AI for Post-Agency Marketing: Roses and Thorns

It’s quite hard to write about Artificial Intelligence (AI) and still be interesting, especially in long form like this.

We aren’t short of content, and one trips over AI news and views all day long, and we all have different levels of understanding, or even interest.

However, I believe that before we continue with our post-agency marketing roadmap that we need to build a common understanding between us of two things: the potential of AI, and the responsibility that comes with using it. In this newsletter we will focus on the first of these, and next week we will look more at the ethical and legal challenges of using AI in marketing (including some recent examples like Air Canada, who lost a court case this week after its chatbot hallucinated a fake policy about bereavement fares and had to repay a customer)

If there was ever a year to have more time on your hands as a curious marketer it was 2023, so I was lucky.

I have enjoyed learning about AI and experimenting with different tools and platforms. I’ve watched OpenAI wipe out so many start-ups with each successive update or launch. And we aren’t even at the end of the beginning of this revolution.

According to a recent report by the Association of National Advertisers (ANA) the main reasons for in-housing are cost efficiency, better knowledge of the brand, speed and nimbleness, and data control and protection. Obvs, as my daughter would say. These are also what AI accelerates, which is why we explore this topic at this stage. marketers are wondering if they still need to rely on external agencies for their marketing needs.

AI is accelerating agency in-housing, but it is not replacing agency outsourcing, as long as those agencies harness it in the right way. It is creating a new paradigm, where brands and agencies can work together in a more efficient, effective, and mutually beneficial way, where agencies focus on higher-value and more complex services, such as consulting, research, branding, and storytelling, that require human judgment, empathy, and vision. Or they won’t. And the march to disintermediation will continue, now even faster.

So let’s dive in, shall we? To where algorithms don't just complement but catalyse our creative endeavours. Where the fusion of AI with the unique identifiers of consumers opens up a realm of possibilities, promising a level of personalisation in marketing that was once the stuff of dreams. But, as Uncle Ben counselled Peter Parker, with great power comes great responsibility. How do we navigate this brave new world with wisdom, ensuring that our embrace of AI fosters genuine connections rather than hollow interactions?

Let’s go.

First, AI is not a new new thing

The first applications of AI in marketing was through its ability to analyse vast amounts of data at unprecedented speeds, uncovering insights that were previously inaccessible or too time-consuming to derive. This data-driven intelligence enabled marketers to understand consumer behaviour, preferences, and trends with a level of precision and depth that was unimaginable until a few years ago.

Now, from predictive analytics to natural language processing and machine learning, AI technologies empower marketing teams to create more engaging, relevant, and personalised campaigns. They allow brands to anticipate customer needs, tailor messages with surgical precision, and engage with audiences across multiple channels seamlessly. media planners, once reliant on spreadsheets and manual calculations, now leverage AI tools to predict the most effective media mix, optimise budgets in real-time, and measure campaign performance with granular precision. AI can automate routine tasks, such as campaign reporting and A/B testing, freeing up marketing teams to focus on strategic and creative initiatives. This agility is crucial in today's fast-paced digital landscape, where consumer preferences and behaviours can change rapidly. The integration of AI into voice search optimisation and chatbot technologies also exemplifies its innovative impact on marketing. As voice searches become more prevalent, AI's ability to understand and process natural language queries becomes critical in optimising content for voice search, ensuring brands remain visible in this rapidly growing search domain. Simultaneously, AI-enhanced chatbots are becoming increasingly sophisticated, capable of handling complex customer service inquiries and providing personalised shopping advice, thereby enhancing customer engagement and satisfaction.

Through this adoption, AI is fundamentally altering traditional roles and functions within marketing teams and agencies. This disruption stems from AI's ability to perform tasks that were once the exclusive domain of humans with unparalleled speed and efficiency. As a result, the roles within marketing departments and agencies are evolving, shifting towards more strategic, analytical, and technologically savvy positions. We will be exploring these talent and capability challenges together in a future edition.

(OK, breathe, then...)

Now, to demystify Generative AI a bit, since the hype around it the past year has created as much disappointment as hope.

Generative AI stands out for its ability to create content—from written articles to images and videos—that is increasingly indistinguishable from human-generated content (a new bar having been set by OpenAI’s launch of Sora this week). This technology enables marketers to produce high-quality, customised content at scale, significantly enhancing brand engagement and opening new avenues for creative storytelling.

It’s potential is unleashed when coupled with AI's application in predictive modelling and hyper-personalisation. By analysing consumer data, AI can forecast future buying behaviours and preferences with remarkable accuracy. This foresight allows marketers to tailor their strategies to meet the future needs of their target audience, creating a proactive rather than reactive marketing approach.

Here's a simplified view into how this approach works:

1. Collection of Data: The first step involves gathering data associated with unique identifiers for each consumer. These identifiers could be anything from customer IDs in a CRM system to cookies tracking online behaviour. This data is not just limited to basic demographic information but extends to interaction histories, purchase records, and even nuanced preferences indicated by online activities.

2. Analysis and Insight Generation: Generative AI algorithms then analyse this data to extract meaningful insights about individual consumer preferences and behaviours. By processing large datasets, the AI can identify patterns and preferences specific to each consumer, such as favoured product types, content engagement patterns, and optimal engagement times.

