May 26, 2025
In Summary
Looking for your next level of growth in email marketing? AI is one of the central considerations to explore when you’re optimizing email strategies. Here’s our lens view to how ESPs are expanding their functionality through AI adaptation.
* A break out of each ESP and how they are utilizing AI
* How these can best be used for scaling your email channel
* Using each AI tool for your own strategic roadmap
Take Aways:
AI within each ESP is different and is a long game to strategic optimization. Diverse approaches to generative, predictive, agentic AI, or data enrichment enhances the capacity of each system. The tools expand into use cases for best practices to optimization.
AI and email is best used with a strategic approach, looking at how it will read and enhance data, segmentation, and personalization within the next 2-5 years. Those who lead in the competitive marketplace will also have stronger mechanics by utilizing deeper data learning and enhancement to optimize and target sending.
The strongest competitors in the marketplace are utilizing advanced tactics with AI - combining layers of AI (e.g. predictability with generative), to target, personalize, and segment emails at new levels. Email marketers that utilize AI to optimize their strategies are leading the way in machine learning for targeted sending.
AI has taken center stage in the global community, and is targeting marketers to adapt. It is quickly becoming the diva that demands us to pay attention to how we market and who we are marketing to. As it continues to sing its overture, it is also setting the stage and introducing us to the new characters that are acting in the play. Surrounding the AI central character are ESPs and platforms that support an unfolding story in the era of technology.
ESPs are adapting to the AI stage quickly, ensuring that they are opening the curtain to the theatrics of disruption while playing on the stage of innovation. Each of these systems are using a combination of technical structures, ranging from data enrichment to generative or agentic AI. Their goal: playing out new ways to speak to our audience.
Those that adapt, especially with specific types of AI, will also embed a competitive edge that transforms fast scalability. And for those that don’t, ramping up when AI is maturing in the marketplace will become an endless staircase to get to the main theater, leading to the side stage performance with results that are not as advantageous. The more marketers adapt now, the more they will stand on the competitive edge of personalized and data driven results.
There are two major shifts in ESPs working with AI and adapting to new machine or deep learning. The first is related to leveraging the amount of data that ESPs have access to as a first party tool. As email lists have grown over time, they have also accumulated thousands, if not millions, of pieces of data. The second is to create prompt related predictability to make sending easier and to work with AI tools for better results.
Data within ESPs is both our strongest asset and weakest component. More data means more personalization and the ability to target and segment in ways that predict growth. However, flooding the system with too much data makes it impossible for marketers to track and target over time. The inundation of data within the system causes many marketers to rely on templated approaches versus the ability to strategically work with personalized data in the system. This constricts scalability and reduces the capacity to target strategically.
Sorting, segmenting, and identifying profiles is now a conglomerate of thousands and millions of pieces of data that marketers have to identify. Leveraging AI takes the groundwork out of data identification and personalization so targeting can become more accessible. This allows marketers to work in tandem with the AI machine with one chorus in mind - send to the right person at the right time.
The use of AI is extending to an experimental ground that helps machines to learn and build more within the system. Platforms offering AI are now focused on optimization through timing of sends, content, creative, segment testing, and personalization practices that introduce new layers to email communication.
The AI craze is causing many marketers to sign up for an autograph, dropping strategy and productive approaches to sending. The new tools offer marketers a free ticket to stop thinking strategically and instead to let AI do the heavy lifting. However, this also places the system at a detriment, making sending practices and strategies whimsical.
The key to leveraging ESPs and new, innovative AI tools is to work in harmony with strategy so accounts scale properly. High level strategy that utilizes AI for optimization is the standing ovation of scaling and sending - a shared recognition between email strategists and technology.
For AI to work with ESPs, strategists and those that understand how the learning systems adapt to personalization and data is key to viewing the entire show of how much AI can benefit and optimize email practices.
To introduce email marketers and ESPs, program development within diverse systems is the central component to adaptation and powering up data driven and personalization based results. The development of competitors today is going to alter the landscape, leading companies to consider migrating to the most sophisticated and innovative tools that support faster growth and a targeted communication infrastructure.
Klaviyo’s approach to AI is a guideline to help create a stronger foundation when you are launching and building your business. Their approach is to use it as a pivotal point or way to optimize and enhance what you have in the system while allowing those at the ground level of marketing to have guided support to create a robust email channel.
Klaviyo has divided their AI tools into major sections that are needed to optimize your email. Subject lines, creative, segmentation, send time, and predictive analytics are used as a prop for better practices. Their foundation is with the use of generative AI, using past data and information to provide insights on what to build next.
AI in Klaviyo uses a combination of pattern recognition, data inputs, and prompts to enhance sends. This means that AI will look at profiles to see what activities they take or what profile information is available, all which are prompts that Klaviyo has integrated into their segmentation building as a complete system, meaning what prompt you enter will match with prompts that are entered in all Klaviyo accounts for the best outcome. This allows Klaviyo to use AI as a guidebook on how to build your next target. Similar to ChatGPT, Klaviyo is centering its approach around prompting, using this as a leverage to give sending foundations.
The sample below highlights the use of generative AI to find specific segments without knowing the major definitions in the system. When entering an AI segment to find a group of people you know have specific attributes or actions (for example), AI will take the learning in the system to find the best data or patterns from the prompt entered.
The more effective you are at identifying segments, content, and send times through prompts, the more AI is able to respond and support the information you are looking for with optimized results. This offers a new, guided data layer to optimize results needed.
Attentive has launched four layers of AI into their system. The software points at the use of a combination of predictive and generative AI, allowing the machine learning to take what works over time and to target with their own systems in place.
