The Power of AI for Marketers

“You won’t be replaced by AI, but you may be replaced by someone who uses AI better than you“

There is a RACE to Artificial Intelligence (AI). We are ALL in it as consumers, professionals, and individuals. Some believe it's already 'mainstream' as it's been ubiquitous through articles, social posts, family, friends, and water cooler discussions.

Name any company who hasn’t announced AI being embedded into their products this year

You don’t have to look far to find it. Whether it’s on social media, at work, talking to your kids, or in the media, it’s there. Google just announced how AI will be incorporated into all their products (if not already) at their annual I/O conference last week. Much of the buzz has been building up over time, but ChatGPT opened up the floodgates last November. I can barely keep up with all the AI posts on LI. If you have played around with ChatGPT and Bard, you have already experienced the potential, the flaws, what works and what is still WIP. However, I would like to discuss the implications for marketers, how far we've come as an industry, how to maintain the edge but also how to get started for newbies. How can we continue the momentum based on some of these new advances? What will the impact be for marketers?

Let’s first make sure we are all on the same page - what is AI?

In simple terms,” AI is the ability of a machine to learn and perform tasks that would normally require human intelligence”. This can include things like recognizing objects, understanding natural language, and making predictions.

Artificial Intelligence (AI) has emerged in so many business discussions about it's disruption to every industry. Many people forget that AI has already been fueling a significant part of marketing - specifically digital marketing. For over 2 decades, it has opened up new frontiers for marketers by creating consumer connections, analyzing mountains of data, increasing the speed to execute, enabling measured impact, and make informed decisions. This article covers the evolution of Marketing AI, its significance in marketing, pros/cons, benefits, who’s doing it well, and practical steps to start your Marketing AI journey.

Fortunately, for now, AI will NOT take over human jobs but we are at an interesting inflection point where the newest capabilities are smarter, faster, more powerful, and of course more ‘human’ than ever before

AI = the confluence of Data, Digital and Technology

Foundational elements of AI require a ‘machine’ or algorithm to take in data to ‘learn’ from it and continue this process as more data is fed into the system.

In the marketing world, we’ve gone through many phases of "new shiny objects". Many ‘buzzwords’ that have been out there for years were precursors or launching pads for AI.

These topics are listed in order that they were introduced:

• Neural Networks - 40's - breakthroughs in 80's and early 2000's. A method in artificial intelligence that teaches computers to process data in a way that is inspired by the human brain. It is a type of machine learning process, called deep learning, that uses interconnected nodes or neurons in a layered structure that resembles the human brain.
• Database Marketing - launched in 80's. Database marketing is a form of direct marketing using databases of customers or potential customers to generate personalized communications to promote a product or service for marketing purposes. This is an old term from traditional CRM but reinvented
• Big Data - peak in 2014. Big data primarily refers to data sets that are too large or complex to be dealt with by traditional data-processing application software
• Data Science 2000's- Data science is the study of data to extract meaningful insights for business
• Digital Transformation slow rise from 2017 with 2021 peak during pandemic needs- the adoption of digital technology by an organization to digitize non-digital products, services, or operations

Again using Search data as a big focus group to understand interest, see the related searches for these terms from 2004 - present

Digital Transformation and Customer experience peaked last year and has been slowly declining/flattening. Big Data also showing steady decline with some spikes here and there. Database marketing was too small of an audience to even show up compared to the others. Data Science however continues to grow steadily. Finally Neural Networks were quite popular in the early 2000's but have slowly declined in searches.

Replace Database Marketing with AI and it dwarfs all of them.

The resurgence of AI can be seen through search trends below.

As you can see, AI has taken over the dialog and replaced many of the other big marketing topics. It's even superceded Taylor Swift searches before and while she's been on tour in the last few weeks!

Who's doing it well

Many industries have been adopting AI for years: manufacturing, finance, technology, auto, retail, e-commerce, and advertising. But there are specific categories that are more ‘marketing’ driven 

Chat GPT tells us that the following industries have already been deep into AI. No surprise here. However, I highlighted in blue the areas that are marketing/consumer-driven opportunities

While these were new at the time, they are all iterations of data driven opportunities. 

