Here’s a frightening reality: while 88% of marketers believe that AI and automation are essential to meeting customer expectations, but they are only of use 56% of tools buy them.
This discrepancy raises important questions: Are marketers investing in the wrong tools or are they simply not using their full potential?
I spoke to several marketers to understand how they are integrating AI into their core stack and what areas still have room for improvement.
From an 82% increase in email conversion rates to dramatic improvements in customer retention, their insights show that the right approach to AI can transform your marketing operations.
Let’s look at their key insights and some key tips on how to make the most of AI in your MarTech strategy.
Table of contents
The state of AI and MarTech today
We must accept that generative AI will become a central part of our organizations and will be integrated into almost all areas.
It is here to stay and will continue to spread whether we like it or not. Instead of trying to avoid the problem, think about how you can make the most of it.
One of the best comments on this topic came from a marketing expert Jessica Apotheker in one of hers TED talks.
Apotheker notes that over the last 15 years, marketing has evolved from a set of general skills to a set of more specialized skills.
This includes digital marketing or marketing technology. She notes that generative AI has now changed the core of marketing activities.
Quite a fascinating discovery – and I think it could make a big difference in our productivity and overall effectiveness as marketers.
That said, let’s dive into the specific operations of marketers where AI and MarTech work together today.
1. Lead generation and nurturing
Before AI BDRs and automation, manual lead generation was a real challenge. If you’ve ever been in this position, you know how stressful it can be. Luckily, AI has stepped in to make this process smoother and more manageable.
This is how AI makes this process smoother and more manageable. 👇🏼
Predictive lead scoring
Predictive lead scoring helps marketers prioritize leads based on their likelihood to convert. Tools like HubSpot’s lead scoring software Actually do this job for marketers.
Such tools use machine learning to analyze and optimize lead scores based on behavioral patterns and predefined criteria and automatically highlight “warm” leads. Additionally, it is self-training and adapts as your business grows.
Pro tip: HubSpot’s rating options let you switch between these traditional scoringwhere you set criteria like form submissions or page views, and Predictive evaluationthat uses AI to prioritize leads based on behavioral patterns.
This flexibility means you can go hands-on or let AI do the work, whichever best suits your goals.
Automated email marketing
AI in email marketing uses data analytics to personalize content, optimize subject lines, automate follow-ups, and segment audiences.
By tailoring emails to each recipient’s behavior and preferences, AI makes campaigns more engaging and effective, leading to higher open and conversion rates.
A great example of AI-powered email personalization:
Revolve increased its email effectiveness using Cordial AI and 16 data points to create tailored recommendations for each customer.
They used to send generic blocks of products, but now each email offers 32 unique product suggestions – like abandoned cart items, favorite brands, and trending products by location.
Tests have shown that these personalized emails doubled engagement, with a 65% increase in click-to-open and click-to-conversion rates, create the conditions for large increases in sales.
Pro tip: Always use behavioral triggers – like abandoned carts or product views – to automatically send follow-ups, creating a great experience that engages customers without overwhelming them.
I recommend Mailchimp for this purpose, especially if you automate personalized campaigns on a large scale.
It helps you welcome new contacts, recover abandoned carts, and win back lost customers with AI-generated automation and ready-to-use emails that you just need to review and send (very often, no customization is even required).
2. Creation and distribution of content
Creating and sharing content is another very time-consuming task. AI helps here too.
Content creation
AI has changed the way we create content in various formats, not just text. It can now generate videos for TikTok, YouTube and Instagram Reels.
I rely primarily on for my writing tasks breezeHubSpot’s AI tool for marketing and sales, and ChatGPT 4o. Breeze Copilot automates all types of content, like blogs and case studies, and allows you to brainstorm ideas for titles and clear subheadings.
What I like most about it is the pre-built templates and prompts that make starting tasks and brainstorming much easier.
When it comes to creating video content, InVideo has always been my first choice because it can produce amazing videos from even the shortest description.
For example, I asked it to make a before and after video of a home office remodel:
All I had to do was choose the audience, choose my preferred style and decide on the format based on the social media platform.
