Contract Intelligence Agent.
A PE-backed multi-state risk-based provider group went from no contracting infrastructure to a validated 6-agent Commercial Contracting Engine in three weeks, with 10–20% of revenue identified as annualized upside ($2–5M).
Company Profile
PE-backed multi-state risk-based provider group, Mid-Atlantic and Northeast footprint. ~$50M annual revenue. Three legacy operating entities merged under one umbrella. Hundreds of payer contracts spanning Medicaid, Medicare Advantage, and commercial. No centralized contracting function. No shared repository. No analytical layer on top of rates, codes, or renewal calendars.
The Challenge
The combined business had grown through entity consolidation, but contracting had never been stood up as a function. Decisions happened reactively, entity by entity, often in the inbox of whoever fielded the last payer email.
Contract terms, fee schedules, and amendments were scattered across email threads, shared drives, and individual laptops. No single source of truth existed for which contracts were live, which were close to renewal, or which were missing entirely. Negotiation prep, when it happened, was assembled by hand on partial information. Limited time was allocated to the work because the people who would have done it were already running other operations.
The PE sponsor’s investment thesis depended on commercial scale across the three entities. The commercial engine couldn’t scale on a foundation of PDFs and spreadsheets.
The 3-Week Engagement
3PS scoped a three-week engagement to stand up the contract management capability from zero.
Week 1: Diagnostic
Sessions with revenue cycle, finance, and the leaders of each legacy entity. There was no formal contracting process to map. The real question was where contracting decisions actually got made and what information the people making them had.
A sweep of shared drives and email archives produced a corpus of over a thousand files. The agent design was specified in a working session, modeled on how a regional health plan runs its contracting shop and sized for a $50M provider group. Before the full build, the design was run end to end against the corpus, sorting contracts from non-contracts, identifying payers, normalizing carrier families across M&A, flagging gaps, and producing a first-pass outreach list.
Weeks 2–3: Build
The validated design became a working contracting capability. Every contract was normalized into a structured record: entity, state, payer, line of business, effective dates, renewal mechanism, rate basis, code set, and VBC provisions where present. Where originals were missing, the next-best documents served as stand-ins until they could be procured.
Six agent workflows were built and tuned in parallel, each able to read vertically within a single payer relationship and horizontally across payers, entities, and states. Three legacy entities meant three NPIs, three Tax IDs, three ways of describing the same clinical service, and three sets of rates. The build reconciled them into a single view so the agents could compare like with like.
Each workflow was tested against contracts where the answer was known, then run against the full corpus, then reviewed with the contracting and finance leads before going live. Issues the leads identified fed back into the workflows so errors were corrected on future runs.
Results
Functional Areas Touched
What started as a contracting build rippled across the organization. The map below shows a typical risk-bearing provider operating structure. Highlighted nodes are where this engagement created downstream impact.