorithms.

This is new, there are many, many challenges to be overcome before the use of generative AI becomes mainstream but it’s exciting to think about. Here we have something that not only enables the flexible processing of vast amounts of information but something that can create customised outputs at a scale that human beings simply can’t manage on their own.

Somewhat ironically, the successful integration of AI will rely heavily on human input over the next 12 months. Firstly, there is a huge need for education and training – marketers will need to learn how to train AI models appropriately – this needs to come from experts. Then a robust test and learn approach is needed to identify whether AI moves the needle in terms of performance, employee satisfaction, and consumer perceptions. Finally, companies that leverage AI to any extent will need robust policies to safeguard data and its use and assess the quality of outputs. This is vital to mitigate potential risks of misinformation or bias and preserve the integrity of both the technology and the brand it serves.

I asked ChatGPT “How will AI transform DOOH advertising” - here’s the output along with my emoji-based response. Overall, very impressive!

?

·?????? Audience Targeting:

o?? Data Analysis: AI can analyse vast amounts of data to understand audience behaviour, demographics, and preferences. This enables advertisers to target specific audience segments more effectively, delivering personalized content.

o?? Real-time Analytics: AI can provide real-time insights into audience engagement, allowing advertisers to adjust their campaigns on the fly based on performance metrics.

My reaction: ??

?

·?????? Dynamic Content Optimization:

o?? Personalization: AI can tailor content dynamically based on various factors such as time of day, weather conditions, and audience demographics. This ensures that the content remains relevant and engaging.

o?? A/B Testing: AI algorithms can conduct A/B testing on different creatives to identify the most effective content and optimize campaigns accordingly.

My reaction: ??

?

·?????? Contextual Advertising:

o?? Context Awareness: AI can analyse contextual information such as location, weather, and events to deliver more contextually relevant advertisements. For example, advertising a cold beverage on a hot day or promoting rain gear during rainy weather.

o?? Location-based Targeting: AI can use geospatial data to target specific locations with tailored content, increasing the relevance of ads to the local audience.

My reaction: ??

?

·?????? Facial Recognition and Emotion Analysis:

o?? Audience Reaction Monitoring: AI-powered cameras can analyse facial expressions to gauge audience reactions. Advertisers can use this information to understand emotional responses to content and adjust campaigns accordingly.

o?? Demographic Profiling: Facial recognition technology can estimate the age, gender, and other demographic information of viewers, helping advertisers refine their targeting strategies.

My reaction: ??

?

·?????? Predictive Analytics:

o?? Behaviour Prediction: AI can analyse historical data to predict future audience behaviour, helping advertisers anticipate trends and tailor campaigns proactively.

o?? Optimised Scheduling: AI algorithms can optimize the timing and frequency of advertisements based on historical performance data, ensuring ads are displayed at times when they are most likely to be effective.

My reaction: ??

?

·?????? Interactivity:

o?? Gesture Recognition: AI can enable interactive DOOH displays with gesture recognition technology. This allows users to engage with content in a more immersive way.

o?? Voice Recognition: Voice-activated features can enhance interactivity, enabling users to interact with advertisements using voice commands.

My reaction: ??

?

·?????? Cost Efficiency and ROI Improvement:

o?? Automated Campaign Management: AI can automate various aspects of campaign management, reducing manual labour and improving efficiency.

o?? Performance Prediction: AI can predict the potential success of campaigns, helping advertisers allocate budgets more effectively and maximize return on investment (ROI).

My reaction: ??

?

For anyone interested in AI in marketing and advertising, the key thing to understand is that what we have available to us today is not the finished product. It’s powerful, it’s transformative, it removes the need for a lot of repetitive manual tasks and can give us insights into data that are unachievable through any other means. But it is not perfect, there are flaws that should be respected not feared. I believe that if we give AI the respect it deserves and embrace it with skill and knowledge then we will see real transformation in this area.