• LinkedIn
• Twitter
• Instagram
• YouTube
• RSS Feed

Apple •Spotify• Pocket Casts •Youtube •Overcast •RSS

What’s up everyone, today we have the pleasure of sitting down with Anna Gelfand, Director and Head of Lifecycle & Retention Marketing at Miro.

Summary: Anna breaks down what codification actually does to the job of lifecycle marketing, why she built digital twins of 100 real customers to test campaigns without ever bothering a live user, and how the smartest lifecycle teams are refusing to be called the email team. She gets into the 20/80 flip between building and thinking, why she’d rather own responsibility than a channel, and the one belief that survived 3 ESP migrations and 7 years at Miro. Stick around for the part where she admits she kinda likes suffering, intellectually speaking.

In this Episode…

• Why Lifecycle Marketing As We Know It Has No Future
• How To Codify The Qualitative Side Of Lifecycle Marketing
• Testing Email Campaigns On Digital Twins Of Your Customers
• Shipping The Criteria Instead Of The Campaign
• Why Lifecycle Teams Should Stop Calling Themselves The Email Team
• Why Your Workflow Should Define Your Martech Tools
• What Lifecycle Marketers Will Do When Agents Run Campaigns
• The Skills Lifecycle Marketers Should Double Down On In The AI Era

About Anna

Anna Gelfand is the Director and Head of Lifecycle & Retention Marketing at Miro, based in Amsterdam, where she’s spent close to 7 years building and running lifecycle and retention for a global product-led company at real scale. She leads a team responsible for email, in-app, and the wider customer journey, and she’s known for treating lifecycle as the connective brain of the organization rather than a single channel.

Before Miro she worked in lifecycle and retention at Carrot Quest. She’s also comfortable holding a contrarian view worth arguing over: lifecycle marketing as the industry knows it doesn’t have a future, and its center of gravity is already moving toward strategy, content, and codified systems that agents run.

Why Lifecycle Marketing As We Know It Has No Future

Spend enough years in lifecycle marketing and the work starts to rhyme. You build the segments, ship the campaign, wait for the numbers, and a large share of the time the test comes back saying nothing at all. Next quarter you run a version of the same thing with a slightly different subject line. When the uplift arrives, it’s a fraction of a percent for months of effort.

Anna hit that wall at Miro, and she’s careful to point out it started before AI was disrupting everything left and right. She’s been there a long time, long enough to watch the product and the market reinvent themselves more than once. By that point you’ve already picked the obvious wins, then the not-so-obvious ones, and what’s left is weeks of work for a result that refuses to tell you anything.

“I would rather get a very negative result rather than an inconclusive one.”

That line sounds strange until you’ve lived it. A clearly negative result kills an idea and hands you a lesson. An inconclusive one burns the same 3 weeks and teaches you nothing, so you’re right back where you started with less time on the clock. Anna’s take on the plateau is that manual, human-level optimization simply can’t reach the relevancy customers deserve anymore. That’s especially true at Miro, where the product is an open canvas and users can build almost anything on it. You know your personas and your core use cases, but hand-tuning content one segment at a time won’t make an email feel genuinely built for the person opening it.

Here’s the part that keeps Anna in the craft anyway. Lifecycle is intimate, even when you’re sending to millions and thinking in the billions. The human aspect of the communication is the whole point. So the real question she’s chasing isn’t whether to throw technology at the problem, because everyone will. It’s what actually survives once you do. The plateau isn’t a Miro problem or a tooling problem. Any mature lifecycle program eventually runs out of manual optimization road, and the teams still treating inconclusive tests as a normal cost of doing business are the ones that get lapped first.

Key takeaway: Pull your last 2 quarters of lifecycle experiments and count how many returned a clear win or a clear loss. If most came back inconclusive, stop feeding new tests into the same tired playbook. Redirect that time toward a faster way to kill weak ideas before they eat a full test cycle.

Back to the top ⬆️

How To Codify The Qualitative Side Of Lifecycle Marketing

Codification is the word doing the rounds right now, borrowed from Luke Kline’s writing on the topic. The idea is simple enough: take the marketing knowledge locked in your head and your docs, make it accessible to AI agents, and let them operate on it directly. It works beautifully for the quantitative half of lifecycle. Propensity models, churn scores, who bought what and when, all of that is already data, so agents eat it easily. The harder question is the other half. How do you codify a gut feel, the sense of what makes a subject line bad, a nudge too pushy at the wrong stage, or a campaign a little too clickbaity?

