Apple •Spotify• Pocket Casts •Youtube •Overcast •RSS
What’s up everyone, today we have the pleasure of sitting down with Stephen Stouffer, Director of Automation Solutions at Tray.ai and the first ever returning guest. We had Stephen on earlier in the year in episode 112 where we unpacked the practical wonders of combining AI tools with iPaaS solutions.
Summary: AI can transform your marketing without overwhelming you. Start with one use case. Watch the results, and go from there. You don’t need to master data science to add AI value, but you need to be willing to experiment, keep what works, and let the tech do the heavy lifting.
Jump to a Section
- Customer Journey Mapping Essentials
- AI’s Role in Automating Personalized Emails
- Automating Personalized Outreach with AI Agents
- Challenges in Implementing AI-Driven Customer Journey Mapping
- Getting Started with AI for Outbound Personalization
- Leveraging AI for Lifecycle Personalization
- Smarter Content Choices with AI
- Adding Human Guardrails to AI Messaging
- Use Case Example: Automating LinkedIn Lead Gen Forms
Customer Journey Mapping Essentials
Customer journey mapping, as Stephen puts it, is best approached as a clear, structured framework. For marketers, this often starts by examining the visitor’s first few seconds on a website. Stephen’s “three, five, seven rule” is a useful guide: three seconds to capture attention, five to build engagement, and seven to prompt action. Reviewing homepage or landing page performance through this lens keeps the focus on essentials. Are calls-to-action (CTAs) clear and accessible? Does the page guide users toward the intended outcome effectively?
Stephen further notes the importance of every element “above the fold.” Content here needs to be concise, visually appealing, and should naturally lead users to the next step. A well-placed CTA, such as a prominent button, encourages forward motion, while a hidden or confusing one can derail the journey. Each interaction should be straightforward and intuitive.
Beyond landing pages, Stephen highlights the journey before a visitor even arrives. Campaign managers, for instance, should ensure that ad copy and visuals align with the landing page, creating a smooth transition from ad to action. Consistency here reduces friction and keeps the experience cohesive.
For advanced mapping, Stephen recommends storyboarding different customer personas and their digital pathways. By tuning each stage to fit these profiles, marketers can craft a journey that feels relevant and trustworthy, engaging each segment from the very first interaction.
Key takeaway: Use the “three, five, seven rule” to evaluate each customer touchpoint on your homepage or landing pages. This approach helps ensure your content captures attention, fosters engagement, and prompts action—all within a few seconds.
Back to the top ⬆️
AI’s Role in Automating Personalized Emails
Stephen recently demonstrated how AI automates personalized emails with just a first name, last name, and email. AI uses data from sources like LinkedIn, company details, and job history to craft messages that feel genuinely tailored to each recipient, far beyond typical generic responses.
This level of automation doesn’t just boost engagement; it saves significant time. Instead of setting up complex variable fields or spending 15-20 minutes per email on manual research, AI handles it all in seconds. Stephen notes that marketing teams can skip intricate field configurations, while sales teams gain back valuable time to focus on high-impact tasks.
AI also serves as a replacement for traditional enrichment tools, pulling in dynamic contact details without third-party data providers. For sales, it means delivering relevant, personalized content effortlessly. AI does the heavy lifting, creating an email that feels custom-built for the recipient—no manual assembly required.
Key takeaway: AI enables efficient, data-rich personalization for customer outreach, saving marketing and sales teams time and resources while boosting the quality of each touchpoint.
Automating Personalized Outreach with AI Agents
AI agents are redefining how teams approach personalized outreach, offering new ways to automate highly customized interactions. Stephen explains how Tray.ai leverages a powerful combination of APIs—OpenAI, Google, LinkedIn, and more—to build out complex automation processes directly within its platform. Each AI agent is designed to use the best tool for the task at hand. Given the right context and instructions, these agents can gather relevant data from press releases, Crunchbase, LinkedIn profiles, blog posts, and other sources to craft an email that feels genuinely tailored.
Imagine a marketing email generated entirely by an AI agent. With the recipient’s email, role, and other contextual clues, the AI might produce a message like, “Hey, congratulations on your recent speaking slot at AntiCon in London. Hope you had a safe journey back!” This level of personalization would usually require about 15 minutes of research by a BDR or ISR. Now, it can be fully automated, freeing up sales and marketing teams to focus on strategy and high-priority tasks rather than time-consuming data gathering and crafting.
