60%
reduction in reconciliation time
Margin visibility extended to individual matter and fee earner level
Firm-wide rollout after initial two-practice pilot
Automated low-margin alerts enabling intervention on active matters
A Magic Circle law firm replaced fragmented, inconsistent margin reporting with a single governed profitability platform that decomposed matter margin into its underlying drivers: rate realisation, utilisation, leverage, write-offs, and lock-up. The work cut reconciliation time by approximately 60%, gave partners evidence-based pricing and staffing insight, and rolled out firm-wide after a successful pilot.
The Problem
The firm knew which matters were generating revenue. What it could not reliably answer was which matters were actually making money, and why.
The Solution
We consolidated matter, time, billing, and resourcing data into a single governed data layer and built profitability reporting that decomposed margin into its underlying drivers: rate realisation, utilisation, leverage (the mix of seniority on a matter), write-offs, and lock-up.
Before building any reporting, we ran a structured exercise with finance and practice leadership to agree definitions for each margin driver. This was the single most important step in the project. Without it, the outputs would have faced the same credibility problem as the spreadsheets they replaced. Each definition was documented, circulated, and signed off. Where practices had historically used different approaches, we recorded the rationale for the chosen standard and provided a mapping from old definitions to new.
We built an ETL pipeline that resolved mapping issues across the four source systems, creating a canonical matter identifier that linked time worked, fees billed, rates applied, and staff allocated into a single record per matter. The agreed data was structured into a dimensional model with matter as the central fact, surrounded by dimensions for client, practice group, office, fee earner seniority band, matter type, and time period. Allocation rules for overheads and support costs were embedded in the pipeline (not applied manually each period), ensuring consistency and full traceability. The reporting layer was built for self-service: partners could see total margin, margin trend, and the contribution of each driver, then drill into specific clients, matters, or seniority bands. Automated alerts flagged matters exhibiting low-margin patterns (declining realisation, rising write-offs, unfavourable leverage) early enough for intervention.
Results and Impact
| Metric | Outcome |
|---|---|
| Reconciliation time | Reduced by approximately 60%, freeing finance for analysis rather than data assembly |
| Margin visibility | Extended from practice-level to individual matter and fee earner seniority level |
| Pricing confidence | Partners able to reference comparable matter economics when setting and negotiating fees |
| Write-off detection | Patterns identified during active matters, not retrospectively at close |
| Staffing adjustments | Leverage analysis prompted rebalancing of seniority mix on several high-volume matter types |
| Adoption | Rolled out firm-wide after initial pilot with two practice groups |
The most visible impact was on client conversations. Partners walking into fee negotiations or relationship reviews could show (internally) exactly what a client relationship cost to serve, where margin was strong, and where it was under pressure. Less visibly, the leverage analysis revealed that several practice groups were habitually over-staffing at senior level on matter types that could be delivered with more associate hours, improving margin without reducing quality.
Key Takeaways
-
A margin number on its own is almost useless; driver-based views show what to change. Telling a partner their matter margin is 35% gives them no actionable information. Showing them that margin is 35% because realisation is strong but leverage is unfavourable tells them exactly what to adjust.
-
Agreed definitions determine whether profitability reporting is trusted or ignored. The single largest risk is that stakeholders dismiss the numbers because they do not match their own mental model. Rigorous definition work upfront, transparent methodology, and reconciliation to the figures finance already trusts made the difference.
-
Alerts need calibration or they become noise. Initial thresholds were set analytically, but the firm's tolerance for alerts was much lower than the statistical model suggested. Practical calibration, done iteratively with the people receiving the alerts, was essential. An alert system that gets ignored is worse than no alert system.