How to plan for successful AI implementations
When you decide to incorporate artificial intelligence (AI) tools and techniques into your marketing program, you can easily jump into the experimentation and “doing” mode. In fact, if you were to pick up (and read) a trade publication or hear any chief marketing officer (CMO) speak, they extol the virtues of getting your hands dirty and experimenting. But I view the experimental approach to AI as a mistake that can take up valuable time without yielding meaningful business results.
The speed with which customer preferences and overall market conditions change leaves little time for experimentation anymore. You must integrate AI into your marketing efforts with the goal of driving meaningful value in the short term for your business.
The following steps give you a logical and stable path that paves the way for smooth AI implementations in marketing:
-
Define your goals.
Without clear objectives, teams may be mesmerized by the technology and force fit it to uses that may not offer the best results for your business.
-
Choose the right AI tools and techniques and plan to measure your marketing outcomes against your goals.
For example, if a company’s goal is to increase customer engagement, they can track user engagement metrics (such as how much time a user spends on a website), click-through rates, or conversion rates to determine whether their AI-powered marketing campaigns are effective.
-
Invest in quality data to train your AI tools.
The data that your business gathers to train the AI engines that it uses for marketing can help the AI provide various insights and recommendations to your teams, such as where and how to advertise.
Develop an explicit data strategy that includes gathering data from various internal and external sources — first-, second-, and third-party data. Bring it all together into a single data lake (a single repository that stores, processes, and secures large amounts of organizational data). When you have this data in place, you can point AI engines to your data lake and let the AI model training begin.
-
Prioritize customer privacy and data protection.
You have the essential task of respecting customer privacy. Ensuring this respect means holding your marketing teams and yourself to a very high standard of transparency about your data usage.
You must clearly communicate with your customers about how you collect, use, and protect their data, including how you connect their information with other data sources. And follow data protection regulations such as the General Data Protection Regulation (GDPR) in Europe and the California Consumer Privacy Act (CCPA) in the U.S.
How to train your marketing team to use AI
You must train your marketing team regarding the capabilities and limitations of artificial intelligence (AI). Instilling this knowledge helps your teammates better integrate AI into existing marketing processes, understand how best to harness the opportunities around collecting and analyzing data, and devise innovative growth marketing solutions — all with AI acting as a co-pilot.
Taking a business-oriented approach to training not only equips your team with essential skills for using AI in their work, but it also encourages a culture of continuous education and innovation. These factors can, in turn, help your business stay agile and competitive in the fast-evolving digital marketing landscape. Plan to offer a multi-layered training program to your marketing team and consider the following important points:
- Begin with a foundational training program that covers the essentials of AI, including machine learning, data analytics, and AI ethics. For this training, you can use platforms such as Section, Coursera, or Udacity, which are designed for business professionals. This initial step establishes a basic understanding across your team.
- Tailor the foundational knowledge to training on specific roles. For example, marketing teams could explore AI in customer segmentation, while more sales-oriented teams might focus on AI-driven customer relationship management (CRM) tools. Salesforce’s specialized training for its Einstein AI platform can provide you with an excellent resource, demonstrating the practical application of AI in specific job functions.
- Add hands-on experience that builds on the foundational training to strengthen the knowledge level within a team. Through workshops and case studies, team members can apply AI concepts in real-life scenarios, a practice enhanced by partnerships with companies such as IBM, known for their practical AI workshops. This experience reinforces the more theoretical learning and also encourages more creative thinking in your team to solve specific business problems.
- Treat AI training as an ongoing journey, not a one-time event. Encourage continuous learning and staying updated with the latest AI advancements through resources such as the Google AI blog or the OpenAI blog.
Fostering a culture of innovation for your marketing team is a necessity in the AI era. Encouraging your team to construct targeted experiments by using AI tools and engage in cutting-edge projects (such as using the generative AI recommendations in Google’s Performance Max) promotes a strong willingness to tackle challenges. And as always, encourage your teams to align their AI projects and experimentation with overall business objectives.
How to compare AI tools for effectiveness
Testing various large language models (LLMs) to see how effective they are relative to one another can be a valuable practice for experimentation among marketing teams. For example, @theneweracore on TikTok showcased a comparison among Microsoft’s Copilot powered by OpenAI’s DALL-E, Midjourney, and CGDream. This test, conducted by marketer Sonya Naboka, aimed to determine which image generator could best create an alien wearing Lacoste.
Copilot emerged as the winner for its accurate recreation of the Lacoste crocodile. This example highlights how different AI models can produce different results, demonstrating the significance of AI’s ability to handle brand iconography — a crucial step for the technology to potentially take over advertising duties.
How to choose partners for marketing with AI
When you decide to bring your marketing efforts into the artificial intelligence (AI) era, you have no shortage of tools, techniques, and companies that can help you accomplish your business goals. In fact, you can refer to a website, There’s An AI For That, which tracks the number of companies that help large businesses with their AI needs.
