CIO briefingERP AI (any ERP)Board and executive reporting

How to present ERP AI to the board: what to include and what to leave out

Question

how to present ERP AI to the board

Also searched as

ERP AI board presentation outline
what should a CIO tell the board about AI on ERP
board slide deck AI initiative manufacturing ERP

Short answer

A board does not need a model tour. It needs one decision, a bounded first step, the risks and who owns them, a cost envelope, and the evidence you will bring back. Structure the session as decision, exposure, pilot, money, and proof, in about eight slides.

Applies to: Any ERP (SyteLine, NetSuite, SAP, Dynamics, Epicor, LN, M3, JD Edwards). Mid-size and enterprise manufacturers with a board or audit committee.

Build the board session in this order

1Open with the decision you need, in one sentence: for example, approve a 12 week read-only pilot on the production ERP with a spend ceiling and a go or no-go review date.
2State the business problem in the board's language: planner hours spent on expedite reports, quote turnaround, month-end close days, audit evidence requests. Use your own baseline numbers, not market statistics.
3Show the exposure slide: where data goes, which models run where, who can see what, and how ERP roles are respected. Boards and audit committees ask this before they ask about value.
4Show the pilot scope: one or two use cases, read-only access, named business owner, success measures fixed in advance, and what is explicitly out of scope (no write-back, no customer-facing output).
5Show the cost envelope as ranges with line items: licences or GPU, integration, partner time, internal time, and ongoing run cost. Say which items are one-off and which recur.
6Show the stop conditions: the results at which you will end the pilot, and the cost of exiting (usually small if the system is read-only and the models are open-weight).
7Close with the evidence calendar: what you will report at 30, 60 and 90 days, and who signs off on moving beyond pilot.
8Prepare an appendix, not slides, for architecture, vendor comparisons and model details. Hand it out only if asked.

What directors actually want to know

Directors are generally asking four things under different wording: are we exposed, are we falling behind, what does it cost, and who is accountable. If a slide does not help answer one of those, it is probably there for the presenter's comfort. Drop it.

Exposure is the question that ends careers when it goes wrong, so lead with it. Explain in plain terms whether ERP data leaves your network, whether it is used to train anyone's model, whether prompts and answers are logged, and how a user's ERP role limits what the assistant can retrieve. If you are in aerospace, defense or export-controlled work, say how the design keeps controlled technical data inside your boundary, and let your compliance lead state your own obligations rather than claiming a certification on the vendor's behalf.

Falling behind is the second. Avoid fear framing. A calm line such as 'the ERP vendors are shipping assistants, and we want to decide deliberately whether we use theirs, a private layer, or both' lands better than competitor anxiety, and it is more honest.

The slides, one by one

Slide 1 is the ask. Slide 2 is the problem with your baseline. Slide 3 is the approach in one diagram: users, assistant, retrieval layer, ERP read-only connection, models on your infrastructure or a private cloud. Slide 4 is controls: authentication, role mapping, query visibility, logging, human approval for any write. Slide 5 is the pilot plan and success measures. Slide 6 is cost. Slide 7 is risks with owners. Slide 8 is what you will report back and when.

Resist a vendor logo slide. Boards read logos as endorsement, and a pilot has not earned one. If the question 'who else does this' arises, answer from references you can actually name with permission, and say plainly when evidence is thin.

Include one honest sentence on limits: the assistant can be wrong, it will show the query or source so a person can check, and the first release will not replace any role. That sentence usually reduces resistance more than any benefit slide.

Questions to rehearse before you walk in

Expect: what happens if the answer is wrong and someone acts on it; who is liable; can the vendor see our data; what if the model provider changes terms; how is this different from the copilot our ERP vendor is selling; and what does it do to headcount. Have a two sentence answer to each and a named owner for the follow-up.

On the vendor copilot question, do not dismiss it. The vendor's own assistant is often enough for single-system, in-product tasks. A separate layer earns its place when you need cross-system answers, on-prem or air-gapped operation, or freedom from a single vendor's roadmap. Say that, and say you will compare them on your own data during the pilot.

On headcount, be direct. Pilots of this kind usually shift time from report building and lookup to review and decisions. If your plan includes headcount change, that is a separate conversation with HR and should not be smuggled into a technology approval.

