Marketing’s org chart has never been built to last. It keeps changing because each structure solves a specific problem of its moment—and then creates the next one as customers, channels and business portfolios shift. The current conversation is less about picking the perfect model and more about spotting when the old model has stopped matching the business.
From brands to channels to lifecycles
Brand management as we know it began with a 1931 P&G memo. Neil McElroy argued for giving each brand its own dedicated owner and team, treating it as a separate business. That logic still works for portfolios, but it creates a hidden cost: every brand starts wanting its own martech stack, analytics setup and email rules. Eventually finance asks why 15 brands are paying to solve the same problem 15 times.
The common fix was centralizing by channel. Channel centres of excellence did improve specialist knowledge, but specialists stop at the edge of their channel. When growth depends on a handoff across paid search, email and social, no single channel team owns the merge.
Lifecycle structures—splitting acquisition from retention—moved marketing closer to the customer journey. Yet many still sit inside channel walls: acquisition teams organize by acquisition channel, retention teams by CRM. The deeper question remains: who owns pricing logic, recommendations, loyalty and identity when multiple teams need the same capability?
Enter the domain team
Domain thinking borrows from software engineering, Domain-Driven Design and Conway’s Law: systems end up looking like the org chart, planned or not. Team Topologies gives it a useful shape—stream-aligned teams owning a capability end-to-end, platform teams supplying shared plumbing, and enabling teams helping adoption.
A domain is a business capability with its own logic, data and expertise: personalization, pricing, loyalty, search, identity and consent, or the content supply chain. A domain team does not run a brand P&L or a channel budget; it owns the models, rules, systems and roadmap that brand and channel teams depend on. Success is measured on aggregate revenue lift, conversion, retention and cost taken out by not rebuilding the same thing five times.
Who has already made the shift
Amazon embedded domain ownership early through single-threaded owners, two-pizza teams and defined service interfaces. ING restructured about 3,500 people in 2015 into roughly 350 squads and 13 tribes organized around domains such as payments, mortgages and daily banking. Mondelez built a shared digital core for AI and personalization so brands plug in rather than reinvent the wheel. Spotify’s tribes and squads became the most-copied model, but the company’s own engineers have since said it does not resemble how Spotify runs today—a reminder that a slide is not the same as an operating model.
When the domain model fits
Domain structures earn their keep only when a real portfolio leans on shared capabilities and there is a data and technology platform underneath. Skip that foundation and a domain team is just a silo with a new label.
- Multiple brands or products depend on the same underlying capabilities.
- Shared data, identity and tooling already exist or can be funded.
- Leadership accepts separating accountability for a result from control of the system that produces it.
- Meaningful duplication is visible and expensive enough to fix.
They fit less well with one dominant product or genuinely separate businesses that share little technology.
Why martech leaders should care now
AI makes this urgent. The model underneath an AI feature is close to a commodity and changes often; the real differentiator is the harness around it—data, business context and integration that makes output reliable and specific. That harness is exactly what a domain team owns. Leaving five brand or channel teams to build it separately means paying for the hardest part of AI adoption five times, badly, at the wrong moment.
The transition has a real cost: decision rights move away from leaders used to having them, and the enabling layer must be funded so downstream teams actually adopt what domain teams build. That coordination cost is unfamiliar, but for organizations with scale, it may be cheaper than the duplication it replaces.
Source: MarTech
