Why the Old Editor breaks
A template produces a document; a living email is an application.

Figure 1 — The template editor produces a document: fixed, identical, inert. The factory produces an application: live, per-recipient, learning.
The cleanest way to see the problem is to name the category. A template editor produces a document — a fixed artefact, complete at the moment of saving. A living email is an application — a small program that runs when opened, responds to the recipient, talks to a data source, and records what happened. These are different kinds of thing, made by different kinds of tool. This is why the instinct to ‘add AMP support’ to an existing editor never quite works. You cannot bolt application behaviour onto a document tool and get an application tool; you get a document tool with a confusing new tab that produces broken applications. The right mental model is not ‘next-generation template builder’. It is an experience compiler: the marketer states what the email should do, and the tool assembles the experience that does it.
Five things a living email needs that a template cannot give.
Be specific about what is missing. First, interactivity — the blocks must do something when tapped, which means each block carries logic, not just appearance. Second, per-recipient composition — the email is a template-of-one, filled differently for each person. Third, live data binding — a balance, a delivery status, a price, fetched at the moment of opening rather than baked in at send. Fourth, memory write-back — every interaction routed back to the brand’s record of the customer. Fifth, dual rendering — an AMP version and an HTML fallback, kept in lockstep, from a single definition. Each is a discipline a template editor has no concept of. The missing capability is not in the marketer. It is in the instrument.
The trap: asking the marketer to become an engineer.
Faced with this, the tempting answer is the wrong one: give the marketer the raw materials and let them assemble the application by hand. Hand-author the AMP. Maintain the HTML fallback in parallel. Wire the data calls. Set up the memory routing. Test across a dozen clients that each render AMP slightly differently. This does not scale, for a simple reason. The people who make marketing emails are marketers, and they will not become engineers to send a newsletter — nor should they. A handful of sophisticated brands will hire the talent and produce a few beautiful living emails a quarter. That is not a channel; it is a craft project. For living emails to be the default, the engineering has to disappear from the marketer’s job entirely. The marketer’s job is to know the customer and the message. The tool’s job is everything else.