I have hired a lot of marketers over the years, and the fastest way I can tell whether someone will do well buying traffic is to hand them a growth goal and watch how they break it down. The people who last do not start with creative ideas or channel opinions. They start with the numbers, because unit economics is the layer underneath every campaign decision you will ever make.
In this guide I want to walk you through how I think about it. We will decompose a revenue goal into its parts, talk about the difference between what you pay and what you earn back, and work backwards from a target return all the way to the most you can afford to bid. By the end you should be able to look at a campaign and say, with a straight face, whether it deserves more money or a quiet death.
Start by decomposing the goal
Say your target is 100,000 dollars in revenue this month from paid acquisition. That number on its own tells you nothing about what to do on Monday. The trick is to break it into the levers you can actually pull.
Revenue almost always factors into three pieces: volume times conversion rate times average revenue per customer. So if your landing page converts clicks to buyers at 3 percent, and each buyer is worth 50 dollars on the first purchase, then you need 100,000 divided by 50, which is 2,000 buyers. To get 2,000 buyers at a 3 percent conversion rate, you need about 67,000 clicks. Now you have something concrete to plan against.
The reason this matters is that every one of those factors is a separate job with a separate owner. Volume is a media buying problem. Conversion rate is a landing page and offer problem. Average revenue is a pricing and merchandising problem. When a goal is missed, decomposing it tells you which lever slipped instead of leaving you to guess.
CPI and CPA are not the whole story
New marketers tend to fixate on cost per install or cost per acquisition, and I understand why. It is the number the ad platform shouts at you all day. But CPI and CPA only describe what you paid to get someone in the door. They say nothing about whether that person was worth it.
The number that actually matters is payback: how long it takes the revenue from a customer to cover what you spent acquiring them. A 40 dollar CPA is wonderful if the customer pays you back in two weeks and great for years after. The same 40 dollar CPA is a slow disaster if the customer churns before you ever recover it. Cost is only half of a ratio, and a ratio needs both halves before it means anything.
Contribution margin and the core identity
Before you compare what a customer earns you against what they cost, you have to strip out the costs of actually serving them. What is left is contribution margin, and it is the real fuel for your acquisition budget.
Here is a quick example. A customer generates 50 dollars in revenue. Payment processing takes about 1.50 dollars, cost of goods or delivery takes 15 dollars, and support and refunds average another 3.50 dollars. Your contribution is 50 minus 20, or 30 dollars per customer. That 30 dollars, not the 50, is what you have to spend on acquisition and still come out ahead.
That leads to the one identity you should tattoo somewhere visible:
- LTV must exceed CAC plus variable costs. Lifetime value has to beat what you paid to acquire the customer plus what it costs to serve them.
- A common rule of thumb is a 3 to 1 ratio. If lifetime contribution is at least three times acquisition cost, you usually have room for overhead, refunds, and the campaigns that do not work out.
- Payback under 12 months is a healthy target for most subscription and repeat-purchase businesses in the US market, though faster is always better for cash flow.
Working backwards to a max bid
This is the skill that separates people who react to the auction from people who control it. Instead of asking what the platform charges, you decide in advance the most you are willing to pay, then let that number set your bids and budgets.
Start with your target ROAS. Suppose you want a first-purchase ROAS of 2, meaning two dollars of revenue for every dollar of ad spend. With 50 dollars of revenue per customer, a ROAS of 2 means your maximum allowable CPA is 25 dollars. Now push it one more step. If your landing page converts clicks to buyers at 3 percent, then your max cost per click is 25 dollars times 0.03, which is 0.75 dollars. Bid above 75 cents a click and, at this conversion rate and revenue, you break your own ROAS target.
Notice what just happened. A vague goal (make it profitable) became a hard operating limit (do not pay more than 75 cents a click). That is what unit economics buys you: the ability to walk into any channel and know your ceiling before you spend a dollar.
Scale it or kill it
Once the math is in front of you, the decision usually makes itself. If a campaign brings customers in under your max CPA and pays back inside your window, you feed it more money and watch whether the economics hold as you scale, because costs tend to rise as you push volume. If a campaign runs above your ceiling, you have two moves before you kill it.
Move one is to improve a factor in the chain: lift conversion rate, raise average revenue, or trim the variable costs eating your margin. Any of those loosens the max bid and can rescue a campaign that looked dead. Move two, if none of that works, is to cut it without ceremony. Marketers get attached to campaigns they built, and that attachment quietly burns budget that a winner could have used.
Unit economics is not glamorous, and it will not show up in a portfolio the way a clever creative does. But it is the thing that decides whether your growth makes money or just makes noise, and it is one of the most common things I probe for in an interview or a test task.
Key takeaways
- Decompose every revenue goal into volume times conversion rate times average revenue, so you know which lever to pull when a number slips.
- The core rule is that lifetime value must beat acquisition cost plus variable costs, and a 3 to 1 ratio gives you room to breathe.
- Work backwards from a target ROAS to a max CPA and then a max bid, so you know your ceiling before you spend, and use it to decide whether to scale or kill.