Stephen points out that the true power of AI agents comes from implementing them in real, tangible ways. For instance, rather than abstract promises of efficiency, Tray.ai demonstrates AI’s impact with practical use cases like this automated email personalization, which resonates more directly with the people using it. By creating a functional demo that allows teams to see this technology in action, Tray.ai bridges the gap between AI’s potential and its practical application.
For anyone curious to test it out, Stephen offers a live demo of the personalized email automation. This hands-on approach helps users understand the realistic possibilities AI agents bring to customer engagement and outreach, transforming the process from concept to actionable, impactful workflows.
Key takeaway: AI agents streamline personalized outreach, combining data sources and automation tools to generate highly customized emails without manual research. By automating these tasks, teams can focus on high-impact activities while still delivering meaningful, individualized interactions.
Challenges in Implementing AI-Driven Customer Journey Mapping
Implementing AI-driven customer journey mapping and personalization comes with its share of challenges. Stephen highlights three primary obstacles teams face: complexity, connectivity and compliance.
- The technology’s complexity. Even technical professionals sometimes struggle to understand the nuts and bolts of AI integrations, making it difficult for organizations to determine how to effectively deploy these tools. The ambiguity around building and customizing AI solutions internally often becomes a barrier to adoption.
- The challenge of data connectivity. For AI agents to deliver relevant outputs, they need access to comprehensive data across systems. Whether it’s a CRM, sales records, or product usage data, these inputs provide the context for AI to make useful recommendations. While crafting a prompt might sound straightforward, gathering and linking all the necessary data to inform that prompt is anything but simple. Stephen explains that AI can only be as effective as the information it’s fed, making seamless data integration a top priority for effective personalization.
- Perhaps most daunting, the challenge is compliance. When feeding sensitive data into large language models, teams must navigate a maze of security requirements like SOC 2, HIPAA, and GDPR compliance. Many organizations hesitate to dive into AI because of the fear of regulatory risks. Legal teams often step in, concerned about the consequences of mishandling data or overlooking specific compliance rules. Without a clear understanding of these regulations, even the best AI strategy can face roadblocks.
These hurdles—technical understanding, data integration, and regulatory compliance—highlight the complexity involved in using AI for customer journey mapping. Each requires thoughtful planning and collaboration across departments, from IT and data to legal.
Key takeaway: Implementing AI for customer journey mapping involves technical complexity, comprehensive data integration, and strict regulatory compliance. Understanding and addressing these challenges are essential steps toward effective AI-driven personalization.
Getting Started with AI for Outbound Personalization
To effectively implement AI in outbound personalization, Stephen recommends starting with a clear, simplified approach. The first step is gathering requirements and defining basic objectives, rather than aiming to create a complex AI agent right out of the gate. He suggests beginning with one or two data sources, like LinkedIn or Google, or even OpenAI’s base model. While this setup may not produce hyper-personalized emails immediately, it’s a practical step that allows teams to experiment with automation and understand the foundational processes.
For teams interested in exploring AI agents in a hands-on way, Stephen notes that Tray.ai offers free weekly workshops. These sessions give participants full access to Tray’s platform, allowing them to explore the tools directly, test different agent setups, and get comfortable with various AI use cases. This approach gives RevOps or marketing teams a low-commitment way to see how AI agents can automate aspects of outreach.
By participating in these workshops, teams can dive into the details of AI-driven email personalization without feeling pressured to become Tray customers. Stephen’s focus here is to demystify the technology, helping people understand how to leverage AI tools for their unique business needs. The workshops provide a way to experiment with everything from enrichment options to API integrations, offering participants a better understanding of what’s possible with even simple agent configurations.
For those seeking additional guidance, Stephen encourages LinkedIn connections and is open to casual coffee chats. His approach is decidedly non-salesy, focused on sharing insights rather than pushing a product. This way, teams new to AI-driven outbound can get practical advice and support as they develop their own use cases.
