I Can't Prove What Marketing Returns

Why the link from spend to revenue breaks

The data usually exists; it just is not linked from spend to deal.

Each tool tells its own story

Ad platforms report the conversions they can see and credit themselves for them. Analytics uses another model, the CRM another. Adding them up counts the same customer more than once.

Leads lose their source on the way to a deal

A form fill carries its source. By the time it becomes an opportunity and a closed deal, the source field is empty, overwritten or set to "direct".

Nobody agreed what counts

Without a written definition of a qualified lead, an opportunity and a marketing-sourced deal, marketing and sales report different numbers for the same quarter.

Tracking broke and nobody checked

Consent changes, a site rebuild or a new form tool can stop events from firing. The reports keep running, showing less and less of what happened.

Trace last quarter's deals back to their source

Take the deals you closed last quarter and try to answer these from your own systems.

• For each closed deal, the CRM shows the first and the most recent marketing touch.
• Marketing and sales use the same written definition of a qualified lead and an opportunity.
• The conversions your ad platforms reported roughly match the leads your CRM recorded from those platforms.
• You can show spend by channel next to pipeline and revenue by channel for the same period.
• Someone checked in the last quarter that your forms, tags and CRM sync still record what they should.
• Finance accepts the marketing numbers, or at least knows how they were built.

Definitions first, then one model

I start with definitions, not tools: what a qualified lead, an opportunity and a marketing-sourced deal mean, agreed with sales and finance. Then I check that each lead keeps its source from the first visit to the closed deal, and repair the tracking where it does not.

With that in place, one model reports spend against pipeline and revenue by channel, with the unknowns shown as unknown instead of filled in. It is usually less flattering than platform attribution, and more useful in a budget discussion. The attribution OS and full-funnel analytics pages describe the build.

Where tracking cannot see a channel, such as word of mouth or AI assistants, a "how did you hear about us" field and sales notes fill part of the gap. The tracking evidence checker linked below is a quick first test of your setup.

Measurement work in two published cases

At Riverside.fm my case study describes building one attribution model the finance team could defend in front of a board, and reporting MRR attributed to paid against real post-trial paying users. At cnvrg.io, as my case study records, there was no marketing-attributed pipeline reporting that finance trusted when I started, and the work began with shared pipeline definitions.

I do not present the measurement work as the cause of the revenue results. The cases describe what was built and what it let the team see.