And here is the result:
Overall, a good job considering the minimal description. What I think is missing is a before and after comparison, but that’s my fault for not providing a more detailed prompt.
Pro tip: When using AI tools, be specific with your prompts. The clearer you are about what you need, the better the results will be.
Content optimization
AI tools are also game-changers when it comes to making our existing content shine. It analyzes what works by identifying popular keywords, suggesting relevant topics and headlines, ensuring content is easy to read, checking for plagiarism, and predicting how well it will perform.
I caught up Irina Maltseva, the growth lead Aura, and she told me how AI helps her:
“AI tools like AI Content Helper from Ahrefs and Clearscope have significantly accelerated marketing tasks, especially SEO and content creation. Before AI, I spent hours manually researching keywords, analyzing competitors, and hoping my clients’ content hit the mark.
Now AI does the heavy lifting – giving me instant keyword suggestions, content optimizations, and real-time ranking insights. Instead of guessing and waiting weeks for results, you can optimize as you go, saving time and increasing accuracy.”
One of the tools I use for this purpose is Surfer SEO. The real-time content rating feature is brilliant. Surfer compares your content to the highest ranking pages and provides informed suggestions for improving keywords and structure.
Pro tip: When it comes to surfer SEO specifically, while we all want the score to be green and over 90, sometimes that just isn’t realistic. If you push too hard, you’ll end up with keyword stuffing and a lot of confusion that you should definitely avoid. So use it as a guide, but don’t rely on it too much.
And another professional tip: always keep your content up to date. Update it regularly based on the latest trends and AI suggestions.
Social media management
Every social media manager knows how much AI has made our lives easier, especially when it comes to scheduling and creating content.
In addition, we all need detailed performance analyzes with suggestions for improvement.
To this end I love HubSpot’s AI social media post generator, This helps me turn my raw ideas into polished posts for Facebook and Instagram.
It adjusts the tone of voice to suit your brand, saves you time editing, and ensures your posts stay within character limits while balancing emojis and hashtags without going overboard (which I really like).
Additionally, the tool allows you to schedule multiple AI-powered posts without having to perform tedious manual updates. It also tracks the performance of your posts and gives you insights into what resonates with your audience.
Pro tip: Try A/B testing your social media posts. This is a great way to find what works best and refine your strategy over time.
3. Customer Experience and Support
AI also brought better customer interactions and support. Now people receive messages that feel natural and personal, making the entire experience more realistic.
Here you can find out how AI actually helps.
Chatbots
First of all, chatbots.
Machine learning allows them to learn from interactions and make predictions. AI-powered chatbots offer 24/7 availability, scalability, cost-efficiency, and improved customer experience.
When it comes to AI chatbot tools, I swear by them Drift. The ability to let leads schedule meetings directly from a chat is simply awesome.
Here’s the story that proves it:
1Password saved 16,000 support hours with Drift and in just six months 75% of support requests were rejected.
Before implementing Drift, customer support struggled with slow response times and inefficient processes.
This integration improved efficiency, allowed agents to focus on complex issues, and resulted in a customer satisfaction score (CSAT) of over 4.6 and an ROI of over four times.
And if you are in e-commerce or a similar field, I would highly recommend you ManyChat.
I’ve used it to create interactive experiences for my customers, especially on Facebook Messenger, and it’s amazing.
There’s nothing better than leaving common questions like delivery times to AI so you don’t have to answer the same thing over and over again. I also like how easy ManyChat’s templates are to customize and set up.
Pro tip: Run A/B tests on different conversation flows or scripts to see which ones get the best responses or conversions. Tests can show you which prompts, tone, or CTA produce better results.
Sentiment analysis
AI helps companies understand customer feelings through sentiment analysis. It monitors social media and online reviews to assess brand perception.
Brandwatch is one of the best tools for this purpose. It stands out for its ability to analyze conversations across multiple platforms in real time.
The best part for me is Listen, the emotion analysis tool that uses a custom logistic regression model to identify anger, disgust, fear, joy, surprise, or sadness.