Tray.ai Workshops: https://tray.ai/build-your-own-ai-agent
Key takeaway: Start AI-powered outbound personalization by defining basic objectives and experimenting with one or two data sources. Tray.ai’s free workshops offer a hands-on, no-pressure environment to explore AI agents, providing insights into how teams can streamline their outreach through automation.
Leveraging AI for Lifecycle Personalization
Lifecycle marketing thrives on personalization, but creating a truly customized experience requires the right data and dynamic content. Stephen outlines a practical approach for using AI and data enrichment to tailor interactions based on where a customer is in their journey. By tapping into tools like Clearbit Reveal and other enrichment providers, organizations can “de-anonymize” web traffic, connecting anonymous visitors back to individual contacts or accounts in their CRM. This linkage opens up a world of possibilities for enhancing the customer journey at every stage.
Once an anonymous user is identified and matched to a lead or contact in Salesforce, Pardot, Marketo, or HubSpot, you can pull relevant information—like their account details, open opportunities, and purchase history—into your website’s content delivery system. This allows for dynamically tailored content on a scale beyond basic personalization. Imagine a homepage that adapts based on a visitor’s status: for prospects, it might highlight a “Get a Demo” button, while existing customers might see options to reach customer support or schedule a call with their account manager. This level of customization can turn your homepage into a highly effective, responsive experience for each unique visitor.
With AI’s help, lifecycle marketing doesn’t just stop at basic dynamic content; it can tailor entire customer journeys based on real-time data. For example, a customer with a basic product tier could be nudged towards an upsell on their next site visit, while a high-value lead could receive content more aligned with their specific needs. This intelligent targeting reduces friction along the journey, making it easier for customers to find relevant content and services as they interact with your brand.
Stephen explains that tools like Tray.ai play a pivotal role here by integrating data from multiple systems and enabling AI-driven decisions based on those insights. AI can automate this “intelligent content” approach, pushing the right information at the right time to maximize engagement and conversion. The end result? A frictionless experience that aligns with each customer’s needs without the manual intervention that traditional lifecycle management often demands.
Key takeaway: AI-driven data integration enables a deeper level of personalization in lifecycle marketing. By connecting CRM insights with dynamic content, brands can craft experiences that evolve with each customer’s journey, reducing friction and enhancing engagement across every touchpoint.
Smarter Content Choices with AI
For most, creating personalized customer journeys used to require rule-based orchestration with intricate if-then pathways and behavioral triggers. Stephen explains how AI can simplify this by selecting the right content dynamically. His approach is to feed AI a central document containing all the necessary details: product descriptions, ideal customer profiles, and messaging guidelines. This document acts as a content hub, accessible to both sales and marketing, where they can add or refine product information and targeting criteria.
With this setup, AI has the context to make informed content decisions. Instead of setting up rule-based conditions—like triggering an email on Day 3 only if certain actions aren’t completed—AI draws from the document to determine what content is most relevant. For example, if a customer is on a base tier, AI can automatically recommend messaging about an upgraded plan that aligns with their needs, without relying on manual rules.
Stephen emphasizes that this document-centric approach replaces the old-school mind maps and whiteboards of traditional journey mapping. By embedding both the user’s current stage and the document’s content, AI identifies and suggests the best product, message, or offer for each user. This means that everything from emails to landing pages can adapt in real-time, adjusting based on where the customer is in their journey without constant manual adjustments.
This streamlined process doesn’t just save time; it enhances the precision of customer engagement. With AI at the helm, journey mapping becomes less about managing triggers and more about delivering timely, context-aware content that resonates with each customer’s unique path.
Key takeaway: Using a centralized content document, AI can dynamically select the best messages and offers for each customer, eliminating the need for complex rule-based workflows and enabling more effective, scalable personalization.
Using Historical Data to Improve AI-Driven Messaging
Integrating historical performance data into AI-driven outreach is a game-changer for tailoring customer interactions if you aren’t doing that already. Stephen is exploring this potential, envisioning how feeding AI past data on open rates, click-through rates, and successful sequences could elevate email outreach. By allowing AI to study the best-performing emails and identify patterns, teams could automate content that builds on past success rather than starting from scratch every time.
