Blog

Notes from the work

Shorter pieces than the research: something we learned delivering an engagement, or a buyer question that came up often enough to answer properly. No schedule, no trend pieces, and the same editorial policy as everything else on this site.

Get new work by email

Roughly monthly, only when something is actually published. No sequence and no drip.

1 / 3
No content calendarMethod statedCorrections dated
  1. AI Agents in ERP, Hype or Real

    A sober look at where AI agents in ERP are production-ready for finance teams in 2026, where they still fail, and the controls that separate the two.

    Read the post
  2. A chart of accounts migration is not a copy job

    Copying a legacy chart of accounts into a new ERP preserves the same reporting problems the implementation was supposed to fix. A migration should start with the reports you need, not the export file.

    Read the post
  3. Closing faster is mostly a reconciliation problem

    Teams trying to shorten the close usually attack the last three days. The time is almost never there. It is in the accounts that are only reconciled once a month, and the reason that is hard is not effort.

    Read the post
  4. What an agent structurally cannot do inside a ledger

    Our agents run at 94% accuracy on the workflow they are best at and 66% on the one they are worst at, measured across 23 customers. The gap is not a training problem. It is a description of which tasks have a checkable answer and which do not.

    Read the post
  5. "Nobody can decide" is a schedule item, not a personality problem

    Requirements that never settled (31%) and no internal owner with authority (19%) are the two largest primary causes across the 41 stalled implementations in our sample. Together they are half of it, and both are usually described afterwards as a people problem. They are a design problem.

    Read the post
  6. Bad data is always discovered during the load

    In 24% of the 41 stalled implementations we have been called into, the primary cause was data that turned out to be worse than anyone knew — and it was almost always found during the load rather than during the diagnostic. The timing is the problem, not the data.

    Read the post

Start with the pages that cost us something.

The manufacturing guide telling you to buy a competitor, the directory where we rank fourth, and the 66% benchmark.