I see personalisation shifting from “who you are” to “what you’re trying to do right now”. Less about static traits, more about live intent. So instead of a broad “new customers” campaign, the email engine reacts to signals like “browsed cancellation page”, “used feature X for the first time”, or “contacted support about pricing”.
The tech that excites me is mostly behind the scenes. First, predictive models that score each contact on things like churn risk or likely LTV. In practice, that means the system decides whether to send a retention offer, an upsell, or nothing at all, based on profit, not just engagement. It forces teams to judge email by revenue and retention, not opens.
Second, modular content. One campaign, many versions. The email is built from blocks that change per person: problem angle, product shown, proof, offer. A SaaS trial user who’s stuck might see a “get set up” path, while a power user sees an annual upgrade offer, all from the same send.
Third, privacy-safe data use. More focus on first-party data (what people do in your product, on your site, in support logs) and less on third-party tracking. I think we’ll see more on-device or platform-side decisioning, where personal data doesn’t have to move around as much.
All this means strategy matters more than tools. You need clear rules like: if someone’s high value and looks likely to leave, what’s the one offer we’re willing to give? If they’re price sensitive, which discount and how often? The AI can test and optimise that, but it can’t set the business logic. The winners will be the brands that combine that logic with restraint, so emails feel helpful, not invasive.
