Skip to content

For CFOs and finance leaders

AI readiness for CFOs: questions to ask before funding more AI

Finance feels a weak operating model first: in the close, in reconciliations, and in meetings spent agreeing whose number is right. The book argues that these are symptoms of unclear definitions and missing owners rather than missing systems, and that funding another platform before fixing them buys the same confusion, faster. That makes the questions a CFO asks before approving AI spending unusually important.

Run the AI-Readiness ScorecardGet the book on Amazon

Kindle edition · 99 pages · a 90-minute readRead a free sample on Amazon

From the book

What the book shows

  1. 01

    One word, two meanings

    In the book’s first (anonymized) case study, Finance and Sales described the same customers in incompatible ways. Each view made sense inside its own function; together they left the company unable to make a decisive move when a logistics disruption demanded a fast reallocation of capital and inventory.

    “…to Finance, a customer was a consolidated legal parent with a credit limit; to Sales, a customer was hundreds of regional buying offices and accounts transacting independently.”
    Chapter 19Case Study: The Company That Looked Ready
  2. 02

    Define it once, at the top

    The fix in that case was not a new platform. Leadership set core metric definitions once and stopped renegotiating them. The book reports that the time to finalize critical capital-allocation decisions then fell from eleven business days to under two hours — one company’s result, not a benchmark.

    “…gross margin, for instance, was defined at the boardroom level to include localized promotional discounts, ending the practice where regions calculated it differently to protect local bonuses.”
    Chapter 19Case Study: The Company That Looked Ready
  3. 03

    System programs are easy to fund — that is the trap

    The book argues that consolidating onto one ERP or one platform is rarely the prerequisite it appears to be. Big system programs win budgets because they come with approvals, milestones and vendor accountability; fixing ownership and decision rights is harder, so it gets postponed.

    “Too often, system standardization becomes a proxy for avoiding the harder work of governing the business itself.”
    Chapter 10Meaning, Not Machinery

Where your numbers really come from

The book’s simplest test for finance: when a number is wrong, stop asking which system is wrong and ask which business process produced it and who owns that workflow. It also warns against forcing performance data into legal-entity boundaries, because operations rarely follow them. In its merger case, once definitions and owners were fixed, monthly global close went from twenty-two business days to four — again, one anonymized company.

“Forcing data into legal boundaries distorts performance, dilutes accountability, and destroys comparability.”
Chapter 8The Operating Tree

Questions to ask

Questions to ask before approving the next AI or data budget:

  1. 1Do revenue, margin and customer mean the same thing in every region and function — and who owns the differences?
  2. 2In our last few leadership meetings, how much time went into reconciling numbers instead of deciding?
  3. 3Which budgets assume we must consolidate systems first — and is that dependency real?
  4. 4Before scaling an AI initiative: which decision does it support, and who owns the outcome?
  5. 5Are we reporting performance by legal entity where the business actually runs by operating unit?
  6. 6What is reconciliation costing us that never appears as a line item?

Where to start in the book

  • Chapter 19

    Case Study: The Company That Looked Ready — definitions, owners and the eleven-days-to-two-hours capital decision.

  • Chapter 20

    Case Study: When Two Strong Companies Could Not Become One — the merger savings and the faster close.

  • Chapter 10

    Meaning, Not Machinery — why one ERP is not the prerequisite it seems.

Get the book on Amazon

Kindle edition · 99 pages · a 90-minute readRead a free sample on Amazon

Book cover: Your data isn't broken. Your operating model is.

Also see

All roles: who the book is for →

About this page: our reading of Your data isn’t broken. Your operating model is. by Adam Ayres and Behrang Karimi, for CFOs and finance leaders. The book itself is written for the leadership teams of complex enterprises. Quotes are word for word from the book; its case studies describe anonymized companies, and their results are not benchmarks.