Approach
How we work
Four questions. One pod. Three decision gates.
Most AI pilots stall because nobody asked the right questions first. Our approach is built so the proof of value is a rehearsal for production, and nothing has to be rebuilt.
The blocker is rarely the technology
Enterprise AI investment is at record levels. Real adoption is not. The gap sits between leadership ambition, workforce trust, functional readiness and governance.
Strategy firms produce decks. Technology vendors sell tools. Training providers deliver knowledge nobody applies. Novus AI advises, trains, governs and delivers. From strategy to production. One accountable team.
Only 36% of employees feel adequately trained for how AI has changed their role.
BCG, AI at Work, 2026
63% of organisations report no EBIT impact from their AI use.
McKinsey, State of AI, 2026
What private equity asks us for
An AI plan tied to EBITDA, a workforce that uses the tools it already pays for, and evidence a buyer or auditor will accept at exit.
Five things we hear on first calls.
- “Our AI pilots never reach production.”
- “We bought AI licences and nobody uses them.”
- “The board wants an AI plan with numbers.”
- “The regulator is asking about AI governance.”
- “We can’t hire a Head of AI fast enough.”
Four questions before we build anything
Most AI pilots stall because nobody asked these questions first. We ask them of every use case, with your team, and we only build where all four answers are yes.
Will your people use it?
Tested against the workflow it lands in, with the people who will run it every day.
Can it be built well, inside your infrastructure?
Data quality, access and controls checked before a line of code is written.
All four, or we do not build it. If a use case fails one, we tell you.
Does the business need it now?
Ranked against every other use case on manual density, data, risk, labour and speed to value.
Will it show on the P&L?
A baseline, a target and a named P&L line agreed before the build starts.
One pod. One owner. One outcome.
Every engagement runs on a dedicated pod of 4 to 6, not a rotating bench. The people who scope the work deliver it, tracked against the metrics agreed on day one.
- LeadAI Transformation DirectorOwns the client relationship, the solution architecture and the delivery outcome.
- 2 to 3Data scientistsModel development, ML engineering and data pipelines. Production-grade, not prototypes.
- BridgeBusiness analystRequirements, process mapping, change management and stakeholder alignment.
- CadenceProject managerMilestones, budget, risk and reporting, from the first sprint to handover.
Scope to production in 30 to 90 days
Production requirements are set as acceptance criteria in week one. The proof of value is a rehearsal on a small scope, so nothing has to be rebuilt.
- Stage 01
Scope
The workflow quantified from your own volumes. Thresholds, data in scope and a named owner agreed before any build.
- Stage 02
Build
Short cycles on real data with your experts alongside the pod. Tested on held-out real cases.
- Stage 03
Production
Integration, monitoring, review workflow and rollback signed off. Live in your own infrastructure.
- Stage 04
Own
Code, configuration, evaluation set and runbook transfer to your team, with training for the people who run it.
Baseline, thresholds, data in scope and named owner agreed.
Measured at real volume against the week-one thresholds.
Integration, monitoring, review workflow and rollback signed off.
Next step
See the approach on your own use case
One value stream, one pod, success metric agreed up front.