How to Automate Emails Based on Subscriber Data

Top U.S. sports retailer levels up with GenAI.  How they improved email personalization: Angela Jing & May Khine's infographic breaks down the opportunity, methodology, and results. Opportunity: Dick’s Sporting Goods (DSG) sends one to two daily emails to its subscribers. Previously, these emails were created manually with limited personalization. To enhance efficiency and personalization, DSG sought to combine Generative AI with its demographic, loyalty data, email templates, interactions, and transaction history. Here’s how they did it: 1\ Identify customer segments ↳ Cluster customers by purchase history to identify their content preferences. 2\ Generat email templates ↳ Create personalized email templates for customer segments using an LLM (DBRX) with Retrieval-Augmented Generation (RAG). 3\ Recommend templates for customers ↳ Rank LLM-generated templates and recommend the best for each customer. Results: Deliverables ↳ New LLM-based automated email generation pipeline. ↳ User-friendly web application. ↳ Clear prompt guidelines. Metrics ↳ 65% predicted increase in email relevance. ↳ 18% predicted increase in clicks. ↳ 29% estimated reduction in creation time. P.S. The graphic dives deeper on data scope and methodology. --- ♻️ Repost to help your network! 📌 Want to level up with Generative AI? 1. Just follow me Lewis Walker ➲ 2. Subscribe to my free newsletter. #generativeai #ai

How to supercharge your email newsletter with Clay: So that you can segment your audience. And personalize your newsletter messages. We set this up for our beehiiv, But this should work with Substack, Kit (formerly ConvertKit), or whatever newsletter tool you use. Step 1️⃣: Grab info on new sign-ups via your newsletter API. You can use Zapier or Make, but in our case we use n8n. Then funnel that data into Clay for enrichments. We also store in Airtable and HubSpot for safe-keeping. Step 2️⃣: Scrape person + company data. You'll want to use a formula to separate personal + professional emails. Since you'll have two separate data waterfalls to find LinkedIn profiles. We included LeadMagic, Snov.io, People Data Labs, Clearbit, MixRank, and Apollo.io in the waterfall. We then use Serper to find any that were missed out. Those LinkedIn profiles were used for: - LeadMagic person enrichment (via HTTP API) - and Clay "Enrich Company" for firmographics. Now for ~60%+ of our sign ups, we know who they are and what company they work for. Step 3️⃣: Use AI to enhance your data. In this workflow, I used AI to: - Confirm their full-time role. - Deduce their native language. - Normalize our UTM/source data. - Validate that LinkedIn URLs are correct. - Shorten company names to reference in emails. - Conduct thorough account research (with Perplexity). - Score the companies + people into Tier 1, 2, 3, or unqualified. - Categorize the industry of the company. - Deduce if the signup fits into one of our 6 main buying personas. Step 4️⃣: Send all this new data back to your newsletter platform. As well as your CRM (in our case HubSpot). There are two main ways we plan to leverage this data: One is to create relevant audience segments for certain broadcasts. Examples: - Writing an edition in another language. - Creating bottom-funnel messages for certain buying personas. - Triggering manual SDR outreach when Tier 1 subscribers engage. The other way is to use custom variables within the message. Examples: - Using first names ("Hi there" being the fallback) - Referencing their company name, job title, or industry. We manage all of this, without asking for any additional info upon sign-up. We're still just asking emails, but that's the power of Clay. If you want me to custom-build this for your newsletter, DM me "newsletter" and we can have a chat. P.S. Shoutout Monika Grycz for writing our weekly edition of "Sales Tech Lab" And to Timothy Eugene for helping set this up 🤝

Level up your email automation in HubSpot with this workflow loop ⚡️ → Behavior driven email that drives action AND loops in reps as needed Here’s exactly how it works step-by-step: 1. 👤 Create a contact based workflow ⚡️ Trigger Criteria: This could either be a change in a specific property value, a form submission, or membership of a specific list. 💡 (Optional) Re-enroll contacts - this will allow users to receive this sequence multiple times if they meet the criteria again. 2. Include any actions you would like all enrolled contacts to receive. → Eg. sending an automated email, updating marketing contact status (if you have prior consent), updating lifecycle stage etc. Next we have a series of behavior driven flows designed to increase goal conversion. → Eg. Converting to a trial, booking a demo, registering for an event etc. 3. 🌱 Branch the workflow based on whether the enrolled contact has completed our desired goal. → In this example we are using product usage traits from June to branch our workflow in HubSpot. 💡 These branching checks will repeat for each milestone on the path to our goal and include steps to guide users who have not reached the milestone. ✅ Success Branch - Milestone Achieved → When the milestone is met, follow the success branch to check whether or not the next milestone has been achieved. → A contact will follow this path until they either reach a milestone they have not completed or has achieved the goal state. ❌ Failure Branch - Milestone Not Achieved → For contacts who have not met the criteria we can use the subsequent workflow actions to encourage the behavior we want to see. 4. 📧 Send an automated email to encourage the goal behavior 5. ⏰ Delay and check again if the milestone has been met. 6. 🌱 If the 1st milestone has been met, use a “Go to action” step to redirect the workflow to check the next milestone. 💪 Sending automated emails may be effective butyou can also loop in your team for 1:1 touch points as needed. 💡 You may only want to add a 1:1 touchpoint for specific segments. Add branches as needed. → eg. Is a member of a “VIP” list or is associated to a company with > 200 employees → You can repeat this loop as many times as you need and for as many milestones as you have defined on the way to a contact goal. 🔦 The new Workflow Performance and Conversion Tracking beta makes it easy to analyze the effectiveness of each section of your workflow and optimize your strategy. → Access the performance insights go to View > Aggregate Performance from your workflow — P.S. I’m building a niche B2B media/education company for HubSpot admins starting with a weekly newsletter on Substack. You can subscribe to get step-by-step use case playbooks to implement in your own account each week. https://lnkd.in/en942kTf

lifecycle marketers are sitting on ai goldmines you have more data than any other marketer. you're barely using it. what most lifecycle teams do with ai: 📩 generate welcome email copy 🔔 write push notification variants 🖋️ create "personalized" subject lines what you should be doing: 📊 predictive scoring: ai analyzes user behavior to predict who's about to churn, upgrade, or go dormant before it happens ↪️ dynamic journey optimization: ai adjusts email timing, channel mix, and messaging based on individual engagement patterns 🕸️ behavioral trigger automation: ai identifies micro-moments that indicate intent and fires campaigns instantly 💭 cohort intelligence: ai finds patterns in your retention data you'd never spot manually - like users who upgrade after getting exactly 3 onboarding emails content fatigue detection: ai knows when someone's getting too many emails before they hit unsubscribe the workflow changes: instead of building static drip campaigns, you're building adaptive journeys that evolve. instead of batch-and-blast scheduling, you're sending at optimal individual moments. instead of guessing why people churned, you're preventing churn before it starts. the lifecycle marketers crushing it aren't just automating tasks. they're automating insights. while you're manually analyzing cohorts, they're getting predictive health scores. while you're a/b testing subject lines, they're testing entire journey architectures. your data advantage only matters if you use it. what's one lifecycle decision you're making with gut instead of data? if you want to know more about how you can automate your segmentation + strategy + execution, drop a comment and i'll reach out! Santhosh T Pavan M Gowda can help you set up Hooly!