Skip to main content
Revenue6 min read

Revenue by source is not profit: subtract the ad spend

A paid channel can top your revenue table and still lose money. How to put spend next to revenue for every source, worked through on a real-looking week.

Vicente TorresFounder of Vero

Once sales are tied to the source that brought them, the revenue table becomes the most persuasive screen in your analytics. It is also the easiest to misread. Revenue by source says nothing about what each source cost, and the channels you pay for are exactly the ones a revenue table flatters. This post shows how to put spend next to revenue for every source, the three numbers worth reading, a worked example from the Vero demo, and the limits of any attribution model when you use it to move money.

Why does a revenue table flatter paid channels?

Paid channels are built to produce visible revenue. You can buy reach on any platform, and some of the people you reach will buy. A revenue table records that sale and ranks the source by it. What it does not record is the invoice from the ad platform, which arrives somewhere else: a monthly charge on a card.

Unpaid channels have the opposite problem. A mention on a forum or a good search result costs nothing per visit, so its revenue is almost all profit, and yet it can sit below a paid source in the table. Reading revenue alone pushes you to spend more where you already spend, and to neglect the sources that were free.

How do you calculate profit, margin and ROAS by source?

Revenue is what the source brought. In Vero that means one thing for every provider: what was charged, without tax, less refunds, before your payment provider’s fees. Spend is what you paid to get that traffic in the same period: ad budget, a sponsored newsletter slot, a paid placement. Profit is revenue minus spend.

Two ratios help you compare sources of different sizes. Margin is profit divided by revenue: how much of each euro the source brought you keep after paying for it. ROAS, return on ad spend, is revenue divided by spend: how many euros came back for each euro you put in. A ROAS below 1 means the source loses money before any other cost. Above 1 it covers its own spend, and whether that is enough depends on your other costs, which no analytics tool knows.

A worked example: the demo week

The Vero demo is a fictional SaaS in its launch week, with €18,423 of revenue and a fixed ad spend entered for four sources. Here is each one, by utm_source.

twitter brought €4,821 against €1,200 of spend: €3,621 of profit, a 75% margin, a ROAS of about 4.0. google brought €1,783 against €1,400: €383 of profit, a 21% margin, a ROAS of about 1.3. It pays for itself, barely. The sponsored newsletter brought €482 against €900: a loss of €418, a margin of -87%. And reddit, where the ads sent 421 visits, brought no sale at all against €1,150 of spend: a straight loss of €1,150.

Now the sources with no spend. The untagged traffic, which includes the Hacker News front page, brought €9,092. producthunt brought €1,845. chatgpt brought €400. None of them cost anything per visit, so all of it is profit before your other costs.

Read as revenue, google and the newsletter look like contributors. Read as profit, the newsletter and reddit together lost €1,568 in a week when the free sources made most of the money. That is the decision this view exists for. The vanity traffic post covers what to do with sources like reddit that send visits and no sale.

Make the names match before you trust the table

Profit by source only works if the spend you enter and the revenue Vero attributes use the same name for the same source. Spend for google next to revenue tagged Google and google-ads is three rows, two of them wrong. The fix is a naming rule you write down once and follow every time you build a link.

Lowercase everything. Use utm_source for the place (google, the newsletter’s name, reddit), utm_medium for the kind of link (cpc, email, social) and utm_campaign for the specific push. Never tag internal links on your own site: a tag says where a visitor came from, and a link on your own page is not a source. In many tools it also overwrites the campaign that really brought them. Build tagged links in one place, a UTM builder like the one in the Vero dashboard, so nobody types a source name by hand. The marketing teams use case shows how a team keeps this consistent.

What any attribution model can and cannot credit

Every profit table rests on an attribution rule, and you should know the one you are reading. Vero credits each sale to the first touch of the visit in which the customer paid: the source that brought them into that session. Renewals keep the source of the customer’s first attributed payment. The rule is simple and the same one the traffic reports use, so money and traffic agree on what a source is. How to attribute revenue without cookies explains how the sale and the visit are linked.

Its limit is the journey that spans visits. Someone who clicks an ad today, comes back next week by typing your address and pays then is credited to Direct, not to the ad. So a channel that opens journeys and leaves the closing to another visit will read low in this table, and a channel people return through will read high. That does not make the table wrong. It makes it a strong signal for channels that convert in one visit and a conservative one for channels that start long decisions. Weigh a borderline source with that in mind before you cut it.

From the table to a decision

Look at profit by source once a month, over a whole month, with that month’s spend. A week is too noisy for most paid channels, and spend is usually billed monthly anyway. Then sort every paid source into one of three groups.

Profitable with room: margin well above zero and ROAS comfortably over 1. Try more budget and watch whether the margin holds as you scale. Marginal: ROAS close to 1. Fix the landing page or the offer before adding money, and give it one more month. Losing: ROAS under 1 for two months running. Cut it, or keep it as a deliberate brand cost with a number you have agreed to lose.

Profit by source is part of revenue attribution on Pro, and accounts in the beta have it. Connect Paddle, Stripe or Lemon Squeezy first, because spend next to an empty revenue column tells you nothing.

Unsure of a term? Read the glossary

Vicente Torres
Vero is built by one person in Spain. I read every waitlist entry and pick the beta accounts myself.
Vicente Torres, Founder of Vero LinkedIn
Hand-picked beta, waitlist open

Ask for a place in the beta.

Anyone can join the list. The accounts we pick get Pro free for the whole beta.

By joining you agree to receive launch updates and accept our Privacy Policy.

See the demo