Anna’s answer starts with a shrug and a strong opinion at the same time. Codification is inevitable, she says, because lifecycle was always the most technical, most data-informed corner of marketing. Expecting AI to leave that part alone would be strange. So the work becomes feeding agents the right tools and the right context.

“I’m a control freak with trust issues. It’s very hard for me to trust AI-generated copy or AI-generated content.”

That tension runs through everything she describes. Anna trusts data and segmentation when the warehouse and the models are in good shape. She does not yet trust a machine to write the message. Her example makes the stakes concrete. Miro is a B2B2C company with a wall of self-serve users on one side and large enterprises on the other. She’s always wanted lifecycle communications that feel like a partnership rather than a vendor broadcast, and doing that by hand only scales to the biggest accounts. Codification changes the math. Now an agent can pull a public customer’s stated goals from their own reports, cross them with Salesforce and Gong data, and paint the email in that customer’s colors with their use cases baked in. The account-based treatment she could only afford for a handful of logos becomes something she can run wide.

The catch she keeps circling is judgment. She won’t call it control, but she wants a clear moment where a human applies quality and qualitative feedback before anything ships. Part of that is buying back time to actually talk to users again, which lifecycle teams almost never do enough of. The other part is a bet on content. As AI floods every inbox with more messages than anyone can read, the scarce thing stops being the ability to produce and becomes the ability to stand out.

“Content is more important than ever because now you really need to stand out.”

The uncomfortable implication for most teams is that codifying the qualitative layer isn’t a copywriting problem, it’s a context problem. If your brand voice, your stage-by-stage rules, and your customer intelligence live only in senior people’s heads, an agent has nothing to operate on and will happily generate confident garbage at scale.

Key takeaway: Write down the judgment calls your team currently makes by instinct. Document what a bad subject line looks like, what’s too pushy at each stage, and the brand rules you’d never let slip. Feed that to your agents as context, then set one named human checkpoint where quality gets approved before anything sends.

Back to the top ⬆️

Testing Email Campaigns On Digital Twins Of Your Customers

Plenty of marketers have built a digital twin of themselves by now, a chatbot they argue with to pressure-test their own ideas. Anna’s team did something more useful. During a hackathon they built digital twins of their customers, a synthetic version of a slice of the user base, and started running experiments against those instead of shipping straight to live production. The hypothesis was blunt: why spend 3 weeks on a test that’s going to come back inconclusive when you could learn the same thing in an afternoon?

The value isn’t only speed. Anna found the twins force better hypotheses, because a lot of experiments fail for a boring reason. As she puts it, sometimes you run a bad experiment because the hypothesis behind it was bad. Testing against a synthetic segment builds the muscle of asking sharper questions before you commit real audience to them. She’s careful to say this doesn’t replace A/B testing. Strong hypotheses and significant changes still need real customers and real statistical validation. What the twins do is kill weak ideas at the starting line and surface learnings you could never get any other way.

“People always have a lot of opinions about emails, but only when they’re angry at them.”

Anyone who’s tried to run a customer interview about an email knows the feeling. Nobody wants to talk about your onboarding flow. They’ll rage about a bad send, but ask them to sit down and give thoughtful feedback on value and they’ve got better things to do. The twins sidestep that. You feed in what you already know, the product feedback, the public reviews, the behavioral footprint, and you can hypothesize how a given persona would react before you interrupt a single real person.

Phil brought up the scenario that makes this genuinely exciting, and a little eerie. The customers you most want feedback from are the ones who ghost you: the people who never open, or who unsubscribe and vanish. Legally you can’t email someone who just opted out to ask why they left. A synthetic version of that user carries no such restriction. You can watch the twin unsubscribe from a trigger and ask it what pushed them away. Anna framed the shift in careers terms that landed hard. Early in her career you watched real people use your product in a session-recording tool. Now you watch how synthetic customers respond and click.