Board one-pager skeleton
1. Decision requested + date
2. Baseline problem (your numbers)
3. Scope: use cases in / out
4. Controls: data location, roles, logging, approval
5. Cost envelope: one-off vs recurring (ranges)
6. Success measures + stop conditions
7. Risks + owners
8. Reporting calendar (30/60/90 days)

Common pitfalls

!Leading with model names, parameter counts or a demo. Directors remember the demo failing, not the benefits.
!Quoting industry productivity statistics you cannot source. Use your own baseline measurements or leave numbers out.
!Asking for a transformation budget instead of a bounded pilot. Large asks invite delay and committee redesign.
!Skipping the data location answer. If you cannot say in one sentence where prompts and ERP data go, you are not ready to present.
!Letting the vendor present. A supplier in the board room turns a governance discussion into a sales call.
!Promising write-back automation in the first phase. Keep the first phase read-only and add approvals later.

How an ERP-grounded AI assistant handles this

Before the board meeting, a grounded assistant such as ERPray can help assemble your baseline: how many expedite or shortage reports planners run, how long open purchase orders sit past due, how many days close takes. Because it shows the query behind each number, the figures on your slides are traceable rather than quoted from memory. It is a research aid for your own preparation; the framing and accountability stay with you.

See ERPray

Frequently asked questions

How long should an ERP AI board presentation be?

Fifteen to twenty minutes of content with at least as much time for questions is typical. Eight slides is enough if each carries one idea. Put architecture, vendor comparisons and model details in an appendix. Directors rarely want depth on technology; they want confidence in decision, control and cost.

Should we show a live demo to the board?

Usually no. Live demos fail for reasons unrelated to the idea, and they pull attention to the interface. A short recorded walkthrough of a real question answered on your own data, with the query visible, is safer. Better still, offer a hands-on session to interested directors afterward.

What is the single biggest risk to name?

For most manufacturers it is data exposure, followed by acting on a wrong answer. Name both, and describe the control for each: data stays in your environment under ERP role limits, and answers show their source so a person verifies before acting. Naming risks yourself builds credibility.

Do we need board approval for a pilot at all?

Often not for a small read-only pilot, which management can approve within existing authority. Go to the board when the spend, the data classification or the strategic commitment crosses your governance thresholds, or when the audit committee expects to hear about new data processing. Check your own delegation of authority.

CIO briefing

Do You Need an ERP AI Governance Board?

Yes, once more than one AI use case touches ERP data - even a lightweight, four-person committee that meets monthly beats no governance at all. Its job is narrow: approve which ERP data an AI tool can touch, set the review cadence for accuracy and access, and own the kill switch if something goes wrong. It should not be a bureaucratic gate that slows every pilot to a crawl.

CIO briefing

How Much Does ERP AI Actually Cost?

A scoped pilot against one ERP module typically runs 15,000 to 50,000 USD; a single-department production deployment runs 50,000 to 150,000; a multi-module enterprise rollout runs 150,000 to 500,000 or more, plus ongoing hosting and usage costs. Native vendor copilots are usually priced per active user per month on top of existing licensing, not included free.

CIO briefing

ERP AI Vendor Due Diligence: What CIOs Should Actually Check

The single highest-signal question in ERP AI vendor due diligence is 'show me it answering a real question against our actual ERP data, live, not a demo dataset.' Beyond that, focus diligence on data grounding method, security and access model, exit/portability terms, and references from customers on your specific ERP platform - not on feature checklists, which most vendors can match on paper.

CIO briefing

Measuring ERP AI ROI After Go-Live

Measure ROI with three metric families - time saved on named tasks, error or rework reduction, and adoption depth - tracked against a pre-go-live baseline you captured before the tool launched. Most CIOs skip the baseline and then cannot prove anything six months later, so capture it in week one even if the AI tool is not fully ready.

CIO briefing

ERP AI budget line items: what to put on the sheet before finance asks

Budget ERP AI in eight groups: infrastructure, software and model licences, integration, data preparation, security and compliance, people and change, run cost, and contingency plus exit. The costs that surprise teams are data cleanup, internal time and ongoing evaluation, not the model itself.

CIO briefing

ERP Upgrade or AI First? A CIO Decision Framework

In most cases, add a grounded AI layer on top of your current ERP first, because it proves value in weeks and shows exactly what an upgrade would need to fix. Reserve a full ERP replacement for cases where the platform itself is end of support, unsupported, or structurally blocking the business, not simply because it feels dated.

AI for ERP