Analysis, grounded.
Practical guides to the analysis work that decisions ride on, and why an AI analyst that grounds every figure in your own data changes the job.
Why AI tools invent numbers, and how source-grounding stops it
Ask an AI for analysis and it often hands back a number that looks exactly right and is completely wrong. Here is why that happens, and what actually stops it.
ReadFrom a raw spreadsheet to a finished analysis
Most of the time you spend on an analysis goes to grunt work, not thinking. Here is the method that gets you to a finished, grounded answer faster.
ReadBudget variance analysis: what actually moved versus plan
A variance number on its own tells you nothing. The skill is decomposing it into price, volume and mix, then explaining what truly moved and why.
ReadWriting an investment memo that stands up to scrutiny
A memo wins or loses on whether the numbers survive a hard question. Here is the structure, the sourcing discipline and the failure modes that decide it.
ReadMarket-entry assessment: which frameworks, in what order
A market-entry assessment is not a stack of frameworks. It is a sequence of questions, each answered with the lens the data supports.
ReadHow to run a quarterly business review that changes what people do next
Most QBRs are a status readout nobody acts on. Here is how to run one that ends in a few clear decisions, with an owner and a date on each.
ReadEvaluating supplier proposals: a scoring approach that survives scrutiny
Gut feel and price-only comparisons collapse the moment someone challenges them. Here is a scoring method that holds up, plus where an AI analyst grounds it in the actual proposals.
ReadReading a data room fast: distilling due diligence into a decision
A data room holds hundreds of files and one decision. Here is how to read it by risk, surface the numbers that matter and land a clear call without writing a hundred pages nobody reads.
Read