As you can see, there is no shortage of AI uses for Marketers. The retailers, ecommerce and advertising companies have led the way in the last 2 decades. Not surprising tech leaders include Google, Facebook, Tik Tok and Amazon use AI to power their advertising efforts through programmatic and targeted ads. However, other consumer brands are also pushing in these areas in the form of personalized content, recommendation engines, chat bots, curated playlists, user preferences, etc.

• Amazon: Uses AI to personalize product recommendations and optimize its supply chain.
• Netflix: Uses AI to recommend personalized content to users.
• Spotify: Uses AI to curate playlists based on user preferences.
• Sephora: Uses AI-powered chatbots to provide customer support and product recommendations.
• Domino's Pizza: Uses AI to optimize delivery routes and personalize marketing messages.
• Stitch Fix -  Uses machine learning algorithms to match the right products with the right clients. The algorithms determine characteristics of particular items - color, material, style, etc. - matter to each customer. Insights used for messaging, media and upsell efforts

Part of the evolution stems from key changes:

1. Technology advances 
2. Where consumers were spending their time  
3. The ability to ‘find’ consumers through connected digital devices

AI is dependent on a strong Data Strategy

AI builds on what we already have. Digital advances are now baked into all components of an interaction with a consumer. In other words, every touchpoint with a consumer is an opportunity to ‘collect, connect or activate’ the data. These are critical inputs for a sustainable AI system [see discussion from previous post - titled Return on the Consumer Data]. The Collect, Connect, Activate flywheel is critical for any data driven organization. The framework continues to be just as relevant in the new world of machine learning. AI can increase the velocity and impact of the Collect, Connect and Activate framework. What's different from the past is that AI augments what you already have. As the flywheel gets more data, the system gets smarter by using AI

A year ago, I was talking about the convergence of many media types based on industry changes such as 3P deprecation, stronger computing power, ability to collect and house more data, and consumer willingness to provide more information about themselves. At the time, I called it DataTech knowing only a select few technologies could really pull it off. Today, much of DataTech includes AI as a significant source to create a unified media system. Connecting the dots between these data sources (ie. the customer experience) is more achievable than ever before. The DataTech (supercharged by AI) will also help determine which media type is most appropriate, by device, channel, location, time, cadence, etc for each audience, lifecycle phase, profitability, etc. The new AI capabilities can take massive amounts of data, look for patterns (some that are too difficult for even the most talented data scientist to do on their own), and provide recommendations, predictions, and decisions. AI is now smart enough to determine the right combination of hundreds and thousands of marketing variables.

Requirements for AI 

Very simply, AI has 3 requirements:

1) Vessel - a computer or some device to ‘house’ information/data

2) Data - any data from somewhere or something - typically a lot of it but AI can work even with limited data

3) Algorithm - in the past by a person like a data scientist. Today it’s a machine that can ‘learn’ with more data

These 3 criteria are essential to enabling the power of AI. If you have these, you have the requirements to ‘collect, connect and activate' your data to drive business impact. You have the data (collect), you have the place to connect them (vessel/machine/cloud), and you can activate the data into insights (algorithm). If you don’t, then you need to build these 3 areas to move into AI for your business. If any of these areas don't naturally fit in your business then AI may not be for you.

Macro vs Micro AI

My research with AI shows that most Marketers are using it at 2 levels: macro level and micro level.

There are both Macro and Micro AI opportunities for Marketers. 

Macro AI exists at the broader levels of how it can help marketers. This might include programmatic marketing, personalization tools, predictive analytics, chatbots, recommendation engines for e-commerce, etc. Tools that drive smarter marketing execution, promotions, and personalization to your consumers.

AI has already been integrated into programmatic marketing, SEO, SEM, e-commerce, display, email, websites, and social. Some of the most sophisticated AI techniques have been used since paid advertising inception (think >20 years ago when Google, Facebook, and Amazon launched their advertising capabilities). Remember the more data you have, the more robust your AI/algorithms become. While AI has been infused into so many marketing digital tools now, it’s been taken to a level because of several things

1. Platforms can collect more comprehensive and granular data
2. Cloud computing can store much data
3. Smarter machines and algorithms exist to process the tremendous data that exists now

Micro AI includes day-to-day tasks that can be replaced using AI tools. While some tools already exist such as Grammarly, spell check, Google suggest, etc., generative AI opened up the floodgates for more opportunities. This might include content writing, visual development for a display ad, video creation based on prompts, writing emails, cleaning data, analyzing data, creating a chart, etc. New tools such as Jasper, ChatGPT, and Notion are gaining significant adoption. Over 2,500+ AI tools have been released in 2023. Zain Kahn (the AI Guy) is always sharing new tools all the time if you are interested in learning more - please follow him on LinkedIn. Here’s a great ‘sampling’ of tools to help many day to day marketing tasks.