AI has focused on the mechanics of how email marketers tend to use ESPs. It has broken into the mechanics of what the system is used for: pop ups, messaging, segmentation, and journeys. These are combined with predictive analytics that target the right person at the right time.
The list growth uses AI specifically to find the individual that has the “most likely to purchase” behavior with the pop up. If the machine is able to see that the individual is more likely to respond to a product vs a welcome journey, for instance, then it will tailor the results to make sure that the person sees the message that will most likely prompt them to sign up.
AI in Attentive for those who have already signed up uses data learning for two areas of optimization. Unlike other systems that focus on building lists and following prompts, Attentive’s AI uses machine learning for predictability.
Set up segments, add in your own layers of exclusions and add in AI. You will get a +10% or -10% on who the message should be sent to with “most likely to” logic from the learning system. The AI identifies the profiles that match to the segments you’ve worked with and optimizes it for better sending practices.
Similar learning and optimization practices are used for send times to journeys. AI based journeys are looking at profile data and changing messages to support what the individual is most likely to respond to. It’s using its own AI prompts through agentic AI to understand audiences and create reactions, boosting demand. By leveraging the information in the system, marketers are able to work in tandem to optimize the next stages of messaging.
To lay out new approaches with a red carpet, Attentive is working through targeted personalization. The use of deeper learning can continue to target each user profile based on the most likely response. Enhancing personalization through AI optimization will lead to a future that takes the right message to the right person to generation capabilities.
While most ESPs are using predictions and patterns with generative AI, HubSpot has taken a turn to focus specifically on B2B demands and markets that need more information based on profile data.
One of the powerhouse tools of AI in HubSpot is the adaptation of data enrichment. Marketers are able to find information without hours of search, clicking on LinkedIn or just “digging in”. Data enrichment brings information to you, cutting back on the research and approach to building.
HubSpot’s role out focuses specifically on best “source of truth” information. In the traditional world, ESPs worked with the use of entering an email, leaving marketers on their own to find any other enrichments related to a profile. This made surveys, custom profile fields, and other email responses the central focus of sends so marketers could learn more about their audience and build this into the system.
The bridge to AI enrichment is similar to the mechanics of CDPs, specifically which has the capacity to take third party data and send predictive analytics to specific segments and audiences. However, using these tools is sometimes unapproachable because of the data sophistication related to interconnectivity within the CDP.
HubSpot’s approach is taking out the CDPs and giving a base level of enrichment with the use of AI.
The tool in use is Breeze Intelligence, and it is able to identify segments not only based on prompts, but also through enrichment tools that are interconnected in the backend of HubSpot. This is a three tier opportunity, where it identifies buyer intent (based on activities), combines this with profile enrichment, and adds into individual records to enrich data that already exists.
This information empowers companies to have three cases for use:
1. Analysis: Identifying the number of people in a database that are compatible for targeted messages with the ability to segment and personalize.
2. Targeting: With the buyer intent analysis, combined with profiling, there is a higher capacity to understand cohorts of people in your database, similar to a foundational CDP. Not only does it support targeting through information, but instead of predicting through third parties, it has the capacity to enrich the data that exists in your system, offering a deeper layer of predictability.
3. Record based enrichment: Ever sent a message to someone where the information pulled was incorrect? HubSpot’s enrichment tool takes this out by updating and enriching records. It’s now possible to override individual profiles with specific and more cohesive information for better personalization practices.
As this builds, the system will be able to enrich more within each profile and work as an agentic AI machine to learn about the profiles in your system. The result is a pivot to understanding deeper components of profiles and getting targeted results.
HubSpot is not only using data enrichment as its main roll out. Breeze Intelligence works similarly with other components for content enhancement and automation predictability, giving those who are just diving into the system opportunities for scalability.
Looking for creative that targets? In the old style of marketing, email marketers would have to study their audience. Take the brand, look at the avatar, make determinations, reiterate, target, hit or miss the messaging, and then try again. Hours of research and thousands of dollars later, a specific strategy is applied, which often creates two segments: a centralized persona that is targeted and a second, marginalized set of profiles that are taken out of communication opportunities.
With Braze’s AI tool, personalized messaging happens at the level of enriched data, taking dynamic content to the next level. Whether it’s a complete copywriting to graphic assistant or personalization to hit relevancy, Braze is taking over the ability to study your audience and insert both creative and technical drive to target each message.
The result - layers of personalization that every email marketer dreams of but takes time to hit at the production level. No longer is the production of technical personalization needed, but the use of AI is. By using predictability in relationship to each profile, targeting messaging becomes prominent, inviting in broad sends with targeted messages.
Each ESP is taking the new challenge that the market demands: give us AI. 🍗 And as this is happening, AI is quickly playing out its best act to help the industry change direction and pivot into sophisticated marketing.
The marketplace demands are leading us into a new loop for targeting and scaling accounts. This is a three pronged approach that is inclusive of ESP functionality, AI adaptation, and the strategic knowledge of those scaling the email channel.
The most prestigious place that AI is taking email marketing at this moment is into higher levels and layered approaches where we use our knowledge in harmony with machine learning to reach and scale revenue. While the definitions surrounding AI are placed into creative, segmentation, and optimization, the root of the technology is in the innovation of data learning, offering personalization and targeting with every send.
This post was written by a real, live human and occasionally aspiring playwright. However, true to the diva nature, AI has decided to give me a one-liner epilogue. Here ya go: “AI didn’t just join the cast—it rewrote the script, demanded a trailer, and now insists on royalties.”
See all
© 2025 Structured. All Rights Reserved.