She also used this to make a bigger point about where lifecycle is heading. Email is just a delivery mechanism, and so is an in-app notification. The impact lives in what happens after the message, which has traditionally sat with product teams. With the ability to build interactive experiences around the product, on the website, mimicking the product itself, lifecycle can own richer surfaces than a send. The twins become the feedback loop for those experiences too.

What Data Goes Into A Synthetic Customer

A digital twin is only as trustworthy as the context behind it, so the obvious question is what actually went into the build. These aren’t classic marketing personas. Anna’s team took a real sample of user data, who these people are, how they use the product, their roles, firmographics, usage patterns, and how they engaged with past campaigns, then generated look-alikes from it.

“We took 100 users’ data and we created 100 similar digital twins with the same behavior and characteristics, but who only can respond to us.”

That last clause is the whole trick. The twins carry the real behavior of real customers, but they exist to answer your questions instead of ignoring them. This is codification pointed at people rather than campaigns. You’re not just building richer segments, you’re building context that talks back. The context starts responding to you. That’s a different relationship with your customer data than a static dashboard, and it’s within reach of any team willing to sample honestly and resist the urge to make the twins say what they want to hear.

Key takeaway: Build a small synthetic panel before your next big test. Sample 100 real users across roles, firmographics, and usage patterns, generate look-alike twins from that behavioral data, and run your riskiest idea past them first. Use it to kill weak concepts fast, then reserve real A/B tests for the hypotheses that survive.

Back to the top ⬆️

Speaking of evolving your perspective on qualitative user feedback… are you still operating like it’s 2013 and you haven’t updated your chatbot? We recently started collaborating with the Docket team and what they are building with their Inbound Demand Agent is super cool. Check it out.

Shipping The Criteria Instead Of The Campaign

One clean way to describe the shift is that you stop shipping campaigns and start shipping the criteria that generate them. Anna likes the framing but pushes back on how tidy it sounds. Somebody still has to ship campaigns, and while the hope is to delegate a lot of that to agents, Miro isn’t there yet. Data gets created dynamically, segmentation gets built on the fly, agents draft content, templates lean on AI. The full end-to-end handoff, where you describe a campaign and an agent builds and runs it, hasn’t arrived.

What has changed is the ratio of building to thinking. When the operational grind of assembling campaigns shrinks, the time has to go somewhere, and Anna wants it going into judgment.

“I think the flip will be 20% of time is spent on building and 80% is spent on thinking, like defining the criteria, feeding the criteria, iterating on the criteria.”

Invert those numbers and you can see how most lifecycle teams run today, with the bulk of the week swallowed by the build. Flipping the ratio sounds like a productivity story, but Anna’s real point is that it’s an organizational one. If everyone suddenly has more time and more reach, the old boundaries about who owns which channel start to break down, and that raises a harder question about contribution versus ownership versus accountability.

“In the corporate world, we are over-indexing on the concept of ownership and under-indexing on the concept of responsibility.”

Her argument is that ownership has become a construct people hide behind, a way of claiming a channel instead of taking responsibility for the quality of what gets delivered on it. When agents can build and deploy across surfaces, the valuable thing to own becomes the quality standards, the core ideas, and the infrastructure that lets anyone contribute responsibly without stepping on each other, rather than any single channel. That’s a genuinely uncomfortable reorganization for a function built on turf.

What Lifecycle Looks Like As A Cross-Functional Team

Anna’s clearest example predates AI, which is exactly why she likes it. A couple of years ago Miro made a big product transformation, repositioning itself as an innovation workspace with multiple formats, and the team had to teach users and customers a genuinely new way of working. Updating the onboarding and running an awareness campaign wasn’t enough. People need many touchpoints to absorb a shift like that, and those touchpoints have to be different in kind: things to read, things to test, things to build inside the product.

So lifecycle became the group leading a cross-functional effort rather than a channel team firing sends. In prep for the conversation the team had called lifecycle the heart of the organization, and Anna offered a sharper version of it.

“Maybe not heart, rather brain, or brain with heart.”

One concrete output was a learning center built inside the app, where users could meet the new concepts, test templates on the spot, and see which integrations were available. That’s a much richer experience than an email announcing a launch. It also points to where educational and lifecycle teams blur together, with lifecycle owning how the material reaches customers in the right format at the right stage. The teams that will thrive are the ones comfortable leading work that doesn’t fit neatly under any single channel’s name.