So why AI now?

Many of my students have asked me about my 'Future of Digital Marketing' lecture because it's evolved so much from last year.

We’ve seen a lot of ‘big’ ideas come and go in Marketing. So how do you differentiate between hype vs reality? One of the biggest trends was the hyperbole from last year to this year. Most recently, there has been a pivot from crypto, metaverse, and NFT discussions to AI. By the way - these other topics are still quite valuable for marketers but it's all about timing and results. In this case, the economic uncertainty contributed to the shift but also the launch of ChatGPT last year. Given the current climate, there is a stronger focus on more tangible, direct, and immediate impact initiatives.

So how do you differentiate between the hype vs reality?

Hype - all talk, Reality - business impact supported by data

The data should support these areas and demonstrate a clear ROI:

• Automation
• Productivity/Efficiency
• New products
• Personalization
• Solve complex problems
• Manage a TON of data

It was the launch of GPT that put ‘AI” on the map and the center of so many conversations. Generative AI opened up the opportunity to drive day-to-day uses from the most senior to the most junior marketer.

  “In the past, AI has been used largely for predictions or categorization. ChatGPT will create new articles, news items or blog posts, even school essays, and it’s pretty hard to distinguish between them and real, human-created writing,” according to Helen Lee Bouygues 

ChatGPT launched as a prototype to the public on Nov. 30, 2022. Within five days, more than a million people were using it. 

According to the latest available data, ChatGPT currently has over 100 million users. And the website currently generates 1 billion visitors per month. This user and traffic growth was achieved in a record-breaking two-month period (from December 2022 to February 2023

ChatGPT helped democratize technology. Finally, AI was accessible and easy to use with a direct impact. The tool was within everyone’s reach: free, easy, immediate  

Besides the obvious ‘magic’ of generative AI, why has this propelled us to new heights this year?

2 big things that differentiate this new version of Generative AI:

1. It gets smarter over time with more data. By giving these new systems a ‘rewards’ system to continue to improve, it gets better over time
2. It’s now ‘human-like’ - using ChatGPT, you realize how personal it can be and conversational. This was the first time it felt less like a ‘machine’ and more like a person

Impact of AI

The most obvious advantage of AI is the ability to save time but how does this translate to business impact?

1. Economic Value - over $ 13 trillion by 2030 (McKinsey article). According to a study by McKinsey & Company, AI could generate up to $13 trillion in economic value by 2030, with a significant portion of that coming from the marketing industry.
2. Time savings - Depending on how many repetitive tasks can be transferred to machines, one study revealed that by adopting AI such as chat bots and automated reporting, businesses can save an average of nearly 40 hours each week, or 2,075 hours per year.*
3. Cost savings - less ‘labor’ costs
4. More innovations - to solve complex problems

AI will drive hyper-personalization, hyper optimization, hyper automation

Time and Revenue Impact on Marketing

AI has the rare and unique capability to impact both the short and long term needs for Marketing. It can save marketers a significant amount of time almost immediately. According to a study by Salesforce, AI can save marketers an average of 11 hours per week. This time can be used to focus on more strategic tasks, such as developing new marketing campaigns and creating content. There are several ways that AI can also drive revenue for marketers. For example, AI can be used to:

• Personalize marketing messages and experiences for each customer, which can lead to increased conversion rates. [Revenue impact and Time savings]
• Predict customer behavior and make predictions about future outcomes, which can help marketers to target the right customers with the right message at the right time. [Revenue impact and Time savings]
• Automate marketing tasks - such as sending email campaigns, scheduling social media posts, and managing ad campaigns. These save marketers time and resources, allowing them to focus on more strategic tasks. [Revenue impact and Time savings]
• Create content, such as blog posts, articles, and social media posts, which can help marketers to reach a wider audience and generate more leads. [Revenue impact and Time savings]

Some specific channel examples take it even further;

• Email marketing: AI can be used to automate email marketing campaigns, such as sending out newsletters, promotional emails, and abandoned cart emails. This can save marketers a significant amount of time, as they no longer have to manually create and send out emails.
• Social media marketing: AI can be used to automate social media marketing tasks, such as scheduling posts, responding to comments, and running contests. This can free up marketers to focus on more creative and strategic tasks.
• Content marketing: AI can be used to create content, such as blog posts, articles, and social media posts. This can help marketers to create more content more quickly and easily.
• Customer service: AI-powered chatbots can be used to automate customer service tasks, such as answering frequently asked questions and providing support. This can free up human customer service representatives to focus on more complex issues.