In this approach, AI isn’t just generating content—it’s learning from existing outcomes. Stephen sees value in providing AI with comprehensive context, including details about the customer, company, and product offerings, layered with performance metrics from past campaigns. Imagine an AI that can analyze a specific rep’s email with unusually high engagement, understand its strengths, and apply those tactics across other outreach efforts. The potential extends beyond email: AI could make similar adjustments on landing pages, examining elements like button color, layout, or offer positioning to replicate top-performing designs.
Stephen explains that expanding AI’s input with Google Analytics data could help automate what currently requires detailed A/B testing. For example, if certain call-to-action buttons convert better in a particular color or certain page layouts drive more engagement, feeding that data to AI could standardize these high-performance traits across all customer-facing content. This way, the AI tailors its output based on a deep bank of insights rather than basic prompts, aligning more closely with what’s proven to work.
While Stephen hasn’t fully implemented this system yet, he sees it as the natural next step in making AI a true partner in outreach strategy. Rather than manually analyzing and adjusting each campaign, teams could let AI handle these adjustments, freeing up time to focus on strategy and creativity.
Key takeaway: Incorporating historical performance data into AI-generated outreach adds a new layer of intelligence, allowing AI to learn from successful past interactions. This approach can improve both efficiency and effectiveness, making every email or landing page tailored to what has already been proven to work.
Making AI Personalization Easy for Every Marketing Team
AI-powered personalization is evolving quickly, yet a major hurdle remains: making these tools accessible to teams that lack dedicated data scientists. Stephen highlights how Tray.ai and his team are tackling this problem, focusing on simplifying advanced capabilities so that any marketer can harness AI’s potential without needing a technical background. As he explains, while enterprise-level companies with data science resources can already run complex propensity models and real-time A/B tests, smaller teams are often left out due to limited resources and budget constraints.
For Stephen, the goal is clear: bridge this gap by creating solutions that bring high-level AI experimentation within reach. His team’s mission is to build platforms that enable marketers to orchestrate personalized customer journeys, test incrementally with control groups, and leverage real-time data—all without the need for a specialized engineering team. In a time when companies are trying to achieve more with tighter budgets, these accessible tools become crucial for staying competitive.
What this means practically is that smaller teams will soon be able to create their own data-informed strategies, using models that continuously test and refine messaging based on real-time results. Imagine a marketer wanting to send a discount offer only to trial users with a high likelihood to convert—AI can automate this, fine-tuning the messaging based on live performance data and providing incremental lift insights. Stephen’s vision is to allow any team, regardless of size, to operate at the level of a fully resourced data science division.
This kind of democratization is pivotal as AI continues to grow in complexity. Stephen and his team aim to take the complex data science that powers high-impact campaigns and package it in a way that’s accessible to all. As a result, AI-driven personalization and optimization are becoming tools for any marketer, not just those with enterprise-scale budgets or engineering support.
Key takeaway: Bringing advanced AI personalization and real-time experimentation within reach is essential for modern marketing. By simplifying these tools, platforms like Tray.ai enable teams to deliver data-driven personalization without needing a dedicated data science team, making high-level AI capabilities more accessible and actionable for all.
Building Data Literacy for the Next Generation of Marketers
For marketers looking to up their data game, Stephen suggests focusing on the fundamentals of how data works, rather than diving too deep into specific coding languages. He believes that foundational skills, like SQL and JavaScript, provide a good grounding since they’re universal, but insists that it’s more important to understand data’s role and potential uses. As Stephen puts it, we’re entering a world where tools like ChatGPT can generate code for us, but if you don’t know what you’re trying to achieve, you’ll still face barriers. Data literacy is now less about syntax and more about knowing how to harness data’s value.
Stephen emphasizes the importance of curiosity and hands-on experimentation. Many platforms offer AI or agent workshops, including those from his own team at Tray.ai, that allow marketers to get under the hood of the tech, exploring different use cases in a low-stakes environment. These workshops are perfect for trying out basic data scenarios, even without a full understanding of coding. Signing up for free trials, Stephen advises, is one of the most accessible ways to build confidence and see firsthand how these tools function.
Beyond technical skills, it’s critical to approach data with the right mindset. Instead of being intimidated by coding or analytics, marketers should aim to get comfortable with data’s building blocks, asking questions like, “What information do I need, and how can it be applied?” Stephen points out that embracing data literacy means developing the habit of being inquisitive and willing to test out various scenarios, even if it’s just basic at first.