Key takeaway: Take your next major launch and design it as a cross-functional effort with touchpoints across email, in-app, and web. Define the criteria and quality standards for each one so any contributor can build to them. Own the standards and the infrastructure rather than guarding a single channel.

Back to the top ⬆️

Why Lifecycle Teams Should Stop Calling Themselves The Email Team

AI adoption in lifecycle is wildly uneven. Some teams were doing more advanced work 5 or 6 years ago than plenty of teams manage today, back when a well-funded lifecycle group might have machine learning scientists on staff and a homegrown propensity model scoring every point in the CDP. Other teams are still sending weekly blasts. So the shift Anna’s describing doesn’t land on everyone from the same starting line, and she’s honest that it never has for her either.

Miro used to run its own machine learning model to predict best-fit audiences and likelihood to upgrade or churn. Then the product changed, and everything built on the old assumptions stopped being relevant. Suddenly you’re back to blast campaigns because you’re mid-transformation. For Anna this is just the shape of the work.

“Sometimes you actually need to go back to the level zero and start building everything from scratch.”

The career progression moves in fits and starts. You get advanced, the ground moves, you reset. With the disruptions coming, she expects more of those resets, not fewer. The foundations still matter, but she’s genuinely unsure whether the specific things her team built before should be the base for what comes next, because how people use a product like Miro is changing that fast.

Underneath the whole conversation is a constraint lifecycle teams put on themselves. They quietly accept being the email team. Not in the batch-and-blast sense, but in the sense that email and in-app are our channels, so that’s the box we work in. Anna rejects the box.

“The title of the team doesn’t define what you can and should build to improve user experience.”

Her point is that you decide who you are. Segmentation and multi-step journeys stay the core, because that’s how lifecycle people think about intent and signals. But the number of ways you can shape a customer’s experience should be getting much wider, and there are finally enough tools to step outside journey campaigns without asking permission. That’s why she keeps returning to the idea of a rename. She wants to call the team something other than lifecycle, maybe user growth, though that’s not quite it either. Naming isn’t cosmetic to her.

“Naming defines how you’re feeling within the space that you’re in.”

She’s still hunting for the tagline that gives the team permission to think bigger. The lesson underneath is worth taking seriously even if you never rename anything. The words a team uses about itself quietly set the ceiling on what it believes it’s allowed to build.

Key takeaway: Write down every type of work your team assumes is off-limits because it lives outside email and in-app. Pick one item on that list, a web experience or an interactive surface, and prototype it this quarter. Let the work redefine the team’s scope before you argue about what to call it.

Back to the top ⬆️

Why Your Workflow Should Define Your Martech Tools

Every vendor pitch in this category makes the same promise. Nothing in your stack gets replaced, you just connect a context layer over your existing data warehouse and light everything up from inside your real stack. It’s a good pitch, and it’s worth asking a warehouse-native practitioner whether it actually holds. Anna’s answer accepts the mechanics and reframes the whole way you should think about tools.

“Workflow defines the tool, not the tool defines the workflow.”

She’s earned the opinion the hard way. Over her time at Miro the team went through 3 different email service providers, a migration marathon she describes as leaving scars on her heart and her resume. The lesson from all that pain is that the tools were interchangeable. Every one of them did the job. The only question that matters is which workflow is most efficient for the business, and then you build the technology around that, refreshing or replacing pieces as the workflow changes.

Where it gets concrete is the creative process, something Anna has strong feelings about. Ask where you start building an email today and the honest answer is no longer the email CMS. The old flow of opening a builder, dragging blocks, and dropping in images is dissolving. She walked through how her own process is reshaping itself: freeform brainstorming on a Miro board because that’s how she thinks, then pulling that through MCP into Claude, then handing it to a CMS wired to her code repo, her visual guidelines, and her copy rules, which composes the actual email. The steps that used to define email production are getting rearranged around how a person actually works.

The rule that falls out of this is simple and a little ruthless.

“If the tools are not following your workflow, you’re replacing the tool.”