By using AI to automate tasks and create content, marketers can save a significant amount of time and money. This time can be used to focus on more strategic tasks, such as developing new marketing campaigns and creating content.

As AI continues to develop, marketers will likely find even more ways to use it to drive revenue.

How can Marketers integrate AI into their business?

Paul Roetzer (Author of Marketing Artificial Intelligence) has a wonderful framework for approaching Marketing AI. I love this model because it covers all the key marketing needs. You can use this 1) to see what’s already been done in the category and 2) what potential opportunities to tap into 3) and break this down by tools to help determine opportunities or current gaps

He breaks down marketer's needs by the 5 P’s 

1. Planning -Building intelligence strategies [micro/macro]
2. Production -Creating intelligent content [micro]
3. Personalization - Powering intelligent consumer experiences [micro/macro]
4. Promotion: Managing intelligent cross-channel promotions [macro]
5. Performance - Turning data into intelligence [macro]

There is a tool for all these areas by each channel. In his book, Roetzer reviews vendors across email, SEO, SEM, Social media, content, website etc - it’s incredible the various tools that can be used. I highly recommend you read the book to get more details as he reviews numerous AI vendors by channel. Each day there is a new tools being introduced to help us every day (1 year before ChatGPT launched 40 tools, a year later over 100, prediction is to have 10K by 2030

Pros and Cons

While AI comes with great advantages, there are also valid concerns on the broad access to such powerful tools. With any new innovations, guard rails haven’t been fully established (we saw this with world wide web in the 90's and early 2000's)

How to get started

Key questions to ask yourself if you want to get started with AI:

1. Goals - What are your goals? What business needs do you have, machine will only address what you tell it
2. Data - Do you have data? Can you get the data? Where is it collect, where is it stored, is it integrated into other data sources, and how is it cleaned?
3. Repetitive tasks -Are there repetitive tasks that can be replaced to allow resources to be reallocated
4. Improvement - Can you improve on what you are already doing - faster, smarter, cheaper
5. Predictive - is there something that can predicted to help the task
6. Human sniff test - does it make sense? Are there any biases based on the data input? People are still needed for any AI initiative. Just like Google search requires someone to 'search' , generative AI requires someone to prompt and assess the output. It's an iterative process with 'humans' required in the drivers seat. True marketers know that the field depends on elements of imagination, creativity, and originality to make any impact with consumers

Justifying AI throughout the organization

I hope by the time you get to this part of the article, it’s clear AI is a critical part of any marketer's toolkit. However, how do you justify the investment to your leadership team? Here are steps you need to take:

1. Get buy-in through the numbers/results
2. Determine priority areas - where are your repetitive tasks, where is there already data, which will benefit the most (most impactful - cost savings, time savings, etc)
3. Use at the Micro and Macro levels. In other words, it’s not just the big projects but it’s also your day to day across each team member
4. Think about the 5 areas of marketing; planning, production, personalization, promotion, performance

Andrew Ng (Founder of DeepLearning.AI and luminary in the field of AI) suggests that you start with small pilots to determine what works through quantitative impact. I agree with this model and it’s very similar to how you would approach any digital marketing campaign, channel, or project. Because it's such a new focus area for many organizations, gaining some ‘wins’ under your belt to get c-level buy-in through the numbers. Once you have this in hand you can justify AI projects across the organization.

Additional tips: Look for a quick win pilots with narrow scope. Gather strong use cases that can be shared throughout the organization to get the alignment and credibility

Finally, the Human Context is probably the most forgotten element

Man vs Machine

I've heard the term “AI for good”.

We need to keep this as an overarching theme because AI can be used for good (autocorrect, mapping/traffic software, recommendation engines)) but also for bad (consumer privacy, biased programming, danger to humans, and unclear legal regulation).

