You ran twelve campaigns last quarter. Name the one that made you money.

Ask a marketing team what they shipped last quarter and you will get a real answer. Twelve campaigns. A webinar, three email sequences, a content push, a paid experiment, a redesign of the newsletter nobody asked for.

Ask which one made money, and the room goes quiet.

The quiet is the problem, and it is not that the team is lazy. They are visibly busy. Being busy and making money were never the same measurement, and somewhere along the line the first one crept in and took the place of the second.

Opens are not outcomes

A campaign measured by opens, clicks and impressions is really being measured on effort. Send more and the numbers climb, which is exactly why they make a poor stand-in for results.

The real question is harder to sit with: what pipeline did this create, from which accounts, and how do we know it was this campaign and not something else running at the same time. A campaign you cannot tie to a number on the pipeline is a donation with a good subject line.

Design backward from the number

Most campaigns are designed forward. Someone has an idea, it becomes a brief, the brief turns into assets, and the assets ship. Nobody stopped to ask what the campaign was supposed to move until the wrap-up meeting, by which point it is too late to have wanted anything in particular.

Design it backward instead. Start from the revenue question: which segment, what stage, what should be true afterward that is not true now. Then work back to the play that would cause it. Now the campaign has a target before it has a subject line, so you can judge whether it worked and fix what did not. The forward version leaves you nothing to grade at the end. You admire it and move on.

Triggers beat the calendar

The batch-and-blast calendar - everyone gets the same thing on the same Tuesday - is easy to run and mostly ignored. It treats a person who just spoke to sales the same as a person who has not opened anything in a year.

The higher-return version fires on behaviour. The message goes out in response to something the buyer did: a stalled deal, a new signup, a renewal window, a page visited three times. That trigger is already sitting in the CRM. Most teams never wired a campaign to it, so the timing gets handed to a calendar that has no idea who the buyer is.

Attribution is imperfect. No attribution is worse.

Every marketer eventually finds out that attribution is a swamp. First touch makes demand gen look good, last touch hands the credit to sales, and multi-touch tends to flatter whoever built the model. Not one of them is the truth, and you can waste a lot of time pretending otherwise. We break down each model, and where it lies to you, in marketing attribution, and the part no model can see in the dark funnel.

People take that, constantly, as a licence to measure nothing, which is the wrong lesson to draw. Pick one model. Learn exactly how it lies to you, and hold it steady from there. A consistent imperfect number at least tells you which way things are moving, and which way things are moving is all you need to decide what to do more of next quarter. Nobody is going to cross-examine the model.

Close the loop, or repeat the mistake

The campaign creates a lead. Sometimes that lead turns into a deal and sometimes it dies. Either way, in most companies the outcome never makes it back to the person who ran the campaign, so they run the same one again next quarter and change nothing.

That loop is the point of the whole exercise. Campaign to CRM to what closed, then back into who you target next time. It falls apart the moment sales stops logging what happened, which is its own ongoing fight, because a campaign is only as good as the activity the CRM can see. It falls apart again when marketing, sales and finance each read their own numbers, three private versions of the truth.

Run twelve campaigns if you have the budget. Just make sure that at the end of the quarter, when someone asks which one made money, the room does not go quiet.