Marketers who engage with data in this way will gain insights into what’s possible, which in turn will improve their ability to ask the right questions—an essential skill in an AI-enhanced environment. Stephen notes that the next hurdle won’t be writing JavaScript or Python but knowing what to request from AI tools to create meaningful outputs. For marketers, this means building intuition around data’s practical uses and applications.
Key takeaway: For the next generation of data-literate marketers, foundational understanding is key. Learning basic coding can help, but cultivating curiosity, exploring real-world use cases, and experimenting with free trials will provide the most value in developing data skills that matter.
Adding Human Guardrails to AI Messaging
Striking the right balance between AI automation and human intervention in customer journeys is essential. Stephen approaches this by setting up “guardrails” within his AI-powered processes. He includes an “out clause” in the document fed to AI—essentially, instructions for scenarios where none of the usual responses are suitable. For instance, if a customer tries to manipulate a prompt to receive a discount that doesn’t exist, the AI knows to notify Stephen or customer support rather than responding inappropriately. This fallback helps maintain quality control and ensures human oversight when needed.
Stephen believes that AI fits most naturally in the pre-sales and post-sales phases of the customer journey. Before a sale, AI can deliver dynamic, personalized content that helps guide prospects through early interactions, like exploring product options or understanding different pricing tiers. After the sale, AI can support customers through chatbots and automated ticketing systems—a space where Stephen is optimistic about improvements in AI-driven chat. Although current chatbot experiences can feel underwhelming, he foresees these tools evolving to provide much more value to customers in need of quick answers or support.
In the critical middle stages, however—where leads are nurtured, qualified, and transformed into marketing-qualified leads (MQLs)—Stephen argues that human involvement is crucial. This phase benefits from real conversations and personal touchpoints that AI can’t fully replicate. Humans excel at understanding complex needs, listening to pain points, and adapting demonstrations to meet specific challenges. For Stephen, these interactions, including product demos and proofs of concept, require the nuance that only human agents bring.
Even after the sale, Stephen sees an ongoing need for human engagement. While AI can handle many routine interactions, high-stakes or complex queries often demand a personal touch. For him, maintaining a balanced approach means continuous testing and evaluation, adjusting automations where they fall short, and ensuring a reliable “out clause” to reroute any issue that the AI might not handle optimally.
Key takeaway: To optimize customer journeys, leverage AI for scalable automation in pre-sales and post-sales stages while reserving the human touch for critical interactions that require deeper insight. Establish clear fallback options within your automations to keep the experience controlled and responsive.
Safeguarding Sensitive Data in AI Workflows
For industries with strict regulatory requirements, like health tech and finance, managing sensitive data in AI-driven workflows presents unique challenges. Stephen points out that in these sectors, it’s critical to tread carefully with automated messaging, especially when dealing with personally identifiable information (PII) or confidential data. Rather than allowing AI free rein, organizations must set up robust protections to ensure compliance with regulations such as HIPAA, SOC 2, and GDPR.
To address this, Tray.ai includes features like Merlin Guardian, which tokenizes sensitive information before any data reaches large language models (LLMs). This tokenization process masks confidential information, allowing teams to leverage AI without exposing sensitive data directly. Stephen emphasizes that establishing these safeguards is foundational—organizations should prioritize data security before diving into advanced AI applications.
For professionals in regulated sectors, the need for manual review and human intervention isn’t going away. While AI can automate and streamline processes, human oversight remains necessary to ensure that all compliance boxes are checked. Stephen recommends that teams handling sensitive data attend Tray.ai’s workshops, where they can explore practical, hands-on use cases tailored to maintaining regulatory compliance while still benefiting from AI capabilities.
In these workshops, Tray.ai demonstrates ways to navigate complex requirements in a way that’s compliant, secure, and effective. By learning from tangible examples, teams can better understand how to implement AI in their workflows without compromising data security or regulatory standards. This approach provides peace of mind and a clear path forward for teams aiming to responsibly incorporate AI into highly regulated environments.
