Who is the main contact for approvals
The starting point is asking the client directly who will be the point of contact for approvals, meetings and KPI discussions, even before any contract exists. I include this question in the immersion checklist I apply before any project kicks off, asking who the main contact will be for approvals, meetings and strategic discussion.
Without this defined role, whoever is running the project wastes time figuring out which internal opinion actually counts. If the marketing manager asks for one thing and the sales lead asks for the opposite the next day, the work stalls and nobody knows who was right. Having one name and one channel from the first conversation solves this, before any paperwork is even signed.
This question is part of a broader set of immersion questions I collect during the pre-selection phase, often before I even know whether I will work with that client. This immersion work does not start only when the contract is signed, it starts from the very first conversation with a potential client, even knowing many of those first contacts never turn into a project.
Aligning expectations on the right metric
Deciding who approves what only matters if everyone is looking at the same number, every month. I share a real episode from my agency where the team reported results in revenue while the client was actually tracking ROAS, and that mismatch created quiet friction.
Revenue was breaking records, but the client's margin was being eaten away by rising ad spend, and that created dissatisfaction despite the agency's positive numbers. The client is not always looking at the same metric as the technical team, so this needs to be aligned explicitly rather than assumed.
This example shows that aligning expectations isn't a one-off step at kickoff, it's ongoing work repeated in every report and every meeting. If the client cannot say which metric they check every day, that is exactly where you should start before touching any campaign or proposing any budget change.
How to convert more leads into clients in a service business
How to use AI in business beyond chatbot prompts
To design execution and handle exceptions, explore the business process automation guide.
Who does what in day-to-day operations
During onboarding, I ask clients a block of questions about their internal operations: who handles marketing tasks, who produces creatives, who tests pages and who plans content. Answers sometimes point back to the external team itself, but what matters is getting each task assigned clearly in writing, so nothing falls through the cracks between sides.
These answers may point back to the external team itself, but what matters is having each piece of work clearly assigned in writing. Without this explicit split, tasks fall through the cracks because each side assumes the other is handling it, and the gap only surfaces weeks later when something visibly breaks.
These answers may point back to the external team itself, but the important thing is that it's written down clearly who owns each piece of the work. Without this explicit split, it's easy for a task to fall through because each side assumed the other was handling it, and the gap only surfaces weeks later when something visibly breaks.
It's also worth asking about the preferred communication method, whether Slack, WhatsApp, email or phone, and fixing that from the first week so information doesn't scatter across channels throughout the project.
Auditing assets before deciding what changes
Before approving any change to a website, landing page or ad account, I first need to understand the current state of those assets. I describe this step as a checklist: is the site secure, is it fast, does it work well on mobile, is the value proposition clear, is the checkout free of surprises.
The same applies to technical tools such as pixels, tag managers, Google Analytics and existing or missing ad accounts. Without this initial check, the risk is sending paid traffic to a site with structural problems and wasting budget on something that was never going to convert, regardless of how good the campaigns are.
This analysis produces a document with concrete recommendations, which becomes the basis for deciding with the client what changes, when it changes, and who is responsible for each fix. That document gives the main contact the information needed to approve or reject each point, instead of loose decisions made mid-conversation with no record.
Meetings only with a defined purpose
An important part of deciding who approves what is also deciding when to meet. I am direct about this: every meeting needs a concrete objective, and I shouldn't walk into a meeting without knowing exactly what needs to be decided.
Scheduling weekly meetings just because it seems professional eats into the time needed by whoever executes the work every day, and it doesn't replace a clear list of what needs approving. At the start of a project it's normal to have closer follow-up, with several meetings concentrated in a single week if needed, and that pace tends to space out over the following months.
What shouldn't change is the underlying criterion: every meeting goes on the calendar because there's something concrete to approve or review at that moment, not because it was simply scheduled since the start of the contract.
Setup fee or a longer contract as a trade-off
When the start of a project requires a large amount of upfront work, that extra load needs to be recognised somehow in the contract terms. I propose two alternatives: charging a separate setup fee on top of the retainer, or negotiating a longer contract in exchange for not charging for that initial work.
Closing a twelve-month contract in exchange for not charging the initial setup is an extremely valid and fair trade-off for both sides involved. Without this explicit negotiation, there's a real risk of putting weeks of heavy work into a client who, three months later, decides not to continue, precisely because the business itself wasn't yet validated in the market.
Deciding upfront who pays for what and under what conditions protects both sides from what might go wrong later on. It also prevents the external team from being exposed to months of unpaid work if the project ends earlier than expected at the client's decision.
Don't try to solve everything in the first week
A common mistake, which I admit to having made early in my agency, is wanting to fix everything at once at kickoff: change the site, create creatives, launch campaigns, configure pixels, almost all at once. This creates an expectation in the client that this intense pace will be the permanent norm.
When the team can't sustain that pace over months, because the business grows and the client roster expands, it generates frustration instead of recognition for the initial effort. I want to understand what changed and what I need to decide, and that's only possible if the action plan is split into realistic phases from the start, with clear stages and review dates, instead of an implicit promise to solve everything in the first week.
Sextas Ímpares #89: Marketing Client Onboarding Process
This class, which I recorded and presented as Sextas Ímpares #89, brings together the pillars I use to think about onboarding marketing clients: understanding the business, building security, aligning expectations, and building a realistic action plan. You can watch the full class here: https://www.youtube.com/watch?v=LEAed-8P2is. Worth noting is the moment around the 14-minute mark, where I go through the question about the main contact for approvals, and the moment around the 12-minute mark, where I tell the story of the mismatch between revenue and ROAS.