Why business immersion has to come before the quote
I can't price or plan serious work without understanding how the client makes money. If I don't know which product sustains revenue, who buys, and why, any technical proposal is a shot in the dark. Onboarding starts before the client even signs, in the very first conversation.
That's where I start putting myself in the client's shoes and understanding the value they sell in the market. Without that, optimizing a marketing process risks speeding up the wrong tasks.
Some questions that help build that picture:
- Which product or service generates the most revenue, and which leaves the highest margin?
- Who is the typical customer, and through which channels do they reach the business?
- What objections come up during the sale, and how does the team usually handle them?
- What is the main business priority for the coming months?
If I close a scope of work without these answers, I have no way to judge whether the project is viable. Much of this can and should be answered during pre-selection, before any contract, because I also need to decide whether the project is worth taking on.
The metrics that decide whether a project has legs
Whether a marketing project is viable depends on knowing margins, customer acquisition cost, and lifetime value. If the company doesn't have these numbers, technical work can end up masking a loss instead of fixing it, and nobody notices until it's too late.
Often the business owner can't answer these questions themselves, and it's my job to guide that discovery before starting. I've talked in class about a concrete case that illustrates the risk of skipping this step: in one partnership, the team looked at revenue and celebrated, while the client looked at ROAS and watched margin get squeezed. Revenue was hitting records and the client was still unhappy, because nobody had agreed beforehand which metric actually mattered.
Questions worth asking before I start:
- Margin per product or service: what's left after direct costs, fees, and fulfillment.
- Current CAC and break-even CAC: the acquisition cost above which the operation stops being profitable.
- LTV and repeat purchase: whether the business relies on repeat sales or a single low-margin transaction.
If a client can't tell me their break-even CAC, any campaign risks producing nice-looking numbers alongside real losses. Give me the data, explain the rule, show me where it broke is, in essence, what I need from these conversations.
How to use AI in business beyond chatbot prompts
AI in B2B prospecting: how to qualify opportunities before reaching out
To design execution and handle exceptions, explore the business process automation guide.
Auditing the assets: don't send traffic to a broken site
A slow site, a form that breaks on mobile, or a confusing checkout can undo much of the value of advertising work. Before connecting traffic channels, I need to check whether the house is in order, because no campaign can fix a page that doesn't work.
If I'm running paid traffic, for instance, I won't agree to work with a page that fails basic requirements: no matter how well-structured the campaign is, if the site is broken, results leak away along the path, and the advertising work unfairly takes the blame.
Points I check before moving forward:
- Site or landing page: load speed, mobile behavior, clarity of the value proposition, visible conversion buttons, a checkout without surprises.
- Data tools: analytics, tag manager, pixels, and conversion events installed and working.
- Accounts and access: ad accounts created, audiences defined, business manager configured.
This review produces a simple document with improvement recommendations. That gives the client visibility into what needs fixing before moving to more advanced work, and it also builds trust: the client sees someone on the other side asking the right questions.
Who does what: defining this early avoids months of friction
A marketing project needs clear points of human decision-making: who approves, who produces, who handles exceptions. Without this defined early, delays and confusion follow. Worth asking during pre-onboarding: is someone responsible for digital marketing, who's the main contact for approvals, who produces creatives and runs tests. Sometimes the answer is "that's on me," which is fine if explicit.
It's worth asking, still during pre-onboarding: is there someone on the client's team responsible for digital marketing? Who is the main point of contact for approvals and strategic discussion? Who produces creatives, who runs page tests, who handles content planning? Sometimes the answer is "that's on me," and that's fine, as long as it's explicit.
Defining the communication channel and cadence also avoids misunderstandings. Meetings scheduled just because "we always meet on Mondays" tend to be less productive than meetings built around an actual agenda, and it's normal to need more check-ins early on than later.
Setup versus retainer: how not to work for free
When the audit reveals a very messy setup, that extra startup work doesn't automatically get absorbed into the regular monthly fee. Trying to fix everything in week one out of pure goodwill, without charging for that effort, sets an expectation that's hard to sustain.
That habit has a cost: it sets an expectation in the client's mind that this level of dedication is the normal pace, something that's hard to keep up after six or twelve months, once there are more clients on the calendar. And if the project ends up not moving forward after months of heavy unpaid work, the loss falls on me.
Two ways I protect myself:
- Charge an upfront setup fee, which can go up to double the regular monthly fee, dedicated solely to fixing pages, accounts, and data.
- Negotiate a longer contract, for example 12 months, in exchange for spreading that initial work into the retainer instead of charging separately.
Which option makes sense depends on the case, but the principle is the same: if there's a lot of setup work, someone has to pay for it. Unvalidated businesses fail often, and I'd rather go in protected than find that out three months later with nothing to show for it.
A simple guide for the discovery meeting
This diagnosis doesn't need to be complicated. It fits into a thirty- to sixty-minute conversation, as long as each question connects to a concrete decision rather than filling meeting time. I ask which product has the best margin, the break-even CAC, whether the site is ready for traffic, who approves what, and which metric everyone will watch.
If a client can't answer half, that's where I start.
Which product has the best margin, what's the break-even CAC, is the site ready for traffic, who approves what, and which single metric will everyone watch during the first months. If a client can't answer half of these, that's where I start.
Source note
This article draws on Sextas Ímpares #89, "Marketing Client Onboarding Process," where I talked through discovery questions, asset audits, expectation alignment, and charging for setup work before starting with a client. You can watch the full class here: https://www.youtube.com/watch?v=LEAed-8P2is (the topic starts around 3:54, the revenue-versus-ROAS misalignment example comes up near 12:15, and the setup fee versus longer contract trade-off is discussed from around 17:40).