Key takeaway: In highly regulated sectors, safeguarding sensitive data is essential when integrating AI. Tools like Tray.ai’s Merlin Guardian can tokenize confidential information, helping teams maintain compliance while leveraging AI-driven automation responsibly.
Use Case Example: Automating LinkedIn Lead Gen Forms
Stephen shared a powerful example from his previous company, where automating LinkedIn lead generation forms dramatically boosted conversion rates. Initially, the process involved a cumbersome export-import cycle: LinkedIn lead data was manually exported to a CSV file, which was then imported into Salesforce every one to two weeks. For anyone in sales or marketing, it’s clear that such a delay can hurt engagement. By the time leads were contacted, they had often forgotten the form submission entirely.
To streamline this, Stephen’s team used the Tray platform to automate the lead intake process. They integrated LinkedIn’s forms directly with Salesforce, setting up AI-driven data cleaning to handle any inconsistencies. One of the main hurdles with LinkedIn forms was the open text fields for state and country, where users could input any value, often needing manual adjustment. With AI, they established an “acceptable values” list to auto-correct entries, and for anything that didn’t match, AI would label it as “unknown” as a fallback. This automation brought down the lead-processing time from two weeks to mere milliseconds.
The results spoke for themselves. With this seamless workflow, the company saw a conversion rate jump of about 40–50% from marketing-qualified leads (MQLs) to sales-qualified leads (SQLs) and created more stage-one opportunities. Stephen emphasized that a simple, three-step automation—intake, AI-driven cleaning, and Salesforce submission—made a measurable impact on revenue growth.
This case illustrates the potential of AI to transform manual workflows, especially in lead management. It’s a reminder that impactful AI implementations don’t always have to be complex. In this instance, the simplicity of the setup allowed for faster engagement, which is essential for lead nurturing and sales success.
Key takeaway: Even basic AI-powered automations can yield significant gains in lead conversion. By automating the intake, data cleaning, and CRM entry, businesses can engage leads in real-time, increasing the likelihood of conversion and revenue impact.
Find Time for Skill Building
For RevOps and marketing ops teams looking to stay ahead amidst constant changes, Stephen recommends a practical, inward-focused approach. Rather than diving into organization-wide changes, he advises starting with a close look at your own daily routines. Track where your time goes—whether it’s hours spent combing through emails or repetitive tasks that eat into your day. Identifying these areas allows you to focus on automating routine tasks, freeing up time to focus on high-impact projects.
Stephen highlights that this freed-up time is crucial for staying competitive. Once automation reduces the manual workload, you can allocate 45 minutes to an hour to valuable activities like attending workshops, diving into webinars, or simply catching up on the latest trends. This combination of optimizing for efficiency and prioritizing self-education allows teams to tackle current challenges while building skills that will serve them well as tech continues to evolve.
One key point Stephen stresses is to keep your initial focus on problems within your current scope. Rather than spreading efforts thin across various departments or looking for solutions organization-wide, zero in on the issues closest to you. This approach enables a quick impact on your workflow, while the learning gained along the way will naturally broaden your capabilities. Over time, as you solve these smaller problems, your understanding deepens, allowing you to take on increasingly complex challenges.
Future-proofing isn’t about implementing every new technology that comes along; it’s about creating a sustainable workflow that continually adapts to immediate needs. As you learn through incremental improvements, the nature of your challenges will shift, bringing new insights and tools into play. Stephen’s approach is all about building a tech stack that serves current needs while also allowing room to pivot as necessary.
Key takeaway: Start by automating repetitive tasks in your daily workflow to free up time for skill development. Focus on optimizing your immediate responsibilities, gradually expanding your knowledge and solutions as you learn. This approach builds a resilient, future-ready tech stack without overwhelming your team.
Getting Started with AI in Marketing Operations
For marketing ops teams just stepping into AI, the path can seem daunting—especially when convincing senior management to prioritize it. Stephen recommends starting with a “champion” approach: appoint one person to explore AI applications while others focus on day-to-day tasks. This dedicated role can gradually build interest across the organization, especially when others notice tangible improvements. Over time, this curiosity often sparks a ripple effect, drawing in more teammates from various departments and cultivating an environment receptive to AI innovation.
