Running a business without these numbers is like navigating without knowing your destination. Higher revenue is not the same as a healthy company. When ad costs rise or the team grows, without these reference points any attempt to scale can turn into a loss.
ROAS is not profit: calculate your break-even
To know whether you are actually making money, you must separate campaign ROAS from business profit, and calculate the minimum ROAS at which you stop losing money. ROAS divides ad-driven revenue by ad spend, ignoring product cost, suppliers, payroll, or other costs, which is why it can mislead anyone reading that number alone.
The most common mistake among people running paid traffic is celebrating a high ROAS without looking at product margin. The example used in the class is a product with 10% margin: spend 10 euros on ads and generate 20 euros in sales, and you get an apparently positive ROAS of 2. But of those 20 euros you keep only 2 in margin (10%): you spent 10 to earn 2, losing 8 on that sale.
With a 10% margin, the break-even ROAS is 10: only from that point is the ad cost actually covered by the product margin. If a team celebrates a ROAS of 7 without doing this math, it's losing money without realizing it. As explained in the class, if you don't know your break-even ROAS, you have no way to judge whether a campaign's result is good or bad, whatever multiplier the dashboard shows.
One-off purchase or recurring business: that decides your CAC
To know how much you can pay to acquire a customer, you first need to understand whether your business lives off a single sale or off recurring revenue over time. That distinction determines whether CAC should be your guiding metric or whether lifetime value should drive the calculation instead.
If you sell a product with little repurchase within a 12 to 24 month cycle, you need to profit on the first transaction, because there is no future revenue to make up for an initial loss. In these businesses, cost per action needs constant watching, since there is no second chance to recover what was lost on acquisition.
When the business runs on recurring revenue, contracts, subscriptions, or retainers, the logic changes. The class gives the example of an agency that can pay 2,000 euros to acquire a client paying 1,000 euros a month, even knowing the first month runs at a loss, because the client typically stays around 16 months, adding up to roughly 16,000 euros over the cycle. The critical point is having enough cash flow to absorb the gap between paying for acquisition and receiving the return, without running dry along the way.
Before switching off a campaign, understand which sales it helps
How to design a dashboard that shows which clients need attention
Growing an online business: what I learned about margin, team, and AI
Cheap or expensive cost per lead depends on what you can do with it
The right way to judge cost per lead is to look at the average profit you generate per lead, not an isolated number copied from someone else's business. Calling a lead cheap or expensive without that reference is a guess with no foundation.
A 1-euro lead can be terrible if you can't convert those people into paying customers afterward. Paying 15 or 20 euros per contact can be excellent in a business with a good close rate. Comparing to what you paid years ago also distorts things: ad platform auctions tend to get more expensive over time as competition and demand increase, much like rents or wages rise. Expecting to return to past costs isn't realistic, and clinging to that comparison distracts from the real problem.
Instead of trying to artificially lower cost per lead at the expense of worse traffic quality, the lever usually lies in monetizing better every lead that already comes in: revisiting the offer, improving messaging, and speeding up the sales response.
Lifetime value and churn: where growth usually lives
Lifetime value and churn reveal whether a business accumulates value over time or keeps starting from zero. Retaining existing customers longer stretches their value without requiring additional acquisition spend. That sustained retention is typically where the missing margin lives, making customer duration central to long-term profitability and consistent growth.
The class suggests calculating LTV within a 24-month window even with longer-standing clients, since overly long horizons distort the present-day read on the business. Churn is the counterpart: if 70% of students renew yearly, that's 30% annual churn. The point isn't just logging that number, it's asking people directly, through a simple survey, why they didn't renew, then working on the product to reduce drop-off and raise LTV without spending more on acquisition.
The class suggests calculating LTV within a 24-month window, even with longer-standing clients, because looking at overly long horizons can distort the present-day read on the business. Churn is the counterpart: if 70% of students renew year over year, that means 30% annual churn. The goal isn't just logging that number, it's asking people directly, through a simple survey, why they didn't renew. Understanding those reasons lets you work on the product and reduce drop-off, which raises LTV without spending more on acquisition.
Finding the right bottleneck in the funnel
A drop in revenue is rarely diffuse: it usually concentrates in one specific funnel stage, and that's where it's worth acting first instead of optimizing everything at once. Identifying that stage requires measuring conversion between each step of the customer journey, from first contact to payment.
In a services business, the funnel moves through lead generation, qualification, call attendance, and proposal closing. If you generate many leads but people don't show up for scheduled meetings, the fix usually isn't more ads; it's better reminders, adjusting the contact approach, or tighter upfront qualification. In e-commerce, the funnel moves through product views, add-to-cart, checkout start, and payment. Understanding exactly where you lose the most people lets you intervene precisely there instead of spreading effort across the whole operation.
Building predictability from these numbers
Predictability comes from connecting money spent on acquisition to money that actually ends up left over at month's end. Without that link, a good month feels like luck and a bad month feels like an unexplained disaster for the team.
A simple way to organize this, without complex platforms, is to regularly track four questions: what was the real ROAS versus the calculated break-even; what was the CAC against the period's margin; how is the 24-month LTV tracking; and where is the biggest drop in the funnel. This mostly requires discipline, well-tagged links (UTMs), and consistent CRM logging, not sophisticated tools or expensive dashboards.
When the team shares these readings, fewer decisions get driven by isolated intuition. Ad spend can be adjusted with clearer criteria, and focus can shift toward profitability that holds over time, not just the current month's revenue figure.
Source note
This article is based on my class, in this class #112, dedicated to the fundamental metrics of any digital business: https://www.youtube.com/watch?v=znR9UtcIBTU.