AI is only as good as the one used to train it. Therefore, human intervention is still a critical component of the process. Nothing should be done without some ability to provide some human context or review. 

“It’s the balance between art and science using a human sniff test.”

 AI is getting so much smarter but it’s not always right. Some form of human intervention is still required. Man+Machine together will have a bigger impact then each one individually.

Remember AI is NOT perfect and is still being refined.

Check out Pepperoni Hug Spot - AI Made TV Commercial

In the fake ad, a Reddit user utilized artificial intelligence to create an advertisement for an imaginary pizza place called “Pepperoni Hug Spot.

While very amusing to watch, it's a hodge podge of some familiar scenes but cobbled together in a bizarre and scary way. We continue to see exponential advances each day but this is a great example of why we still can't rely solely on the machines right now.

Future of AI

Many have asked about the future of AI for Marketers. The role of marketing is to identify, satisfy, and retain customers. To do this requires driving personalization, better optimizations, more automation and new consumer experiences. These goals don't change but they will be boosted by improved AI capabilities. We'll be able to get there faster with more customization and less human labor work.

AI development is growing at an exponential rate. Given the role that marketers have with consumers, there is a responsibility to do the right thing. While AI has great potential, there are profound questions that are still being discussed. Google has declared their #1 principle of AI is to "Be socially beneficial". We should start thinking this way in our AI efforts.

Ongoing discussions include the following:

Ethics/Responsible use

Diminish Humanity

Society adaptions

New regulations, laws, and treaties among nations for safety to others

Data privacy

How to align AI development with human values and beliefs

Key Takeaways

AI is here! You can’t avoid it and you don’t want to.

AI is not scary - it can help you do things faster, smarter, and more efficiently

There’s still a lot more to be learned and developed with AI

Learn and expand your own knowledge with AI tools

Generative is just the beginning….

Key highlights 

1. There are a ton of micro-level developments. AI has been incorporated into many marketing tools already. Every task can be aided and assisted to help you. Generative AI has opened up new doors for day-to-day tasks. There are so many tools already in development, you need to find and use them.
2. Jobs won’t be lost (in the short term) but you can spend more time doing the non labor work. Studies show that the jobs that will be lost will be lower than the jobs that will be created. AI will help us be more productive.History has also proven that technology revolutions have created more jobs than it has destroyed. The beauty of AI is that it will help us focus on what people are good at and remove the repetitive, mundane tasks that are better off done by a machine. AI shouldn't take jobs away but instead assist and complement your current job.
3. As marketers, this is a time for MORE creativity. AI may not be ready for prime time but they are great as starting points. Generative AI will help us get to our final outcomes faster. You don’t have to spend all your time cleaning data, you can test more ideas, and You can Input more ideas, so you get to spend more time coming up with more creative ideas, insights, and innovations - this is brilliant - what an exciting time for marketers to push the limit
4. Data Quality - It’s all about quality of the input and output data. AI requires a strong data strategy. It works with good input (data) and output (impact from data) to be able to understand and learn from the data. Continue to get more data to drive more inputs/outputs to make AI smarter for your needs.
5. Human context is still needed. AI can’t run without any human oversight. There is an art and science to ensuring things make sense and telling the machine what to do. Nothing can replace the human ability for rational and emotional understanding. Humans still need to direct the machine on what to predict and do with these predictions. People will still need to be part of the equation :-) .

The future for marketers is very exciting! 

As a marketer, start incorporating AI into all your activities. Take a 'Learn and do' approach and start small. The micro and macro levels will impact the either the day-to-day (generative) but also broader scalable applications (machine learning, deep learning, etc) that will drive more efficiency, personalization, and remarkable consumer experiences. This is a time of marketing reinvention to realize the possibility of things we only imagined a few years ago. We talked about 1:1 cross channel personalization across the omni channel consumer experience. With AI, we continue to get closer to that vision.

For some, it will be a good test to begin pushing their digital and data efforts further, for others, it will be setting up infrastructure for an AI future

Good luck and happy AI’ing

Sources:

SF Marketing Report 2023

McKinsey AI 2023 Report

Marketing AI Institute, Paul Roetzer

AI for Everyone, Andrew NG

Customer Experience in the Age of AI, David Edelman- HBR

Jim Stern - Intro to AI

HBR: How Machine Learning Can Improve the Customer Experience

*AI Time Savings