Connectivity is what makes that possible, which is why she’s genuinely excited about MCP. When tools expose their capabilities cleanly, you can wire together things that never connected before, and the workflow stops being hostage to whatever a single platform decided to build. The practical read for anyone evaluating a shiny new context layer is to start from your workflow and ask what it needs, not from the demo.

Staying Curious Without Chasing Every Shiny Object

There’s an obvious objection, and Phil raised it. AI is changing tools so fast that something you’ve never heard of today could reshape your work in 3 months. At what point does insisting the tool follow your workflow become an excuse to ignore what’s out there? Anna’s answer holds the line without getting rigid about it. The principle doesn’t change, but the emphasis does. You always have to be improving your workflow, because your current workflow being familiar doesn’t make it good.

“Your current workflow doesn’t mean that this is a good workflow.”

Habits are comfortable, and comfort is the trap. Anna forces herself to reassess by watching where non-experts struggle. Her team owns email for the whole company, so they periodically open the CMS to everyone, letting anyone build an email even though only the team can send. Watching outsiders fumble through the tool is what makes her question whether the workflow is really optimal. The second half of her answer is a straight endorsement of curiosity as a job requirement. You can’t afford to not be curious right now. Plenty of shiny objects are noise, but some carry real potential, so the move is to test constantly, watch how others work, and pull what improves your workflow. She notices it even in how her day starts. She used to open a Google Doc or a Miro board, and now the day begins in a completely different tool, still connected to her craft but reached through different steps. The workflow still defines the tool. The workflow itself just keeps evolving.

Key takeaway: Map your actual creative workflow end to end, from where you brainstorm to where the email gets composed, then judge every tool by how well it fits that flow. Open your core tools to non-experts once in a while and watch them struggle, because their friction is the fastest signal that your workflow needs work.

Back to the top ⬆️

What Lifecycle Marketers Will Do When Agents Run Campaigns

Fast forward a few years, assume codification has fully played out, and picture the lifecycle marketer who’s winning. What are they actually doing all day, and does the job even keep the name? Anna won’t pretend to know, and her honesty about the whiplash of this moment is refreshing.

“I have the days of, we will all be unemployed tomorrow, and then I have the days of, now we will be the most powerful function in marketing.”

That swing between doom and dominance is the actual emotional texture of working in marketing right now, and most people won’t admit to feeling both. Once she sets the mood, her prediction gets specific. The winning marketer spends far more time thinking and strategizing, and far more time genuinely knowing the customer. Lifecycle has always claimed to know the user well. She wants to take that to a deeper level, understanding both the customer and the product in real detail. In B2B especially, knowing your product inside and out becomes non-negotiable, since that’s what you’re driving adoption and conversion for.

The career shape changes too. The industry went from generalists to specialists, through the whole T-shaped debate, and Anna sees it swinging back toward generalists. A lot of lifecycle managers, she thinks, will become closer to content marketers, and many more will turn into something like a product manager or program manager for a broad slice of the business. With codification handling the technical execution and agents running the machinery underneath, the human job shifts up a level.

“You’re spending more time on connecting the dots, building the content, understanding the customer, and making marketing a little bit more full-bodied.”

That phrase, full-bodied, is a good one to sit with. For years lifecycle got thinner and more technical, all triggers and segments and deliverability. The version Anna’s betting on adds the flesh back: content, customer understanding, connective thinking across a business. The uncomfortable flip side she names herself is that this undermines some of the specialist value people spent careers building. The winning marketer of a few years from now looks less like a channel operator and more like a general manager of customer experience.

Key takeaway: Shift your development time from mastering tools toward 2 things agents can’t do for you: understanding your customer at depth and knowing your product inside out. Practice connecting insight across the business the way a product manager would, because that judgment is what survives once execution is codified away.

Back to the top ⬆️

The Skills Lifecycle Marketers Should Double Down On In The AI Era

Ask anyone what skill to bet on for the next few years and you usually get critical thinking, which is true and also a little unsatisfying. Anna knows this. She’s a self-described soft-skills person, so her instinct is to say critical thinking and step outside your job description, then she immediately calls her own answer a bit of a blah answer. The more interesting part comes when she works through which concrete skills survive.

