AI readiness for finance functions is a structured diagnostic that determines, pillar by pillar — sponsorship, champion, data access, governance, training, scoring, project cadence — where the team stands (Green/Amber/Red) and which pillar needs strengthening first for AI adoption to land. For finance teams: most functions are mixed, not uniform — the assessment prevents three months burned on the wrong pillar.
Before you start an AI roadmap in a finance team, you have to know where you stand. Most finance functions are not Green or Red across the board — they're mixed: sponsorship on Green, data access on Red, skills on Amber. A readiness assessment makes that difference visible and stops you from burning three months on the wrong pillar.
For finance teams the generic SME diagnosis mostly works, but two pillars get a finance-specific reading: data access (where the accounting system always sits at the center) and governance (where audit requirements weigh heavier). This piece walks through the seven pillars in their finance version.
Why diagnose first
AI adoption projects in finance rarely fail on one big thing; they fail at the weakest link. You can have a sponsoring CFO and an enthusiastic controller, but if there's no AI policy stating what's allowed with customer numbers and no IT decision on the Exact connection, no pilot will clear compliance. The reverse: excellent tools and data access without leadership commitment yields three experiments nobody scales.
A readiness assessment forces you to look at every pillar systematically rather than your favorite pillar (usually the technical one). It's a counterweight to bias.
The seven pillars in finance form
- Leadership sponsorship — CFO or finance manager actively owns it.
- AI champion in finance — someone with time and a mandate.
- Bottom-up culture — the team dares to experiment with AI.
- Tools, data access, and AI policy — Exact connection, GDPR approach, tier policy.
- Training and hands-on practice — not only power users.
- Project cadence — ownership per pilot, rhythm of check-ins.
- Use-case prioritization — a process to bring in ideas and choose between them.
For each pillar, ask three questions: is this explicitly assigned, is it actively driven, and are there concrete results in the last 90 days? Three yeses = Green. One or two yeses = Amber. Three nos = Red.
What Green, Amber, Red mean
Green per pillar
Owner named, time or budget allocated, visible outcomes in the past three months. For sponsorship in finance: the CFO has explicitly discussed AI in the last MT meeting, there is a budget (even if only protected time for the controller acting as champion), and decisions have been made based on pilot results.
Green doesn't mean "done." It means "enough foundation — keep maintaining it."
Amber per pillar
Something exists, but it's fragile. The CFO mentioned it once, but barely talks about it. The champion has enthusiasm but no time. The policy is in draft but not communicated. Something is happening, but it depends on individuals.
Amber is the most common state in SME finance — and the most dangerous, because it feels like progress. An Amber pillar not actively pushed to Green slips back to Red within a quarter.
Red per pillar
Nothing assigned, no owner, no budget, no policy, no activity. Staff don't know whom to approach, managers don't know what to ask for, pilots die before they start.
Red pillars in finance are surprisingly often not the technical part. "Project cadence" and "use-case prioritization" are typically Red: no process to pick up finance ideas, no rhythm to measure progress, so it stays at occasional experimentation.
Checklist per pillar — finance version
Three ticks = Green, one to two = Amber, none = Red.
Leadership sponsorship
- A named sponsor (usually CFO or finance manager).
- An explicit AI budget for finance, however small.
- The sponsor has visibly communicated about AI in finance in the last 90 days.
AI champion in finance
- One or more champions named within finance.
- Protected time (half a day or a full day per week).
- A mandate to work with IT, legal, and the other finance roles.
Bottom-up culture
- Education paths available (internal or external).
- A ritual (weekly win, demo, Teams channel) that is running.
- Explicit permission to experiment — finance staff otherwise won't.
Tools, data access, and AI policy
- The AI tools in use are licensed for business (no consumer accounts for customer numbers).
- Access to the accounting system for relevant AI tools is arranged (or being prepared).
- There is a communicated AI policy document for finance, including tier classification of customer data.
Training and hands-on practice
- A training roadmap for the whole finance team (not only power users).
- Recurring practice moments (monthly hands-on session, office hours).
- Shared learning (team demos, peer review, knowledge sharing).
Project cadence
- Every finance AI pilot has a named owner.
- Every pilot has a deadline and a success criterion.
- Progress is discussed at least monthly.
Use-case prioritization
- A process for submitting ideas (form, channel, list).
- Criteria for prioritization (impact in hours or errors, feasibility).
- A prioritized roadmap of at least 3-5 finance use cases.
What you do with the result
The goal isn't a total score. The goal is to identify one priority for the next 30-90 days. Pick the weakest pillar that also offers the biggest lever for finance, and take a concrete first step there (see 30/60/90-day plan for AI in finance).
Typical patterns from finance assessments:
- Everything Amber, nothing Green: the classic "we're working on it" finance function. Pick sponsorship or project cadence — they pull the others along.
- Tools Green, the rest Amber/Red: tech-first, people-later. Invest in finance-specific training and rituals, not another license.
- Sponsorship Green, champion Red: the CFO wants it, but nobody has time to pull. Name a controller or senior staff member with explicit time.
- Everything Red: start with sponsorship. Without it, no other pillar ever reaches Green.
Repeat the assessment every quarter. Movement from Red to Amber to Green is slow and incremental; without measurement it looks like nothing is happening, with measurement you see progress.
Audit-grade perspective
In a finance context a readiness assessment has a second function alongside diagnosis: it is evidence inside your internal control framework. An external auditor asking "how have you embedded AI in your finance process responsibly" wants to see a trajectory, not an ad-hoc tool purchase. A quarterly assessment with data classification, policy status, and pilot cadence is exactly that trajectory.
Saldus in practice
A readiness assessment is usually a two-week exercise followed by a 30-60-90-day roadmap. Tools only come into play after the roadmap. The Start 2 Scale assessment (/assessment) uses the same seven pillars in finance form and delivers a concrete first step per pillar, plus three prioritized use cases to start with — independent of which platform or vendor enters the picture afterwards.