Copieux signature framework
Systems, data and process before AI.
A practical method for helping lean HR teams move from manual work and uncertain information to trusted, AI-ready people operations. It gives you a clear answer when leadership asks for an AI plan before the foundations are stable.
For CHROs, HR Directors and Heads of People in scaling organisations.
The principle
The framework names the work Copieux does across projects: understand the complexity, strengthen the systems, make people data reliable, simplify processes, then select AI use cases the organisation can support.
AI is not recommended just because a tool is available or the board has asked for a plan. The data, process, ownership and human review need to be ready for the use case. If they are not, the roadmap says what to fix first and what that will make possible.
The method produces a scored view of the foundations, a maturity level, a prioritised set of fixes and a clear basis for measuring progress.
The five stages
Complexity gives the work its context. Systems, data and process determine whether the foundations are ready. AI readiness follows those checks.
STAGE 01
Set the context for the work.
Map the countries, entities, payrolls, contract types, growth plans and system changes the HR team supports. Complexity shapes what a practical operating model needs to handle, but it does not set the maturity level.
Typical outputs: Complexity map, entity and policy register, and a clear view of capacity constraints.
STAGE 02
Check whether current systems support the work.
Review the HRIS capabilities already owned, the configuration, workflows, integrations, self-service and the spreadsheets or side tools that fill gaps.
Typical outputs: System use and configuration review, integration map, and named ownership for system changes.
STAGE 03
Establish whether people data can be trusted.
Test key definitions, data ownership, quality, reconciliation with finance and the effort required to answer leadership questions.
Typical outputs: Agreed data definitions, priority data fixes, a reconciliation routine and a core metrics pack.
STAGE 04
Find where manual effort and rework remain.
Map how core HR processes actually run, including handoffs, approvals, duplicate entry, delays and workarounds. Simplify and assign ownership before automating.
Typical outputs: Prioritised process maps, clarified ownership and a practical improvement plan.
STAGE 05
Choose a useful AI use case the foundations can support.
Only consider AI when the relevant data is owned and reliable, the process is stable enough, human decision rights are clear and the business question is specific.
Typical outputs: A prioritised use case, governance requirements, measures and a responsible pilot plan.
Maturity levels
Each question is scored from 1 to 4. The overall level is set by the lowest score across Systems, Data and Process. Complexity informs the interpretation, rather than raising or lowering the level by itself.
LEVEL 1
Spreadsheets carry the real work. HR records and reports are difficult to trust.
LEVEL 2
Core data sits in the HRIS and key processes are documented, though workarounds remain.
LEVEL 3
People data has a reliable source, workflows run in the system and HR and finance numbers align.
LEVEL 4
Leaders use insight directly, routine work is handled well and HR time is available for business priorities.
The AI foundations gate
The 25 questions are scored from 1 to 4. Before recommending an AI use case, the Data stage must average at least 3.0 and Process must average at least 2.5.
If the foundations are below the gate
The plan prioritises the data and process fixes first, and names the AI use case those changes could make possible. That gives leaders a practical sequence they can explain and measure.
Used across Copieux work
The framework informs the Copieux diagnostic, foundations reviews, system and data improvements, process work and AI readiness planning. It gives projects a consistent way to assess the current state, agree what matters and measure what changes.
Score the foundations and make complexity, bottlenecks and ownership visible.
Prioritise system configuration, data quality and process changes that release capacity.
Choose an appropriate use case, define safeguards and track capacity and quality.
Put the framework to work
Start with the Work Design Readiness Assessment for a first view of where your foundations need attention.
Find your capacity gaps