3. Content Generation: Leveraging these insights, generative AI tools then create personalised content for each consumer. This could range from personalised email messages, tailored product recommendations, to customised content feeds on a platform. The AI can adjust the tone, style, and format of the content based on what it has learned about the consumer's preferences, ensuring that each piece of content feels uniquely relevant.

4. Dynamic Adjustment: Perhaps one of the most powerful aspects of using generative AI in this way is its ability to dynamically adjust content based on ongoing consumer interaction. As the AI receives feedback—through metrics like engagement rates, conversion rates, and direct consumer feedback—it refines its understanding of consumer preferences and adjusts future content accordingly.

5. Scalability: This approach allows for the personalisation of content at scale, something that would be logistically impossible for human teams to achieve manually. Generative AI can create thousands of personalised content pieces in the time it would take a human to craft a single one, enabling brands to maintain a high level of personalisation across their entire consumer base.

There are many tools where you can have a play yourselves. Here are two examples to check out:

ClickUp serves as an all-in-one project management tool with an AI content assistant feature, making it ideal for marketing project management and AI-assisted content writing. It helps with content research, ideation, copy editing, and summarisation, among other tasks, making it a comprehensive tool for managing the entire content lifecycle within one platform.

Narrato focuses on end-to-end AI content creation, planning, and collaboration, along with workflow management. It's specifically designed for content teams, providing AI assistance for ideation, content creation across multiple use cases, and content optimisation. Narrato also features a powerful SEO content brief generator, AI images, and workflow automation tools, making it a versatile option for teams looking to streamline their content processes.

When considering these tools, you should weigh factors such as the specific content creation needs of your team, the level of AI integration required, budget constraints, and the desired balance between content creation and overall project management capabilities.

When you have a play, you will get excited, and start sparking as to what can be made possible for your brand. But, it's not all roses. There’s also thorns. Here's a quick consideration of the pros and cons of the types of GAI tools you might explore:

1. AI Content Generators (e.g., text, images, videos):

• Pros: Automate content creation, saving time and resources; generate diverse content types; enhance creativity with AI-driven ideas.
• Cons: Content may require human review for brand alignment and quality; potential ethical and copyright considerations.

2. AI-driven SEO and Content Optimisation Tools:

• Pros: Improve search engine rankings with AI-optimised content; identify high-value keywords; automate content analysis for SEO best practices.
• Cons: Over-reliance on AI recommendations may overlook nuanced human creativity and intuition; potential for generic optimisation strategies.

3. AI Customer Data Analysis and Segmentation Tools:

• Pros: Offer deep insights into customer behaviour; enable hyper-personalised marketing campaigns; automate customer segmentation.
• Cons: Dependence on data quality; privacy and ethical considerations in data handling; complexity in integration with existing systems.

4. AI Chatbots and Customer Service Tools:

• Pros: Provide 24/7 customer service; automate routine inquiries, freeing human agents for complex issues; enhance customer engagement.
• Cons: May struggle with complex customer queries; risk of impersonal or unsatisfactory customer interactions if not well-designed.

5. AI Marketing Automation and Personalisation Platforms:

• ?Pros: Automate marketing workflows; deliver personalised customer experiences at scale; optimise marketing campaigns in real-time.
• Cons: Requires significant setup and fine-tuning; potential for data silos; balancing personalisation with customer privacy concerns.

Sharpen the Saw

For marketing leaders looking to invest in AI training and literacy for their teams, there are several reputable providers and courses designed to enhance understanding and application of AI in marketing. Two notable institutions offering specialized training in this area are the Marketing AI Institute and eCornell.

Marketing AI Institute offers a comprehensive suite of AI courses tailored for marketers, including their Piloting AI and Scaling AI courses. These are designed to help both beginners and seasoned professionals understand and apply AI in marketing strategies. The Institute's AI Mastery Membership Program also provides access to exclusive education and training resources, including live briefings on AI trends and quarterly sessions on generative AI technologies. This approach is aimed at rapidly transforming businesses and careers by building AI literacy across all organizational levels.

eCornell, Cornell University's online learning platform, offers a Marketing AI Certificate Program authored by Clarence Lee, a former assistant professor at the Johnson Graduate School of Management. The program focuses on designing performance marketing strategies using AI, prioritizing resources across media channels, and identifying opportunities to apply AI in marketing processes. This facilitator-led, 100% online program spans two months and is designed for a variety of marketing professionals, including marketing managers, content marketers, and CMOs, aiming to equip them with the skills to apply AI in enhancing customer journeys and achieving data-driven decision-making. Stanford has a great program too, as do others.

Phew. You made it. This was quite an dense one, necessarily, otherwise it’s just another breathless post about some new AI vapourware.

In embracing AI, marketers open the door to a world of possibilities where data-driven insights, operational efficiency, and personalised consumer engagement converge to create marketing strategies that are not only effective but also resonate on a human level. The potential for AI to transform marketing is immense, but realising this potential requires a thoughtful approach that balances innovation with responsibility.

And it’s that responsibility that we will turn our attention to in next week’s Edition.

Until then, I look forward to your thoughts, challenges and builds. Cheers, Justin