Take the unglamorous stuff: QA, testing, deliverability, knowing how different email clients render. You might assume codification makes all of that irrelevant. Anna argues the opposite. As agents run more of the work at greater scale, the odds of small details slipping through the cracks go up, so someone has to make sure the agents are actually testing and checking all of it. The technical craft gets more important precisely because there’s more of it running unattended.

“We’re not in a situation where you need to know less. We’re in a situation where you need to know only more and more and more.”

Then she makes a claim that should reframe how lifecycle people see themselves. Marketing is becoming more important inside lifecycle marketing. For years the discipline got so technical that the actual marketing, the creative and the customer understanding, got outsourced to partners. You waited on a good brief from a content person or a product marketer. That dependency flips when you’re the one directing agents, because a generic instruction produces generic output.

“We are the ones giving agents the direction of what to do, and we need to be very opinionated.”

Alongside that sits the obvious new layer: AI skills. How to build, how to prompt, how to manage multiple agents, how to validate the quality of what they produce. Anna is refreshingly candid that she’s still learning this by facing her own bad results, and that there are probably skills coming she can’t even name yet. The honest posture is to keep trying and learning rather than pretending the map already exists. The marketers who compound fastest are the opinionated ones who can also steer an agent, because taste with no technical control stays stuck as a brief nobody executes.

The 2 Career Paths In Lifecycle Marketing

Phil sharpened this into advice he’d given a mentee, framing 2 paths opening up as codification lands. You can go deep on the craft of content, carving out a real brand personality in a world where everyone can batch and blast for free. Or you can move closer to the data and the technical side, becoming the person who operationalizes something like a synthetic customer database. His conclusion was that neither is wrong, and the right one depends on what actually pulls you out of bed in the morning. Critical thinking makes you better on either path.

Anna answered from the seat of someone leading a team, and her view is that a good team holds both paths at once. For a long time teams were built as interchangeable operators, one owning activation, another adoption, another monetization, all assumed to share the same skills and just aimed at different objectives. She sees the future differently, with people leaning into distinct strengths. Someone strong technically builds the architecture and the structures. Someone drawn to creative and visual work handles that. The differences are the point.

“If you’re not owning everything, you’re not owning anything.”

She’s firm that ownership has to run end to end, from the strategy down to building and QAing an individual email. Handing the build of your strategy to a separate marketing operations team is, in her view, a mistake, because you lose the feel of the work in your fingertips. She likes a team that stays deep in the woods with what it’s making. Historically lifecycle teams worked more outside than inside, everyone heads-down on their own piece with little real collaboration. The new formation she wants trades that isolation for internal partnership, mixing different calibers and compositions of talent so the differences enrich the output instead of fragmenting it.

Key takeaway: Pick the path that genuinely excites you, deep content craft or the technical and data side, then get good enough at directing agents to act on your own opinions. If you lead a team, staff for complementary strengths and keep ownership end to end, from strategy through the QA of a single email.

Back to the top ⬆️

How To Decide What Deserves Your Energy At Work

We close every conversation with the same question: how do you decide what deserves your energy at any given moment, and what keeps you aligned with what actually makes you happy? Anna, recording late on a Friday afternoon during a heat wave, answered first with a laugh that she’s not really good at this. Then she got honest.

She’s persistent and resilient by default, the kind of person who pushes to get things done even when she isn’t enjoying the road there. So her system is really a running check-in with herself about whether she’s actually happy, paired with a clear-eyed belief that you can’t feel happy in the work all the time.

“I can work on whatever if the people I’m working with are great.”

The people are the first non-negotiable. She likes partnering closely enough to share the best and the worst and the most annoying parts of the work with someone. The second set of things she knows will land her even after a brutal day: art and beauty, going home to hug her dog, watching her cats melt on the terrace in the heat. Small, concrete, reliable.

The last piece is the one that explains a lot about how she works. She needs to be intellectually stimulated, and she’s changed jobs before simply because she felt bored or not challenged enough.

“I kinda like suffering, to a large degree, from the intellectual standpoint.”

That one line ties the whole episode together. The person arguing that lifecycle should hand its grunt work to agents and spend the freed time thinking is the same person who gets restless without a hard problem in front of her. Wanting the intellectual challenge is a good compass for anyone deciding which parts of the job to protect as automation takes the rest.

