The way most businesses approach D2C follows a pattern that has been borrowed from other types of digital projects. They hire a consultancy for the strategy. They brief a digital agency to build the platform and run the marketing. They bring in a systems integrator for the ERP connection. Each partner is competent. Each delivers their piece. And the D2C channel still does not work properly, because nobody owns the full picture.
This is not a criticism of any individual partner. It is a structural problem with the way the industry has organised itself around categories that do not match how D2C actually works.
The consulting industry, the agency world, and the systems integration market all evolved to serve specific needs. Consultancies diagnose, strategise, and advise. Agencies build digital products, create content, and run campaigns. Systems integrators connect enterprise platforms. Each of these services is valuable. The problem is that D2C requires all of them simultaneously, and it requires them to be designed as one system rather than assembled from separate workstreams.
When a business splits D2C across these categories, each partner works within their scope. The consultancy delivers a strategy that assumes someone else will handle the operational detail. The agency builds a platform based on a brief that assumes the operational questions have already been answered. The systems integrator connects the ERP based on a specification that may not account for the edge cases that only become visible when real consumer orders start flowing. The handoffs between partners are where the gaps appear, and the gaps are where D2C projects fail.
The brief itself is often the first casualty. An agency works from a brief: the client defines what needs to be built, the agency scopes it, prices it, and delivers it. That model works brilliantly when the brief is complete. The problem in D2C is that a complete brief requires someone to have already answered the operational questions: where the stock is, how orders get fulfilled, how the systems connect, what happens when something goes wrong, how the new channel sits alongside existing trade relationships. If nobody has answered those questions (and frequently nobody has), the brief is incomplete, and the agency builds exactly what it was asked to build on top of an operation that cannot support it.
There is a natural bias in the traditional model towards the consumer-facing side of D2C. The website, the brand, the content, the campaigns. These are visible, measurable, and familiar. Most agencies are built around these skills, and they are good at them.
But a D2C channel that looks beautiful and converts well is still a problem if the order cannot be fulfilled properly. Fulfilment is not a frontend discipline. It involves warehouse operations, ERP data flows, logistics networks, returns management, and customer service design. These are back-office capabilities that fall outside the scope of a traditional agency brief. Not because the agency is negligent, but because the model was never designed to include them.
The result is a D2C build that is strong on the consumer-facing side and weak on the operational side. The website works. The marketing is running. But the first 500 orders expose integration gaps, fulfilment errors, and customer service failures that nobody was briefed to address. The client is frustrated. The agency is confused, because they delivered what was asked. Both are right.
The fundamental issue is that D2C does not divide neatly into ‘strategy’ and ‘build,’ or into ‘frontend’ and ‘back office.’ It is one system. The platform choice depends on the ERP integration requirements. The integration design depends on the fulfilment workflow. The fulfillment workflow depends on the warehouse capability. The customer service model depends on all of it. Splitting these decisions across separate partners, working to separate briefs, on separate timelines, is how you end up with a D2C channel that was technically built correctly by everyone involved but does not function as a whole.
The businesses we have seen get D2C right are the ones that refuse to split it. They start from the operations and work outward, rather than starting from the website and working inward. They map the fulfillment, the integration, the customer service, and the returns process before anyone chooses a platform. The platform is then built on top of a functioning operation, rather than the operation being improvised around a platform that was chosen too early.
This approach requires a partner that does not fit neatly into the traditional categories. Not a pure consultancy, because the work does not stop at the strategy. Not a pure agency, because the work extends well beyond the platform and the campaigns. Not a pure systems integrator, because the work includes the warehouse, the customer service design, and the commercial strategy that sits around the technology.
A D2C project needs someone who can sit in the strategy meeting and the warehouse on the same day. Someone who understands why the ERP cannot process a split consumer order, and also understands why the homepage conversion rate is dropping. Someone who can design the data architecture and write the product descriptions. Someone who can brief the paid media and redesign the returns workflow.
That sounds like an unreasonable ask of any single partner. And if you are looking for a 500-person agency that happens to also have warehouse consultants on staff, it probably is. But D2C does not need scale. It needs breadth. A small team with senior people who have worked across strategy, technology, operations, and marketing can cover the full scope of a D2C project in a way that a large agency with deep but narrow expertise cannot.
The operating model matters here. Most agencies are built around repeatable digital deliverables: design, development, content, media. Their margins depend on predictable scoping and efficient delivery. Operational consulting is slower, less repeatable, and harder to scope in advance. It requires people with different skills sitting in different meetings: warehouse walkthroughs, ERP workshops, customer service design sessions, commercial planning with the sales team. That does not fit neatly into a sprint-based delivery model. It requires a different kind of business.
The industry has organised itself into consultancies, agencies, and integrators because those categories serve most digital projects well enough. D2C is the exception. It crosses all three categories simultaneously, and it adds operational complexity that none of them were designed to handle.
The businesses that struggle with D2C are the ones that try to assemble the capability from separate partners and hope the handoffs work. The businesses that get it right are the ones that find a partner who was built for the gap: someone who can write the strategy, build the platform, design the integration, sort out the warehouse, train the team, and stay long enough to see whether it actually works.
The traditional categories are not wrong. They just do not describe what D2C requires. And until the business finds a partner that operates across them, the gap between the strategy and the warehouse will keep producing the same expensive disappointments.
