From the first conversation with a new client, we are already planning our exit. That sounds counterintuitive for a business that charges for its time, and I will be honest: it is harder to sell than a retainer model. But it is the only model that actually works for the client, and over 25 years I have become convinced it is the only one that works for us either.
The goal of every engagement at Voodoo is to make ourselves unnecessary. Not eventually. Not as a nice aspiration on the About page. As a practical, measurable outcome that we design for from week one.
I have spent a long career watching businesses become dependent on their external partners, and it is rarely because the partner is doing a bad job. Usually it is because the partner is doing a good job, but doing it in a way that keeps the knowledge and the capability outside the building.
The agency runs the campaigns, but nobody internally understands the strategy behind them. The systems integrator maintains the connection between the ERP and the ecommerce platform, but nobody internally knows how it works. The consultancy designed the channel strategy, but the rationale behind the decisions was never transferred to the team who have to live with them.
Each of these relationships is individually reasonable. The partner is competent, the client is getting results, and everyone is comfortable. The problem becomes visible when the partner leaves, or when the business needs to adapt quickly and cannot because the knowledge sits in someone else’s building. A client who still needs us for every decision six months into an engagement is a client whose D2C channel is fragile. The knowledge lives with us, not with them. That is not a success.
Stepping back is not the same as walking away. Walking away is what happens when a consultancy delivers a strategy deck and moves on. The client has a document but not the capability to act on it.
Stepping back is something specific. It means that by the end of an engagement, the internal team can run the D2C channel independently. They understand the systems, the workflows, the reporting, and the decision-making that keeps it running. They do not need to phone us to make a change to the product feed, or to interpret the analytics dashboard, or to handle an operational problem in the warehouse. They can do it themselves, because we built the operation with them rather than for them.
That distinction matters more than it might sound. Building something for a client is faster. You bring your own people in, do the work efficiently, hand over the login details, and leave. Building something with a client is slower, because it means involving the internal team at every stage: explaining why a decision is being made, not just what the decision is. Showing someone how to configure the integration, not just configuring it. Sitting with the customer service team and designing the workflows together, not delivering a process document by email.
It is messier and it takes longer. But at the end of it, the internal team owns the operation. They do not just have access to the tools. They understand the logic behind them.
The handover is not a phase at the end of the project. It is a thread that runs through the whole thing.
In the first few weeks, we are doing most of the work. That is inevitable, because the internal team is often learning D2C from scratch. In manufacturing particularly, the person given responsibility for direct-to-consumer may have come from sales, operations, or marketing and has never run an ecommerce channel before. But even at this stage, we are documenting everything, explaining the decisions, and making sure the internal person is in the room for every conversation that shapes the project.
By the middle of the engagement, the balance is shifting. The internal team is taking on more of the operational work, with us alongside them. They are running the platform, managing the content, handling the customer service workflows. We are still there, but the role has changed from doing to supporting. When a problem surfaces (and problems always surface), we solve it together rather than solving it for them. That is deliberate. The problem-solving is where the real learning happens, not in a training slide deck.
By the end, the internal team is running the channel. We are available for questions, for the occasional tricky integration issue, for a second opinion on a decision. But the day-to-day operation sits with them. The knowledge is in their building, not ours.
We document the lot. Not as a 200-page manual that nobody reads, but as process maps, workflow diagrams, and practical guides written in the language the team actually uses. We call it the glass box handover: everything is visible, nothing is hidden, and the team can see how every part of the operation connects to every other part.
I should be straightforward about the commercial reality of this model, because pretending it is easy would be dishonest.
A retainer model is more predictable revenue. The client pays a monthly fee, you provide an ongoing service, and both parties settle into a rhythm. That rhythm can last years. It is comfortable for the partner and familiar for the client. The entire agency industry is built around it.
Our model is the opposite. We go in, do the work, build the capability, and leave. The engagement has a natural end point, and we work towards it deliberately. That means our revenue from any single client is front-loaded. It means we are constantly needing to find the next engagement rather than sitting on a recurring income stream. And it means we have to be genuinely good at what we do, because our only route to repeat business is a client who valued the work enough to call us back for something new.
That, I think, is the part that actually makes the model work. A partner who is incentivised to stay is incentivised to make themselves indispensable. A partner who is incentivised to leave is incentivised to make the client capable. The incentives align with the outcome the client actually wants.
The test of whether the handover worked is not the day we step back. It is three months later, when something unexpected happens and the internal team has to deal with it without us.
A supplier changes their shipping rates and the fulfilment costs need recalculating. A product launch requires new content, new product data, and a campaign plan. The integration throws an error that nobody has seen before. The board asks a question about unit economics that was not in the last reporting pack.
If the team can handle these situations independently, the handover worked. If they cannot (if they need to call us for every decision, if they are stuck because they do not understand the systems or the logic behind them), then we have not done our job properly, regardless of how well the D2C channel performed while we were in the room.
We are not always perfect at this. Some engagements end with the internal team more reliant on us than we would like, usually because the business did not invest in the internal capability we recommended, or because the person we trained moved on and their replacement had to start from scratch. But the intention, the design of every engagement from the first week, is that the knowledge transfers and the team becomes self-sufficient.
There is a pattern we have noticed over the years, and it is the best evidence I have that this model works. The clients who no longer need us are the same clients who call us back.
Not because something broke. Because the D2C channel worked, the business grew, and a new challenge appeared that needed the same kind of operational thinking: a new market, a new product range, a new integration requirement. They come back because they trust the way we work, not because they are dependent on us.
That is a different kind of relationship than a retainer. It is less predictable, less comfortable, and less lucrative in the short term. But it is built on something real: a client who is genuinely better off for having worked with us, and who chooses to come back rather than having no alternative.
That is why we want to fire ourselves. Not because we do not value the relationship. Because firing ourselves is proof that we did the job properly.