Stephen also advocates for internal AI hackathons as a powerful way to generate excitement and uncover practical applications. Hosting a one-day event across departments can encourage diverse teams to explore AI tools like OpenAI or Anthropic, tackling predefined challenges or testing out experimental ideas. These hackathons give teams a chance to play with AI without any pressure, making the learning experience engaging rather than overwhelming. Stephen notes that seeing AI’s potential firsthand often convinces executives of its value, prompting budget allocations and broader adoption.
According to Stephen, senior leaders often need to witness AI’s direct impact on the organization to fully buy in. Whether it’s a simple automation or a creative AI-driven project from a hackathon, these efforts can make the business case for AI clearer, helping it become part of the strategic conversation. In larger companies, even board members can gain insight from these initiatives, recognizing how AI can support long-term business goals.
Ultimately, Stephen underscores that getting AI off the ground in a corporate setting involves both grassroots enthusiasm and executive buy-in. By fostering a culture of experimentation, organizations can gradually expand AI’s role, transforming it from a curious experiment into a valuable asset.
Key takeaway: Start your AI journey by designating a champion to explore and share AI’s possibilities. Consider hosting AI hackathons to spark interest, allowing team members to experiment and showcase value to executives. These grassroots initiatives can help AI gain traction, leading to greater support and integration across the organization.
A Shift in How We Approach Paid Media and SEO
Stephen points out a critical shift happening in marketing operations, especially in how we approach paid media and SEO. Historically, marketing has focused on “hacking” search engine algorithms to push pages to the top of search results through precise keyword targeting and optimization tricks. But this strategy is losing ground. The search landscape is evolving as new AI-driven models, such as ChatGPT and Anthropic, step in, changing how people discover information. As Stephen observes, platforms like Bing already look very different from just a few years ago, reflecting this shift in information retrieval.
Rather than obsessing over keyword placements and search rankings, Stephen suggests pivoting to a strategy focused on genuinely solving customer pain points and building quality content. As search engines and AI tools become more advanced, they’re increasingly tuned to recognize value and relevance over optimization “hacks.” Stephen emphasizes that brands will need to stop trying to outsmart algorithms and focus on understanding and addressing what customers need, creating content that naturally aligns with those needs.
In this new era, Stephen sees a future where lead sources won’t necessarily flow from traditional search engines. Instead, AI models trained to assist users with specific tasks will guide them to resources, products, or brands that best match their queries. This shift means marketing teams should prioritize creating authentic, helpful content and making the user experience as seamless as possible. By fostering trust and accessibility, companies can position themselves to succeed regardless of where the leads come from.
The advice for marketing teams is clear: double down on content quality and user experience rather than techniques aimed at gaming the system. In Stephen’s view, the future rewards brands that are customer-centric, solving real problems with straightforward, valuable solutions.
Key takeaway: As search and discovery tools evolve, traditional SEO tricks will matter less. Focus on crafting valuable, customer-centered content and improving the user experience. This approach aligns with the direction of AI-driven information sources, helping brands stay relevant and competitive.
Generate Content Ideas from your LinkedIn Profile with AI
Stephen shares a creative approach to generating LinkedIn content ideas by leveraging AI tools like Anthropic and ChatGPT. His strategy goes beyond simply prompting these models for generic suggestions; instead, he exports his entire LinkedIn profile data, uploads it into a knowledge base within his AI tool, and creates a customized content recommendation engine. By feeding the model this detailed data, he replicates his tone, topics, and industry focus, making the AI’s suggestions far more relevant and aligned with his brand.
For those unfamiliar with this process, LinkedIn offers the option to download your profile data. Once requested, it provides a comprehensive JSON file containing your connections, posts, comments, and overall activity history. Stephen finds this incredibly useful, as the exported data allows the AI to understand his typical interactions, audience, and interests. Feeding the AI this detailed dataset enriches its understanding of his ideal content topics, eliminating the need to continually “hack” the tool for appropriate ideas.
Stephen emphasizes how effective this process can be for maintaining a consistent and authentic tone across posts. With this setup, he can prompt the AI for content suggestions that genuinely align with his professional voice and the subjects his audience cares about. It’s not just about saving time; it’s about building a scalable system that ensures quality and relevance, keeping his messaging cohesive and impactful.
