The AI Close.
Your AI, run like your books.
Every serious business closes its books once a month — what came in, what went out, what it means, what's next. Almost no business does the same for AI. So AI drifts: a tool here, a dead pilot there, a staff member quietly pasting customer data into ChatGPT. The AI Close is a monthly rhythm that puts one owner in charge of all of it — the way a fractional CFO owns your numbers.
The problem it solves
Most companies don't have an AI problem. They have an ownership problem.
95% of company AI projects fail to deliver a return — and the research is clear the cause is organisational, not the technology. No owner, no roadmap, no go/no-go. (MIT NANDA, 2025.)
You don't need more tools. You need someone accountable for making AI live, killed, or clearly next — every month, in writing, in language an owner can read.
The method — one monthly rhythm
Each month, Dagaz runs your AI Close and hands you one owner-readable pack. Five questions, always answered:
| Question | What it asks | What it means |
|---|---|---|
| LIVE | What's actually running in the business? | Real use, real staff or customers — not demos |
| PARKED | What's on hold, and why? | Good ideas waiting on data, budget, or sequence |
| KILLED | What did we stop, and why? | Dead pilots closed in writing — like writing off bad inventory |
| COST | What are we spending on AI this month? | Every tool, every trend, overlap cut |
| NEXT | What are the next 30 days? | The two things that matter, with owners |
That's it. No dashboards you'll never open. One page a month that answers "what's our AI plan?" in under two minutes.
The artifacts you own
The Close produces a small set of durable documents — you own them, not us:
- AI Readiness Scorecard — where you stand today, and the gaps, scored.
- Use-Case Portfolio — every AI opportunity, ranked by impact × ease × data-readiness × risk. The two we pursue; the rest parked with reasons.
- The monthly AI Pack — the five-question close, in writing.
- Green / Yellow / Red data policy — a one-page rule so staff know what may never go into an AI tool (your PDPL front line).
How an engagement moves
AI leadership isn't a project that ends — but it isn't forever-dependency either. It moves through three honest stages:
- Cleanup (month 1–2): stop the bleeding. Map the tools and spend, kill the waste, name the owner, set the data rules, pick the first two use cases.
- Build (month 2–6): get real things live — directed by us, built as fixed-scope sprints, quality-gated, with runbooks and human review.
- Institutionalise (month 6+): governance that runs itself, a trained internal champion, and a roadmap that keeps regenerating as the models change. The value shifts from fixing to stewarding.
We document everything and train your people. The point is an owned, capable AI function — not a consultant you can never leave.
What The AI Close is not
- Not a chatbot you buy once and forget.
- Not a workshop or a slide deck.
- Not "#1 on Google," not guaranteed revenue, not "replace half your staff."
- Not unlimited engineering — build is scoped and quoted separately, so the leadership stays sharp and the economics stay honest.
Proof, honestly
Dagaz runs its own portfolio of live businesses almost entirely on AI systems. We sell what we run on. We don't publish inflated ROI numbers or borrowed logos — when we cite a result, it's real and checkable. Until then, we'll show you the method and let a fixed, guaranteed Reality Check prove it on your business, not ours.