Humans of Martech

So pumped about this lineup of upcoming episodes on Humans of Martech 👀 🇨🇦 Simon Heaton -- Director of Growth, Marketing at Buffer and former Growth Leader at Shopify 🦋 Jacqueline Freedman 🦋 -- CEO and Founder at Monarch Advisory Partners and former Martech leader at Grammarly and WeWork 🦅 Pranav Piyush -- Co-Founder and CEO at Paramark.com, Reforge Instructor and former marketing leader at BILL, Pilot.com, Adobe, Dropbox, Padlet, PayPal 📹 Benoit Leggieri -- Head of Growth at Livestorm and former Growth leader at HUB Institute, Paris’ top think tank ☕ Liam Moroney -- CEO and Co-Founder of Storybook and contributing writer at MarTech ❄️ Erin Foxworthy -- Industry Lead, Advertisers & Agencies at Snowflake and former Advertising leader at Horizon Media and Microsoft Advertising 📊 Ron Jacobson -- Co-founder and CEO of Rockerbox 🧙♂️ Jared DeLuca -- Director of Operations at Appcues and former IoT Automation and Connected Panel Ops at Keurig Dr Pepper Inc. Don't miss any of these awesome conversations but subscribing at humansofmartech(dot)com 📧

Marketing and technical roles often seem to exist in separate universes. Data engineers and data scientists gravitate to product and UX and avoid marketing and sales at all costs. But this isn't the case for everyone. Sundar Swaminathan's experience building technical teams at Uber that focus on marketing problems shows us that things like curiosity and customer obsession can turn 'technical folks who work in marketing' into 'marketing thinkers who are technical'. During his time at Uber, Sundar was dedicated to the company's success and it motivated him to take the time to understand the inner workings of marketing’s role in achieving business objectives. He would take an Uber home and chat with drivers about their experiences and gathered first hand knowledge that he could bring back to his business teams. His curiosity led to meaningful improvements in Uber’s driver retention strategy. After picking up on new bits of information from his chats with drivers, like learning that new drivers struggled with their first trips during busy Friday nights, he analyzed the data and discovered significantly lower retention rates for drivers who started during those peak hours. So he tested simple changes, recommending Saturday afternoon starts for new drivers instead of Friday nights. Wins that don't happen without caring about customers. If you're trying to find technical talent who embrace marketing, find folks who are driven by curiosity and customer obsession. The best data scientists and data engineers can move around from role to role on different teams because they’re motivated by understanding and solving customer problems, even if that means being 'in marketing'. Full episode with Sundar > https://lnkd.in/gdkPD23X

Sundar Swaminathan spent 5 years at Uber building Data Science teams focused on Brand, Performance, and Lifecycle Marketing. He hired and led a team of 15 in charge of measuring over $1Bn in Marketing spend and delivered over $150M of incremental revenue. Aside from the Uber experience, here’s what makes Sundar different than any other data science expert I’ve chatted with: He doesn’t think he’s a technical specialist who happens to work in marketing. He’s a customer obsessed marketer who happens to have strong data science capabilities. Nowadays, Sundar is a marketing and data science advisor, he’s also working on an upcoming podcast and has a newsletter where he shares frameworks, how-to guides to help B2C marketers. In our episode on Humans of Martech, he shared stories from Uber and some of the cool things his team built: • How Uber saved $26M in yearly spend by cutting 1 channel • How Uber used propensity matching for ROI analysis • How Uber used LTV prediction methods • How Uber proved brand impact with geo tests But it’s not all learnings from huge teams and resources, Sundar has plenty of love and insights for startups: • How to run experiments with tiny sample sizes • How to explain unmeasurable results to execs • How to get buy-in to build a marketing data science team • How to embrace measurement despite unfair ROI pressure Listen now 👇 YouTube: https://lnkd.in/gnrK55NX Spotify: https://lnkd.in/g7H6dWKd Apple: https://lnkd.in/gV_VmJKP Summary: https://lnkd.in/gxAxJZ2k -- Big thanks to our sponsors for supporting the show: 🦩 Census — Universal data layer that unifies & cleans data from all your sources and makes it available for any app and AI agent to use. 🎨 Knak — No-code email and landing page creator to build on-brand assets with an editor that anyone can use. 📧 Customer.io — Marketing automation platform to build intricate, multi-step customer journeys across all channels. 🦸 RevenueHero — B2B scheduling and routing product to instantly connect prospects with the right sales reps to drive qualified meetings. -- 💡My top takeaways: You don’t need a ton of users to pull off incrementality tests. Focus on high-impact changes through controlled pre-post tests. Isolate variables but target big minimum detectable effects. Go for those big swings. You can measure the ROI of a campaign without needing control groups using causal inference and propensity matching. This method mimics A/B testing by comparing similar groups to estimate causal effects. Prove the impact of your brand campaign with a 3-6 month geo test with consistent presence. Establish clear metric movement, implement proper control groups, and maintain long-term commitment to brand building activities.

