Cegid

Cegid Loop exposes several object types: client files (DossierClient), missions (Engagement), accounting entries (EcritureComptable), third parties (Tiers), collaborators (CollaborateurCabinet), and accounting balances. This integration covers two objects: client files and engagements. Accounting entries, which would represent the deepest layer of production activity, were excluded because the entry object carries no direct user attribution field and the endpoint imposes a hard cap of one hundred records per request, making bulk attribution unreliable. Third parties and balances were excluded for similar attribution and aggregation reasons. Six KPIs were retained, selected against three criteria: ability to attribute to a collaborator by email, resistance to gaming, and balance between leading and lagging indicators.

Client portfolio: reading workload through stock and flow

Three KPIs cover the DossierClient object. Active Client Files measures the total number of files in active status assigned to a given collaborator at any point in time. Client Files Opened measures the number of new files created and attributed to that collaborator in the rolling period. Client Files Closed measures the number of files that exited the portfolio over the same window.

Reading the three indicators together reveals the portfolio dynamic that no single indicator exposes. Active Client Files is a stock: it reflects the current state of a collaborator's workload but says nothing about how it is evolving. Client Files Opened and Client Files Closed are flows: they explain how the stock changes. A collaborator whose Active Client Files count remains flat may be in a stable steady state or in a situation where inflows and outflows cancel each other out at a high rate of turnover. A rising stock paired with low Client Files Closed suggests accumulation without resolution. A falling stock paired with high Client Files Opened and high Client Files Closed indicates rapid cycling, which may signal efficient throughput or superficial processing depending on context.

Client Files Opened also serves as an early signal for capacity planning. An acceleration in new file attribution to a specific collaborator, without a matching increase in Client Files Closed, is a leading indicator of overcommitment before it becomes visible in engagement delays. The combination of the three stock-and-flow indicators provides a manager with the information needed to rebalance portfolios before pressure materialises.

Mission delivery: throughput and velocity of engagements

Three KPIs cover the Engagement object. Open Engagements counts the missions currently in progress for a collaborator. Engagements Closed counts the missions completed and closed within the rolling period. Average Engagement Duration measures the mean time in days between the start date and the close date of completed missions.

Open Engagements and Engagements Closed are in a natural tension that makes each more informative when read alongside the other. A high Open Engagements count with a low Engagements Closed count in the same period signals that work is accumulating without resolution: the collaborator is taking on new missions faster than they are completing existing ones. The inverse pattern — low Open Engagements with high Engagements Closed — reflects a collaborator who is efficiently clearing work and whose pipeline may need replenishment. Neither condition is self-evidently positive or negative without the other.

Average Engagement Duration adds an efficiency dimension that the two volume indicators cannot provide. A collaborator with a high Engagements Closed count and a long Average Engagement Duration is processing missions of genuine complexity or chronically delayed ones. A low Average Engagement Duration paired with a high closure count may reflect either genuine efficiency or superficial mission scope. The pairing of Average Engagement Duration with Engagements Closed also guards against a specific gaming risk: a collaborator under pressure to improve closure counts could close missions prematurely. An implausibly short Average Engagement Duration reveals that pattern and flags it for review.

Scope and limits of the integration

Cegid Loop records file assignments and mission lifecycle events, but does not measure the quality of accounting work produced. A closed engagement may correspond to a thorough statutory audit or to a routine bookkeeping task; the API data does not distinguish between the two. Mission duration is an efficiency proxy, not a quality measure: a short engagement may reflect either competent execution or an underscoped mandate. The accounting entry layer — where production activity is most granular — remains outside this integration because the entry object does not carry a collaborator attribution field, and the hundred-record cap on the entry endpoint makes bulk extraction via standard REST pagination impractical.

Furthermore, Cegid Loop only captures the activity that is recorded within the platform. Informal client interactions, work performed outside the system, and partial engagement completion are not reflected. The reliability of these KPIs depends on the discipline with which collaborators and managers maintain file and mission records: consistent use of the maitreDossier field and timely engagement status updates are prerequisites for the indicators to carry interpretive weight.