Three sites, three ways of counting, one baseline.

The company had grown quickly across three plants. Leadership wanted a monthly view of how the business was performing and discovered that the same measure meant three different things depending on which site reported it. Some measures were not being calculated anywhere. Others were calculated from data nobody could point to.

This is the ordinary condition of a company that has grown faster than its reporting, and it is not a failing. It is what happens when every site solves the problem locally because there was never time to solve it centrally.

One definition per measure, agreed by the people who produce the number and the people who use it. A written method for each. An honest list of the measures the company cannot yet produce, and what is missing for each one. And a view that refreshes without anyone assembling it by hand.

1. Collect what exists

   Every measure already in use at every site, with the calculation actually being applied rather than the one on the slide.

2. Surface the disagreements

   Where two sites calculated the same measure differently, both versions were put in front of the people who use the number. Most disagreements resolved in one conversation once they were visible.

3. Write the single method

   One definition, one source, one calculation, one owner per measure, documented so a new controller can reproduce it.

4. Name the gaps

   Measures the company wanted but could not yet produce were listed with the specific data that was missing, rather than quietly dropped.

An integration office cannot report the return on anything until the numbers underneath it mean one thing. Standardizing the measures is not glamorous work and it is almost always the step that has been skipped. It is also what makes every later claim about value defensible rather than arguable.