Ed, who writes this blog

Twenty-five years of building management reports, and of paying for the ones that turned out to be wrong. Every post here is mine, written in part with AI.

Where the posts come from

Over a twenty-five year career I have started companies, worked as a management consultant, owned a multi-million-dollar P&L, and led business intelligence and strategy for large enterprises. Four different vantage points on the same problem, and the problem was always the same one: the number in the deck did not match the number in the system, and finding out why took three days.

As a consultant I built utilization and realization reports for firms who could not tell me, without a week's work, which of their clients actually made money. As a P&L owner I was on the other side of that — waiting on a pack that arrived on the eleventh working day, by which point the decision it was meant to inform had already been made. Running BI for large enterprises, I watched teams spend more effort reconciling two versions of revenue than analysing either one.

None of that is unusual. It is the normal condition of a business that runs on spreadsheet exports, which is most of them. It is also why the posts here dwell on the parts other guides skip: what the denominator should be, which column your export actually writes, and which direction an error pushes the number.

Why I build Quiriz

Quiriz started as a side project to answer questions about my own files without rebuilding the same tab every month. It turned into something that saved real time, so I shipped it. It reads messy exports as they come — merged cells, headers three rows down, a total row in the middle — and answers questions about them in Excel, Sheets and Slack.

I am candid on this blog about what it does not do, because the alternative is that you find out after signing up. When a post says Quiriz cannot compute something from a given file, that is not modesty — it is the actual limit, and knowing it up front is worth more than a demo.

How I write these

• Written in part with AI, and here is which part. I use AI to draft and tighten prose, and I edit what comes back. What it does not do is supply the experience — every field note on this site, and every claim about what I have seen firms do, comes from me. If a passage says I sat in that meeting, I sat in it.
• Every method gets tested before it is published. If a formula is in a post, it ran in a real workbook first.
• No invented statistics. Numbers here are either sourced and linked, or they are labelled as my own experience rather than research.
• Competitor claims carry a verification date, because they go stale in weeks.
• Corrections are welcome and get made. If something here is wrong, tell me and I will fix the page rather than quietly leave it.

Start with utilization rate in Excel or why your team gets different numbers — those two cover most of what I learned the expensive way. The rest is in the blog.