Analyzing over 2 million posts, it examines features such as words, phrases, slang and emojis to predict the dominant emotion in the text. An accuracy of 60-70% is achieved for most search queries.
With these deeper insights, companies can respond appropriately to customers’ feelings, address negative sentiment quickly, and improve overall brand perception before everything “escalates” due to negative reactions.
Pro tip: Use sentiment analysis in conjunction with other analysis tools. For example, compare sentiment data with sales numbers or customer feedback to gain deeper insights into the impact of emotions on business results.
4. Advertising campaigns
AI analyzes data quickly so marketers can show personalized ads to the right people at the right time. This technology helps automate ad buying, improve ad designs, and perform A/B testing, which increases engagement and produces better results.
This is exactly how it works:
Optimized ad creation
AI tools make ad creation easier by testing colors and fonts to find what works best for different audiences, helping to eliminate personal biases. AI is doing excellent work in this area, supported by many positive case studies.
For example, RedBalloon, Australia’s leading online experience retailer, used Albert AI to optimize its ads and address rising customer acquisition costs, which peaked at $50.
By running over 6,400 keywords in 24 hours, they increased their audience reach from 1% to 99%. achieve a reduction in acquisition costs by 25%, a 40% reduction in total cross-channel costs and a 751% increase in Facebook conversions.
End result? An impressive ROAS of 3434% on new buyer campaigns.
Pro tip: Set clear success metrics – like CTRs or conversion rates – and let the AI run tests over a defined period of time. Use the insights you gain to quickly adapt your campaigns. For example, if a headline consistently outperforms the others, make it your first choice.
Optimization of ad design
AI increases ad creative optimization by testing different ad versions to determine the most effective ones.
For example, Facebook Ads Manager uses AI to analyze performance metrics and help marketers find the visuals and CTAs that resonate best with their audience.
His newest feature, meta advantage, automates ad performance even further with machine learning and achieves impressive results:
Similarly, Google Ads allows users to create multiple ad variations – different headlines, descriptions and display URLs – and automatically rotates them to collect performance data.
All of this promotes continuous improvement in conversion rates as the best-performing ads achieve higher engagement over time.
Pro tip: Upload diverse creative assets and switch to data-driven attribution for better bidding. Consider advanced inputs like customer acquisition goals and profit data to refine your strategy.
How to implement AI into your marketing tech stack
In order not to jump into AI thoughtlessly and try to implement everything quickly, you need to start with a plan and a strategy.
Eight steps I recommend you:
1. Assess your current MarTech stack.
Step one is to check yours MarTech stacks and analyzes each tool to see where AI can fill gaps.
For example, if data analysis takes a lot of time, consider looking into an AI tool that can quickly interpret and visualize data.
When reviewing each tool, consider its compatibility with AI. I usually check whether the tool has built-in AI capabilities or whether it can easily connect to other AI platforms.
It’s important to see if it’s flexible enough to grow with your needs or if it’s too limited to adapt as AI technology evolves. This step helps avoid investing in tools that may be outdated or expensive to update later.
2. Define clear goals.
When I define goals for AI in marketing, I start by identifying the specific results I want to achieve. Increase customer loyalty, achieve higher conversions or optimize our marketing budget?
Note: Always ensure that these goals are not isolated. And when I say “standalone” I mean they need to be in line with the broader business strategy.
For example, if the company is focused on expanding its customer base, I would aim to use AI to improve engagement and personalize experiences that attract but also retain new users.
And if my goal is to improve efficiency, I would turn to AI tools that reduce manual effort but at the same time reduce costs. With this focus, I ensure that AI is not just a flashy add-on, but a strategic part of my approach to achieving my business milestones.
3. Build a strong data foundation.
To build a solid data foundation, you need to ensure that the company’s data is clean, accurate, and consistent. I achieve this by regularly reviewing and cleaning the data, for example by removing duplicates and updating old customer contact information.
I also enforce strict privacy and security measures such as encryption and access controls to comply with GDPR and protect sensitive information. I also bring together data from various sources – CRM systems, social media and sales databases – into a data warehouse or data lake.
When you lay such a solid foundation, you set the stage for effective AI implementation and better decision making.