Key takeaway: Name the 2 or 3 concrete things that reliably restore you after a hard day and protect them on the calendar the way you protect deadlines. Use a simple question, am I actually happy and challenged, as the check-in that tells you when it’s time to change something.

Back to the top ⬆️

Fiction Book Recommendations From A Marketing Leader

We recently started a sci-fi fiction book club on the show, so we asked Anna for a recommendation, guessing she leans more toward fiction than nonfiction. She confirmed it with a distinction that’s genuinely her own. When she tried to name a favorite nonfiction book, she couldn’t, because she doesn’t think a nonfiction book can be a favorite at all.

“I do not allow myself to feel feelings about nonfiction. I think that’s not the purpose of them.”

A nonfiction book can be interesting and teach her a lot, but it doesn’t get to move her. Her actual recommendations, both worth adding to a list:

• A Little Life by Hanya Yanagihara, a piece of literary fiction that left a deep mark on her and that she says she deeply loves. It carries almost a million ratings on Goodreads at 4.27 stars.
• The Year Before The Storm, a book about small human moments among artists and politicians in 1913, just before the First World War. She loves how you can feel the storm coming through ordinary day-to-day stories. Goodreads files it under history nonfiction, which she cheerfully admits bends her own rule.

What she values in that second book is the thing she says nonfiction usually can’t deliver: big, deep feeling drawn out of very small, personal details. It’s a fitting note to end on from someone who spends her days trying to make communication at massive scale feel personal and human. The best recommendations tell you how a person thinks, and Anna’s say she’s chasing emotional truth in her reading the same way she chases customer intimacy in her work.

Key takeaway: Add A Little Life and The Year Before The Storm to your reading list if you want fiction that stays with you. Borrow Anna’s filter too: let the books that move you be the ones you return to, and read nonfiction to learn rather than to feel.

Back to the top ⬆️

Episode Recap

Anna Gelfand makes one central argument across this whole conversation: lifecycle marketing as the industry practices it has no future, because the campaign-by-campaign, ops-heavy version of the job is being codified away. Her argument comes straight from running the same optimization loop dozens of times and watching months of work return a fraction of a percent, or nothing conclusive at all. Once you accept that manual, human-level optimization has hit its ceiling, the interesting question becomes what lifecycle should actually be centered around instead.

The tactical thread running through the episode is a steady move from execution to judgment. Codification hands the mechanical work to agents, which is why Anna predicts a flip from 80% building to 80% thinking. Her digital twin project is the most concrete version of this: 100 synthetic customers, generated from real behavioral data, that let her team kill weak ideas in an afternoon instead of burning 3 weeks on an inconclusive A/B test. The same logic pushes her to reject the email team label, to insist that workflow defines the tool rather than the reverse, and to argue that the valuable thing to own is quality standards and responsibility across every surface.

For practitioners, the implication is a real change in what the job rewards. The skills that survive are deep customer understanding, deep product knowledge, opinionated creative direction, and enough technical fluency to steer agents and validate their output. Anna expects many lifecycle managers to look more like product managers or content leads, running broad slices of the customer experience while codified systems handle delivery. She’s equally clear that the unglamorous craft of QA, testing, and deliverability gets more important at scale, not less, because more automated work means more places for small details to break.

What makes the conversation honest is that Anna doesn’t resolve the tension. She calls herself a control freak with trust issues who struggles to trust AI-generated content, and she swings between believing the function will be gutted and believing it will become the most powerful team in marketing. She hasn’t fully handed campaign building to agents, she’s still hunting for a better name for the discipline, and she admits she’s learning the AI craft by facing her own bad results. That mix of strong conviction and open uncertainty is exactly what makes her take worth taking seriously.

Connect with Anna Gelfand on LinkedIn to follow her thinking on the future of lifecycle and retention marketing.

Full episode ⬇️ or Back to the top ⬆️

✌️

—
Intro music by Wowa via Unminus
Cover art created with Midjourney (check out how)

Apple •Spotify• Pocket Casts •Youtube •Overcast •RSS

Related tags

See all episodes

Future-proofing the humans behind the tech

Apple •Pocket Casts•Google •Overcast •Spotify •Breaker •Castro •RSS

• LinkedIn
• Twitter
• Instagram
• YouTube
• RSS Feed