Why AI is creating more work in email and lifecycle marketing, not less

For the past few years, the conventional wisdom around AI has been pretty straightforward: AI makes work faster, so companies will need fewer people to do that work. In lifecycle marketing, we think something different is happening.

AI is making it dramatically easier to write copy, generate creative concepts, build segments, analyze performance, personalize content, and even produce emails themselves. But instead of teams saying, “Great, now we can do the same amount of work with fewer resources,” many are saying something closer to: “We can do so much more now. Let’s step on the gas.”

There’s actually an economic concept for this.

The Jevons paradox

In the 1800s, economist William Stanley Jevons noticed something strange happening with coal. Steam engines were becoming much more efficient, meaning they could produce the same amount of power using less coal. The obvious prediction was that Britain would therefore use less coal.

Instead, coal consumption exploded.

Making steam power cheaper and more efficient made it useful for far more things. More factories adopted steam engines, new industries found applications for them, and cheaper production created even more demand. Each individual task required less coal, but society found so many more reasons to use steam power that total coal consumption went up.

That’s the basic idea behind the Jevons paradox: when something becomes dramatically more efficient to use, we sometimes end up using more of it, not less.

We think something similar is happening in lifecycle marketing.

AI is lowering the cost of making a great email

Not long ago, creating a sophisticated lifecycle campaign could require a surprising amount of work. Someone had to come up with the concept and write the copy. A designer had to create the email. Someone had to build it in the ESP. Data and engineering might need to get involved for personalization or segmentation. Then came QA, approvals, testing, deployment, and analysis.

Multiply that across dozens of campaigns, automations, customer segments, products, markets, and languages, and teams quickly hit a ceiling. There were always more ideas than resources, so marketers compromised. Maybe everyone got the same welcome series. Maybe there was one abandoned cart flow instead of different journeys based on product category or customer behavior. Maybe a campaign went to five million people with the same creative because producing ten meaningful versions would take too long. Plenty of good ideas simply never made it onto the roadmap because the team didn’t have the bandwidth.

AI is starting to change that equation. Writing variations takes far less time. Creating personalized copy for different audiences is easier. Campaign analysis is faster. Creative production and email development are becoming semi-automated. Data that once required an analyst to interpret can increasingly be explored directly by marketers.

The cost, in both time and effort, of producing sophisticated lifecycle marketing is falling quickly. And that doesn’t mean fewer emails. It means a lot more of them.

Lifecycle teams are stepping on the gas

This is what we’re starting to see with lifecycle teams. When people realize they can create better work faster, their ambition expands.

A team that previously struggled to get four campaigns out the door might start asking why they can’t do twelve. Then the questions get more interesting. Why does everyone receive the same email? Why don’t new customers get a different experience from longtime customers? Why aren’t we personalizing based on browsing behavior? Why don’t high-value customers have their own lifecycle? Why aren’t we building different journeys around different products? Why are we testing one creative concept instead of five?

The limiting factor in lifecycle marketing has rarely been a lack of ideas. It has been the cost of executing them. AI is attacking that constraint, and when the constraint starts to disappear, the backlog of things teams have always wished they could do suddenly becomes possible.

One email becomes ten

This is where we think the biggest change will happen.

The promise of true one-to-one marketing has existed for decades, but the operational reality has always been difficult. Consider a retailer sending a campaign around a new product launch. Today, the practical approach might be one beautifully designed email sent to most of the list, perhaps with a few segments or dynamic content blocks around the edges.

But there are potentially hundreds of meaningful variations. A new customer might need more education, while a loyal customer might respond better to exclusivity. Someone who browsed the category yesterday should probably see something different from someone who hasn’t purchased in two years. Geography, purchase history, price sensitivity, loyalty, browsing behavior, and dozens of other signals could all shape what someone receives.

Historically, creating all of those experiences would have been operationally ridiculous. Increasingly, it won’t be.

AI makes it possible to move from creating one campaign for one large audience toward creating many versions for many smaller audiences, without multiplying the workload at the same rate. That is where the Jevons paradox becomes especially interesting for our industry. If producing an email becomes 10x easier, companies probably won’t just produce the same emails with one-tenth of the effort. They’ll find many more things worth producing.

There is probably some AI mania right now

Some of what we’re seeing today is undoubtedly hype. Every major new technology goes through a period where companies try to apply it to absolutely everything, and AI is no different.

There will be AI-generated campaigns that never needed to exist, tests that generate more noise than insight, and personalization that is technically impressive but completely meaningless to the customer. Some teams will confuse the ability to produce more with a reason to produce more. Eventually, some of that will normalize.

But we don’t think the industry simply returns to its old level of output once the excitement fades. The underlying economics have changed. If a lifecycle team can build campaigns faster, analyze results faster, create more variations, personalize more deeply, and automate previously manual production work, the logical outcome isn’t that lifecycle marketing becomes less important. It becomes more ambitious.

The future is more lifecycle marketing, not less

The nature of the work will change. Teams will spend less time resizing images, rewriting minor copy variations, manually assembling templates, pulling repetitive reports, and performing other mechanical tasks. They’ll have more capacity for customer journeys, segments, experiments, behavioral triggers, creative variations, personalization, and coordination across channels.

In other words, the unit of work starts to shift. Instead of asking, “How do we build this email?”, teams can spend more time asking, “What is the best possible experience we can create for this customer?”

That is a much bigger question, and there are nearly infinite answers to it.

We’re probably in a period of AI mania right now, and not every experiment will survive. But long term, we think email and lifecycle marketing become significantly more sophisticated and more personalized. Companies will finally have the time and resources to execute ideas that have been sitting on roadmaps for years, or that previously would have been too expensive and complicated to even consider.

The companies that benefit most from AI won’t simply use it to do yesterday’s lifecycle marketing more cheaply. They’ll use it to do lifecycle marketing they never had the time, people, or resources to do before.

And that’s the interesting lesson from Jevons: when you make something dramatically easier to do, people often don’t do less of it. They find more reasons to do it.