4. Choose the right AI technologies.
Different AI tools use different algorithms for specific tasks.
For example, machine learning (ML) examines past data to find patterns and make predictions. It checks a customer’s purchase history and can guess what they want to buy or do next.
On the other hand, Natural Language Processing (NLP) focuses on understanding human language. It analyzes customer feedback to see how people feel, summarizes long texts for quick insights, and runs chatbots that answer customer questions in real time.
Finally, deep learning is a subset of machine learning that uses neural networks with many layers to process data. It is ideal for tackling complex tasks such as image and speech recognition.
For example, a deep learning model can be trained on huge amounts of image data for object or face recognition.
5. Choose the right AI tools.
When choosing the right AI tools, consider things like cost, ease of use, scalability, and how well they integrate with the tools you already have.
Sometimes ready-made solutions do not cover all your needs. Then you could think about building custom AI models.
For example, a customized model can help predict customer churn by analyzing specific data patterns unique to your business.
Friendly advice: With all the new tools it’s easy to get caught up in the AI trend, but stay focused and don’t spend money on things you may not need.
Write down which tools will bring you value and which might just be a waste. Look for options that offer free trials or demos so you can test them out before incurring any costs.
Take part in a free training program for small business owners and marketers who want to improve their strategies with AI. Learn to drive content creation, personalize customer experiences, and ethically evaluate AI tools while developing skills to create effective prompts and optimize your marketing efforts.
6. Train your team.
Your team needs to understand what data literacy is and how to read insights. Consider conducting workshops or online courses that focus on reading and analyzing data effectively.
Next, provide training on AI concepts and tools. Set up sessions where team members can learn how to use specific AI tools that are important to your business.
Finally, consider the ethical side of AI, including issues such as bias and privacy. Host discussions or training sessions to ensure your team is aware of these challenges and knows how to use AI responsibly.
Here I have to quote the brilliant Jessica Apotheker again:
“You need to identify the true artists, the true differentiators and the true innovators in your function. If you’ve ever worked in marketing, you know who these people are. They are the ones who always disagree with you.
Now you take these people, and you have to strategically retrain them so that they can use AI well, for example, to be inspired by new ideas, to be inspired by new trends, to also crack quick prototypes, to their Multiplying the effect once they are cracked is a great idea.
“But you have to protect them and teach them to use AI to generate and produce original ideas. To do that, they need to use their human brains to keep those human juices flowing, and that in turn will protect your brand’s identity and your differentiation in the marketplace.”
7. Start small and scale smart.
Start with pilot projects that are small and manageable to test the possibilities. This allows you to experiment without overwhelming your resources.
For example, start segmenting your email list based on customer behavior and using AI to tailor content for each group.
Use an iterative approach to refine your AI initiatives based on feedback and results. After launching your personalized campaign, collect data on open rates and engagement. Use this feedback to optimize your messaging and target audiences as needed.
Additionally, track key metrics to assess how AI is impacting your overall marketing performance – for example, monitor conversion rates and customer retention to see if AI-driven changes are producing better results.
8. Monitor and optimize.
Regularly track the performance of your AI initiatives to see what’s working and what’s not.
Make adjustments as necessary to improve results. If you find that certain campaigns, such as targeted ads or chatbots, are not performing as expected, optimize your messaging, targeting, or algorithms used.
If a tool does not work as promised, contact customer support for help or consider switching to another solution if necessary.
Plus, stay up to date with the latest advances in AI and MarTech. Subscribe to industry newsletters, attend webinars, or join professional groups to stay up to date on new tools and trends.
Tips on how to make the most of AI in your MarTech operations
I don’t want to leave you hanging without some great expert perspectives, so I’ve put together a few more brilliant tips and case studies of what the HubSpot team has achieved with AI.
1. Make your customer chat more personal.
Customer chat today needs to feel like you’re talking to a real person.
The more natural and friendly you can make it, the better. It’s also crucial to give customers the answers they need directly in chat, without them having to dig through your website.