Increasingly, both traditional marketing agencies and digital agencies offer a suite of services to help you integrate AI and, more specifically, generative AI into your organizations. If you work with agency partners, you may want to begin by asking them whether they have the capability to help you with your AI needs.
Look for AI assets that match your business needs
When you’re choosing the right partners and the appropriate tools for integrating AI into your marketing programs, you have key factors to keep in mind. Look for solutions that offer scalability (so that they can grow with your business), ease of use, and comprehensive support. Some options include
- IBM’s watsonx.ai: Stands out with its versatile AI services suitable for various business needs — from enhancing customer engagement to training, validating, fine-tuning, and deploying AI models
- Google Cloud AI: Offers best-in-class machine learning (ML) solutions that are scalable and also integrate smoothly with other Google services
- Salesforce Einstein: Helps you out with marketing-focused AI tools, including AI-powered customer relationship management (CRM) solutions that are equipped with predictive analytics and automated task management
- Smaller companies: Such as C3.ai, Dante and Instalily can help you create knowledge bases, autonomous agents, and chatbots relatively easily
Look for partners with an AI track record
Choose partners and tools that have a solid track record and deep industry knowledge of AI implementation. For example, you might look to companies such as
- Accenture: A leader in offering comprehensive AI consultancy and implementation services across a range of industries, ensuring that AI integration aligns with specific business objectives.
- Deloitte: Renowned for its strategic approach in AI and analytics. Deloitte can help your business integrate tailored AI solutions that align with your broader business goals and can lead to significant improvements in efficiency and decision-making processes.
Find AI partners with ethical company policies
Verify that the ethical standards and data privacy practices of the AI tools and partners that you choose agree with your standards. Ensure that they not only comply with data protection regulations — such as General Data Protection Regulation (GDPR) in Europe and the California Consumer Privacy Act (CCPA) in the U.S. — but also align with your commitment to ethical AI use and customer privacy. This scrutiny is critical not just for legal compliance, but also to maintain customer trust and protect your business’s reputation.
How to avoid over-reliance on AI marketing automation
The evolution of artificial intelligence (AI) may be in relatively early stages, but over-reliance on its use for automating marketing practices already presents distinct challenges for businesses. As a marketer, you must protect your customers from the overuse of automated systems and AI-generated content, which can sometimes misunderstand customer intent, ignore cultural nuances, and generalize more than it should.
Don’t degrade the customer experience
Overuse of AI automation most directly affects the customer experience. Automated systems, although efficient in handling routine tasks, often falter when dealing with the nuances and complexities of human interactions. This shortfall becomes glaringly apparent in customer service scenarios, where AI-driven responses can come across as robotic and unsatisfactory, particularly in complex or sensitive situations such as dealing with medical questions.
Customers expect personal treatment and empathy, elements that AI just can’t fully replicate. This can lead to a diminished customer experience and erode the human connection that’s fundamental in building brand loyalty and trust.
Don’t offer marketing content that bores customers
Over-relying on AI in marketing can potentially create content homogeneity (a too-regular condition that lacks specifics and appeal). AI excels in analyzing data and optimizing content for certain parameters, such as search engine optimization. However, using this focus only often leads to formulaic and uninspired content that lacks the creative spark or emotional richness required to resonate with audiences on a deeper and more human level.
The distinct voice and personality of a brand can get lost in translation when AI takes the helm unguided. You especially see this problem in content-heavy marketing strategies — for example, with brands like LendingTree — where the need to stand out from the competition is crucial and filled with lots of explanatory content about personal finance topics. In the absence of human creativity and intuition, even the most data-driven content can fail to engage and inspire the intended audience.
How to use text-to-video AI tools: A case study
Toys“R”Us, a child-centric retail company, showcased the potential of artificial intelligence (AI) in filmmaking at the 2024 Cannes Lions Festival with a short promo film created using OpenAI’s text-to-video tool, Sora. The film, depicting a young Charles Lazarus and the brand’s mascot Geoffrey the Giraffe, garnered mixed reactions online. Developed in partnership with creative agency Native Foreign, the project required significant human oversight and iterative refinement. However, some critics found the imagery unsettling and called the film an abomination.
Toys“R”Us highlighted this effort as an exploration into innovative storytelling methods and is considering future advertising uses for the promotional video. Sora, while not yet publicly available, is anticipated to disrupt the digital entertainment market upon release, although (arguably) it may only be able to produce relatively boring content.
Keep operations on target
I can’t overlook the operational risks associated with an over-dependence on automated systems. AI-driven automated systems, although robust, aren’t immune to technical failures or external disruptions. Reliance on these systems can
- Lead to fragility in marketing operations, where sudden technology malfunctions or changes in external algorithms can have disproportionately large effects. For example, every time Google updates its search algorithms, some brands see significant changes in their website page ranking positions in the Google search results.
- Unintentionally promote atrophy within marketing teams. Reliance on AI tools can diminish the development and application of essential marketing skills. This decay becomes particularly problematic in situations that require quick, creative problem-solving or sensitive handling (such as responding to complicated customer service complaints).
