Three revenues, one deal

A deal closed last month. Everyone who touched it is pleased. The agency put it in their quarterly update. Our finance team booked it. My sales team logged it in the CRM and started arguing about the bonus. Three reports, one deal, and the number in each of them is different. None of them is wrong.

That took me longer to accept than I'd like to admit. When I started selling media I assumed "revenue" was a word with one meaning. It isn't. It's a word with a footnote, and the footnote depends on who is holding the report.

The waterfall

Here is how a CTV or in-app campaign turns from a budget into money the team can point at. The numbers below are illustrative; the proportions are the kind you meet in practice, not from any particular contract.

Start with gross. That's the line in the media plan, the figure the advertiser signed off on, the amount the agency will list under billings. It is the biggest number the deal will ever have, and it's the one people remember at conferences.

Then the agency commission comes off. The agency won the brief, did the planning, holds the relationship; their fee lives inside the gross figure. What lands on the platform's invoice is net. No report stops here, but this is usually the number my salespeople quote when they say "I closed X". It's already smaller than what the client thinks they spent.

Then the third-party technology takes its share: the exchange or SSP fee, ad serving, verification, sometimes a data or measurement provider. Who contracted them decides where they land. If the client did and we only collect on their behalf, it's pass-through: it crosses our invoice and never becomes our revenue, so it comes off here. If we contracted them ourselves, it's our cost, and it belongs below the revenue line, next to the media. The figure takes the first case and folds our own tech in with the media at the bottom.

Then the month closes and the adjustments arrive. Under-delivery credits (a makegood would cost us media instead and leave the invoice alone), discrepancies between our numbers and the client's, a credit note because a creative ran in the wrong slot for two days. Individually small. Together, a steady few percent that the deal loses after everyone has already celebrated it. What survives is net-net, and that's what finance will recognize. One assumption hides in that sentence: the platform buys the media in its own name, as principal, so the media is its cost rather than the client's money passing through. An agent would recognize only its fee, and the drawing would look different.

And finally the media itself. The inventory came from publishers and apps, and they get paid for what was delivered, whatever the credit note said. Our own tech sits here as well. What remains after that is the only number that has ever bought anyone a coffee: the contribution the team actually generated.

Move the sliders and watch how quietly the last bar shrinks. A few points of agency commission, a few points of pass-through tech, a modest adjustment: none of them sounds like much on its own. Compounded, they change the shape of the deal. Then switch whose report you're reading, and notice that the red line lands on a different bar every time.

Three honest reports

The agency reports gross. That's not vanity; it's their business model. Their scale is measured in billings, and the commission is a share of it. When they say "we manage a budget of X", they mean the top of the waterfall, and they are right to.

The platform reports net-net. Finance has rules about what counts as revenue, and those rules don't care how big the media plan was. They care what was invoiced, what survived the adjustments, and what is actually collectable. Pass-through tech never reaches the revenue line at all; tech we contracted ourselves is a cost below it, next to the media.

The team reports what's left. Or it should. A sales dashboard that shows gross is a dashboard that makes everyone feel good and tells you nothing about which deal was worth the time.

Revenue is not a number. It's an agreement about where to stop subtracting.

Once you see it that way, a lot of arguments dissolve. The agency isn't inflating. Finance isn't shrinking. They stopped subtracting at different lines, because different lines are what their readers need.

Why a seller should care

Three reasons, in ascending order of how much each has cost me personally.

The bonus. Whatever your commission plan is tied to, it's tied to one specific line in the waterfall. If the plan says "net revenue" and you've been mentally counting gross all quarter, the payout will come as a surprise, and not the fun kind. Read the plan. Find the line. Then sell towards that line, not towards the biggest number in the deck.

Priorities. Two deals with the same gross can leave the team wildly different amounts. A direct-to-brand deal with a modest budget, little third-party tech and clean delivery can outperform a much larger agency deal that arrives with a full stack of pass-through fees and a history of end-of-month disputes. If you only look at gross, you'll chase the wrong one. Every time.

The conversation with finance. The expensive one. When a finance colleague says the deal "came in lower than forecast", they are not accusing you of anything. They are looking at net-net after adjustments; you were looking at net before them. Bring the waterfall to the meeting. Ask which line they're using. Half the tension in those conversations is two people using the same word for different bars.

What I do about it

Our internal dashboard shows every line, with the fee steps visible, and a switch for which view you're looking at. It isn't elegant. It's a table with too many columns. But when someone says "revenue", we can point at a column and agree that's the one.

Two rules came out of it. First: any number that leaves the team goes with its label attached. Gross, net, net-net. A bare "revenue" in a message is a bug. Second: sellers see the last bar, not just the first one. It's occasionally demoralizing. It's also the only way I've found to get people to care about a discrepancy clause before the month closes rather than after.

The deal in the first paragraph is still a good deal. It's just three different sizes depending on who's asking. Learn which size is yours.