And here’s something Kyle Denhoff, Sr. Director of Marketing at HubSpot says on this topic:
“Deliver a better customer experience by improving on-site chat. Create a more personal, more contextualized experience for customers searching for information instead of having to manually navigate our knowledge base.”
HubSpot’s marketing team tested AI chat features and highlighted their importance for better customer interactions and increased sales. ⤵️
HubSpot onboards 3,500 users with AI Chat in week 1
How did the experiment begin?The initiative began with a focus on website chat as it had great potential to provide real value to users.
Since many customers turn to chat for support and product inquiries, they chose high-traffic sites like the knowledge base for their initial testing.
In the first week alone, they interacted with 3,500 customers and gained some valuable insights from those interactions.
How was the process? Once the AI chat was live, a dedicated team was assigned to monitor the project and collect data for training.
They started with the chatbot on the knowledge base pages and analyzed historical conversations to help the AI better understand customer needs.
The first use case was all about chat, but it quickly expanded to include the in-app pricing page, which aimed to guide potential customers through product options and pricing.
By the second week, they had already processed over 1,000 requests through the AI!
What were the results? During the testing phase, they initially noticed a decline in customer satisfaction, which dropped to 70% as the AI learned from real-world interactions.
However, as “AI became smarter,” customer satisfaction scores rose back to 85%, eventually matching those of human interactions.
This really reinforced the idea that AI, when properly trained, can significantly improve customer experience and increase conversion rates.
2. Create hyper-personalized emails.
I don’t even have to tell you not to use it “Dear Customer” as an intro, right? Use the customer’s name instead. But even that isn’t enough.
Try to go beyond that – point to something current, such as: “I saw that you downloaded our digital marketing e-book. What part do you like best?”
Tailor your messages to their interests. Small gifts, like birthday greetings – “Happy birthday, Sarah! There is a discount here” – make a difference. Track purchases with: “Hello John, I hope you like your coffee maker! Have you tried any recipes?”
Finally, ask for feedback: “Are you happy with this or is there something wrong?” Anyway, let us know!”
These tips will help you connect better with your customers. Kyle Denhoff supports this with the words:
“We worked with the AI/MarTech team to create personalized outreach emails for high-value contacts in our database. These hyper-personalized emails significantly increased response rates.”
But it doesn’t just increase the response rate. HubSpot’s AI experiment revealed that personalized emails can also significantly increase conversions. ⤵️
The AI strategy that Increased Email conversion rate up 82%
How did the experiment begin?HubSpot started by identifying projects with the best ROI potential. They used a simple 2×2 matrix to narrow their focus to around 10-15 key ideas.
Suggestions came through Google Forms and they made sure everything was organized on Slack.
Your mantra? Speed trumps perfection – get it to market quickly and then tweak as needed.
What was the process like? Once their ideas were solidified, HubSpot formed a central AI team and began an iterative process.
They started each project immediately, collected feedback, and made adjustments to the AI models based on real-world usage.
This allowed them to craft personalized AI-driven emails This was really well received by certain users.
What were the results? The results were breathtaking.
By tailoring emails to specific audiences, HubSpot saw a huge increase in campaign engagement and an impressive 82% increase in email conversion rate.
3. Let the AI process the data while you tell the story.
Yes, AI is great for data management, but don’t let it “take the chair.” You are still responsible for shaping the narrative and engaging with your audience.
As the AI crunches numbers and identifies trends, it’s up to you to transform those insights into something understandable and engaging.
As Irina Maltseva says, “Don’t just use AI tools as a quick fix, but integrate them into your entire workflow.” AI should give you insights, but it’s your human touch that makes the content engaging and authentic.
Basically, let the AI do the data processing, but you take care of the storytelling. This way you get the best of both worlds – speed and optimization without losing the personal touch.”
AI is here. Are you?
Let’s finally stop using this expression “AI is the future of marketing.” It is the present. It saves us a lot of work.
Sure, we may not spend that extra time meditating or just unwinding, but we definitely have more time for A/B testing, ideation, and even extra money from the manual work of the past.
It all starts with integrating AI into your marketing technology stack. Use it wisely and you will definitely not overlook the benefits.