Measurement plan first, marketing second

Here are three common ways to spend money on marketing without knowing what it returns. All three can be fixed within a few weeks.

One: the same thing is measured three times

A site that has existed for a few years often has three measurement tools, installed by three different people. One comes from the first agency, one from the ad platform and one from someone who wanted to try something. The tools count page views twice and conversions in different ways. The reports disagree depending on who pulls them, and decisions are made on the one that looks best.

Two: measurement gaps are undocumented

Visitor consent choices and technical limitations affect what your analytics tools can see. Document what is measured, what is missing and which comparisons the data supports. A lower number in a tool does not necessarily mean a channel has stopped producing business.

Server-side tagging changes where data is processed. It does not override visitor choices or automatically make missing visits measurable. Read Google’s explanation of consent with server-side tagging.

Three: the report says what happened, not what to do

A report with twenty charts and no recommendation rarely gets read. The person who has to decide skims through it and does what they did last month.

How to measure marketing with a measurement plan

The work starts with a plan where every important event has a name and a source. Duplicate tags are removed and the different roles of the data sources documented. Consent is decided before the measurement is built, and server-side tagging is assessed against requirements, costs and data handling. The report has three parts: what happened, why and what should be done.

The scope depends on the accounts and systems involved. What matters is doing it before the next campaign is bought, not after.

How this works in an engagement is described under the method, and what a measurement setup covers under web analytics.