📻 Tune in now! In this recent episode of the Humans of Martech podcast: ☕️Stephen S., Tray Director of Automation Solutions, discusses the transformative role of AI in customer journeys and messaging. Check out some key takeaways 💡 ⬇️ And listen or read for more insights: bit.ly/3VrvZvr

The 'best-in-breed' martech stack as a concept is dead. When Austin Hay was interviewed on Lenny's podcast a few years ago he had an answer when he was asked what was the 'golden' martech stack. But today, things are a lot different. Building a warehouse-first stack is often the recommended approach, but it's not the perfect fit for every situation. There isn't a on-size-fits-all. Stacks change like some of your awkward teenage fashion choices, usually by accident and often because of the ghosts of marketers past. IYKYK. We've all seen companies running on mysterious and random stacks. The ghost of a long departed founding team member who grabbed a tool in year one, the company starts growing, migrating to a more robust solution just wasn't in the cards and suddenly that random tool choice became the backbone of your entire marketing operations. The idea of a “best-in-class” stack keeps changing, especially now that AI has crashed the party. Austin takes us back to 2017, when life was simpler and clear patterns existed for different types of businesses. But building a stack today is like playing a jazz concert, there's a lot of key principles to follow, but the art is in the improvisation. For companies that haven't 'made it' yet (which is most of us)... just ignore the entire concept of the PeRfEcT stack. Focus on the stack that keeps the lights on. And that looks different for a lot of folks. Full episode with Austin: https://lnkd.in/gxyDBzzZ

We build our emails in [insert MAP] — what value would a campaign-creation platform provide? Here’s a few reasons why companies come inbound: → They want to future-proof themselves. What if Adobe completely changes its editor or Salesforce halts support on the Account Engagement (Pardot) Email Builder? (NOTE: this has happened). What if they change MAPs one day? What if they go through a re-brand or go through another acquisition? Each of these changes obviously has HUGE ripple effects operationally. → Their overall process is full of bloat. Feedback loops, spreadsheets, no central place to collaborate — a frakenstack of systems and tools that falls on campaign ops to manage. → They’re looking for a single source of truth for asset creation. For example, if they tweak their email design system … they don’t want their team going to a bunch of different places to update a template. → Way too over-reliant on agencies. They cut out unnecessary workflows and save BIG in savings. → Localization. It might take their team 2-5+ days to localize an asset. → They’re interested in putting in place a more ‘self-serve’ way of asset creation (marketers create VS everything going through central MOPs)…but not sure how to make it happen while enforcing brand standards. → They audited their asset creation journey and realized internal collaboration is a hot mess. People are sending test emails / taking screenshots and slacking them to people for feedback … hopping between PM tool > spreadsheet > slack. Huge waste of time. + sooo many more.

Personalization is where Product Marketing and Martech teams come together 🤝 Sarah shares in episode 152 what personalization means for her Growth team at Prefect. - PMMs drive the messaging frameworks for each persona based on use case maps and help figure out how can we be helpful to those folks - Martech folks drive the operationalization of those frameworks. They figure out the tooling to identify those signals, properly segment on those personas and then push all of that into the actual messaging channels Full episode with Sarah 🎥 > https://lnkd.in/g8BSzbR9

Let's all pour one out for the button pusher in Spotify's lifecycle department...

We don't always see eye to eye and they rarely have bandwidth for marketing, but I have a ton of respect for data engineers... even more for the ones who turn to the dark side and become marketers 👀 Sarah Krasnik Bedell spent the first 8 years of her career as a data scientist and data engineer, but one day she led the technical implementation of a marketing automation platform and it changed her life forever. During COVID she started writing about data engineering for marketing and growth use cases. She started attracting analytics and GTM advisory roles and eventually started getting paid to write. She wrote one of the most in depth, technical but approachable articles on marketing attribution that I've ever read. Today Sarah is Director of Growth Marketing at Prefect -- a workflow orchestration tool for data and ML engineers. On Humans of Martech, Darrell and I had the pleasure of hearing her unconventional path from data to marketing. In the episode we also cover: - Why first principles should always beat best practices - Why systems thinking = marketing survival - Why marketing teams need embedded analysts - How to use batch processing and webhooks - How to reconcile web analytics discrepencies - And so much more! Listen now 👇 YouTube: https://lnkd.in/gv-yNp8W Spotify: https://lnkd.in/giWsmzP2 Apple: https://lnkd.in/grPuiiEd Summary: https://lnkd.in/gWM7vr7C -- Big thanks to our sponsors for supporting the show: 🦸 RevenueHero — B2B scheduling and routing product to instantly connect prospects with the right sales reps to drive qualified meetings. 📧 Customer.io — Marketing automation platform to build intricate, multi-step customer journeys across all channels. 🎨 Knak — No-code email and landing page creator to build on-brand assets with an editor that anyone can use. 🦩 Census — Universal data layer that unifies & cleans data from all your sources and makes it available for any app and AI agent to use. -- 💡My favorite takeaways: 🗺️  Develop systems thinking by mapping data flows, understanding transformation points, and identifying bottlenecks. This will help you communicate effectively with technical teams and make better decisions about martech. 🤹 Embed analysts directly within marketing teams rather than isolating them in a central data team. Focus on answering specific business questions rather than conducting unfocused data expeditions. ⚡ Blend different data architectures to complement your stack. Determine what sources need real-time vs batch updates. Webhooks for speed and warehouse-based processing for comprehensive integrations.