Engage · Engagement
Scale Inversion Review and Assurance
Design the transformation. Protect its intent through delivery.
An independent review that retargets a material AI-era transformation, and assurance that holds that position while the work runs.
The problem
The people delivering the transformation also report on whether it is working.
A transformation large enough to matter has an execution architecture that is itself a decision: how much of the work is scaled, by whom, and under whose control. The plan that was approved was usually shaped, priced and staffed by the people who will deliver it.
Nobody in that structure is behaving badly; the structure simply has no independent reading in it. Delivery organisations protect scope and momentum, programme leadership is measured on the plan it committed to, and a steering committee reviewing a status report cannot see the choices that were never put in front of it.
AI sharpens the problem. More execution capacity in fewer hands changes what should scale, so a delivery model designed around large teams can be the wrong architecture before the first milestone — and stays hidden until it is expensive to change.
You will recognise this if
- The programme was scoped and priced by the partner now delivering it.
- Status reporting shows progress against the plan, but nobody tests whether the plan is still right.
- The staffing model reflects a delivery provider’s default rather than your target-state economics.
- The intent that justified the investment is getting harder to recognise in the work.
Who it is for
- Executives accountable for a material AI-era transformation
- Boards and investment committees overseeing large programmes
- Transformation sponsors working with consultancies, SIs, vendors or mixed third-party execution
For initiatives large enough that their execution architecture is itself a decision, and where the people delivering the work also own the assessment of whether it is working.
What the engagement answers
- Is the plan still aimed at what the work now actually requires?
- Is the engagement model right for the outcome, or for the provider?
- Where does scale create value, and where is it a delivery default?
- Is the operating model the programme is building the one the business needs?
- Do the objectives and value model still justify the spend?
- What would a governed two-speed model look like for this programme?
- Who in governance holds an independent read, with the standing to be heard?
What you receive
Review is a moment; assurance is a position held over time. They are separable, because a programme can need one without the other.
01
Independent review of the transformation
The plan, engagement model, operating model, value model, priorities and execution architecture, assessed against what the work requires now.
02
Retargeting recommendations
What to keep, change or stop, and the decisions required to do it, stated plainly enough to act on.
03
Execution architecture options
Where scale is warranted and where a more concentrated core achieves the outcome, including a governed two-speed model where it fits.
04
Assurance position in governance
An independent seat, with a formal voice and, where needed, a formal vote, held while the work runs.
05
Intent, architecture and value tracking
A standing read on whether the programme still delivers what it was commissioned to change.
How it works
1. Read the transformation
The same evidence your delivery partners read: plans, contracts, staffing, governance and progress.
2. Test it against the outcome
Whether the plan, model and architecture still serve the intent and value that justified the investment.
3. Retarget
Recommend what changes, with the decisions and sequencing required.
4. Hold the position
Where assurance is engaged, participate in governance and keep an independent read for as long as the programme runs.
BdG is the trusted independent counterpart to the accountable decision-maker, reaching conclusions that are not the delivery partners’ to approve.
The boundary
This is
- ✓Independent review of plan, model and execution architecture
- ✓Assurance held inside governance
- ✓A trusted counterpart to the accountable decision-maker
- ✓A formal voice, and a formal vote where that is what it takes
This is not
- ✕A PMO
- ✕A shadow SI
- ✕A second delivery organisation
- ✕A replacement for your transformation lead
- ✕A transfer of your accountability
An independent counterpart who absorbs accountability has stopped being independent, and one who runs delivery is a second SI.
The thinking behind it · Perspective 2
Why AI Transformation Fails
AI programmes do not fail because the models are weak. They fail because the organisation was never redesigned to absorb what AI does to decision authority.
Its core thesis
- 01AI redistributes decision authority, and the organisation pushes back.
- 02Pilots succeed; scaling is where the resistance starts.
- 03Probabilistic systems break deterministic governance.
- 04The people asked to lead the change built the structures it must dismantle.
Where this connects
- How AI changes execution →
The mechanism: more execution capacity changes the value of coordination.
- Deciding what should scale →
The design response the review applies to your programme’s execution architecture.
Related perspectives
- Model Access Is Not a Services Moat →
Frontier models now reach every integrator through every cloud, and the labs and clouds are building delivery arms of their own, so an enterprise should choose whom to trust with its AI estate on accountability and independence, never on model access.
- The Ratio Is 7:1. Most Enterprises Invert It. →
AI scales only when roughly seven parts of the effort go into governance, controls and authority structures for every one part spent on new execution capability. Enterprises that invert the ratio buy capability they cannot absorb.
- You Cannot Fix Legacy Governance from Inside It →
Structural transformation led by the people whose authority rests on the legacy will stall; it needs an outsider with explicit authority over the operating model and over personnel, for a defined term, who then leaves.
Begin with the decision
Design the transformation. Protect its intent through delivery.
Bring the situation, the constraints and the decision you are facing.
The other engagements