He notes that this approach could benefit anyone looking to elevate their social media presence. By feeding the AI personalized data, users can produce better-targeted posts and deeper connections with their audience, without manually crafting every post. It’s a powerful method for transforming an AI model into a true content partner.
Key takeaway: Downloading and feeding your LinkedIn profile data into an AI tool can personalize content recommendations, align with your tone, and maintain message consistency across posts. This method provides a scalable way to generate relevant LinkedIn content that resonates with your audience, streamlining your social media strategy.
Embracing AI and Building Use Cases
Stephen’s final advice to those diving into AI for marketing operations is refreshingly practical: just start. For those unsure where to begin, he encourages reaching out for a coffee chat, where he’s open to sharing his own use cases and lessons learned. His approach is hands-on, offering to provide templates he’s developed that can be easily integrated into other organizations—a helpful kickstart for anyone navigating AI without a solid direction yet.
He underscores that there’s no need to overcomplicate this journey. Starting with a few basic AI use cases that demonstrate immediate value can be a game-changer for your team and your career. Stephen emphasizes that by finding small wins with AI now, you’ll position yourself as a forward-thinking, indispensable team member, able to bring scalable efficiencies and insights. For Tray.ai customers, Stephen’s even prepared to share ready-to-go templates to make the AI setup process straightforward.
What’s at stake is more than just operational enhancement; it’s about future-proofing your role. Stephen believes that mastering these initial AI steps isn’t simply about tech novelty but about creating enduring value within your organization, particularly during a time when professionals feel increased job security concerns.
Getting started in AI, even if it’s a few small projects, builds both capability and confidence. As Stephen suggests, these initial projects aren’t just experiments—they’re essential for anyone looking to adapt, stay relevant, and keep an edge in an industry where AI will only continue to evolve.
Key takeaway: Start small with AI by implementing basic use cases that add value, build your expertise, and future-proof your role. Reaching out to others for guidance and sharing resources can make this journey collaborative and actionable from the start.
Episode Recap
AI for customer journeys might sound complex, but Stephen helps us out with a few practical use cases that he’s been tinkering with. BTW you don’t need a PhD in machine learning. His approach to journey mapping is the “three, five, seven” timing rule for websites that grabs attention in three seconds, engages in five, and guides users to act within seven. Think of it as a quick diagnostic: is the call-to-action visible right when users land? Is the path to that next step intuitive? If not, you’re losing people in a matter of seconds.
Stephen gets fired up about AI’s ability to turn personal outreach into a scalable strategy. By pulling in data from sources like LinkedIn and Crunchbase, Tray.ai’s AI agents create emails that feel human and individualized without the hours of manual effort. For any sales or marketing team feeling stretched, this kind of automation is a game changer. Now, instead of slogging through repetitive tasks, teams can focus on strategy, testing, and creative work.
And this isn’t just about automating emails. Stephen shows how AI can be applied to learn from past campaign successes, seamlessly integrating historical data to optimize new ones on the fly. Imagine an AI that observes a certain category of emails getting higher clicks and applies that insight across other touchpoints. Suddenly, AI isn’t just repeating tasks—it’s actively improving results, adjusting in real-time, and giving smaller teams the power to run like an enterprise marketing department.
Of course, Stephen’s realistic about the hurdles: connecting systems, handling sensitive data, and balancing compliance. But his advice here is refreshingly simple: start small. Appoint an “AI champion” who can experiment and share wins with the team. And if privacy is a concern, use tools like Tray.ai’s Merlin Guardian connector to anonymize data, allowing industries with strict regulations to move forward safely.
Listen to the full episode ⬇️ or Back to the top ⬆️
Follow Stephen👇
✌️
—
Intro music by Wowa via Unminus
Cover art created with Midjourney (check out how)
Apple •Spotify• Pocket Casts •Youtube •Overcast •RSS
Related tags
<< Previous episode
Next episode >>
All categories
- AI (99)
- career (63)
- customer data (60)
- email (64)
- guest episode (173)
- operations (127)
- people skills (34)
- productivity (10)
- seo (6)
See all episodes
Future-proofing the humans behind the tech
Apple •Pocket Casts•Google •Overcast •Spotify •Breaker •Castro •RSS