Case Study · Healthcare · Origin Story
We beat Google and Microsoft Consulting. Then we built a company.
A 100+ hospital network ran a blind bid competition to surface decades of trapped institutional knowledge. A lean team of AI specialists beat two of the largest technology firms in the world. This engagement founded CommonLogic.
The Challenge
"If we only knew
what we knew."
Better information leads to better care and lower costs. That principle was the foundation of this engagement — and the source of the client's frustration. They had the information. They just couldn't reach it.
Across 100+ hospitals and decades of operations, this international healthcare network had accumulated an extraordinary institutional asset: terabytes of high-value knowledge. White papers from renowned surgeons. Meeting notes capturing critical decisions. Research papers. Podcasts. Internal analyses spanning every clinical and operational domain.
None of it was surfaced systematically. None of it was scored for quality or relevance. None of it moved through the organization in a way that could change what a clinician knew before walking into a room, or what an executive understood before making a capital decision.
The Department of Excellence knew AI could provide a way forward. What they needed was a team that could show them how — and prove it, at scale, against the best alternatives available.
The Competition
A blind bid. Six months.
Eight known criteria. Two hidden.
The CIO structured a competition that removed subjectivity from the evaluation. Three firms — us, Google, and Microsoft Consulting Services — each had six months to deliver a working AI workflow. Eight success criteria were published. Two were held back and applied blind at evaluation time.
Global tech giant
Boutique AI specialists
CommonLogic
Global tech giant
Microsoft Consulting Services
Why we approached it differently
Having worked at firms like Google and Microsoft, our team understood how large technology consultancies approach competitive bids: minimum team sizes, extensive planning phases, broad proposals designed to impress during pitches. We chose the opposite. Lean team. Early iterations. A working solution in month 1, not month 4.
The Solution
24 agents.
One orchestrated system.
We built a multi-agent orchestration of 24 full-time agents, each with a specific role in a content intelligence pipeline. Together they ingested, scored, prioritized, summarized, distributed, and classified the client's entire content estate — starting with SharePoint, designed to scale to any media format that could be transcribed to text.
8 known success criteria
Content ingestion agents
Ingested content from SharePoint at launch, designed from the start to expand to any media format that could be transcribed to text — meeting recordings, surgeon podcasts, research papers, white papers.
Scoring & prioritization agents
Evaluated each piece of content across dozens of quality and relevance signals — recency, source authority, clinical specificity, cross-reference density, engagement history — and produced a ranked content queue.
Summarization & distribution agents
Produced structured summaries of high-priority content and routed them to the right internal feeds — department-specific digests, executive briefings, surgical specialty channels — automatically.
Taxonomy & search agents
Built a linked content taxonomy in parallel — classifying and tagging the corpus to power enterprise search, internal knowledge bases, and downstream services beyond the primary distribution pipeline.
How we ran the engagement — POINT
Prioritize the Problem
The problem was clearly defined by the client: "If we only knew what we knew." Terabytes of high-quality institutional knowledge — white papers from renowned surgeons, decades of meeting notes, research, podcasts — existed but was effectively inaccessible. Better information leads to better care and lower costs. That was the KPI.
Outline the Win
The evaluation structure did this for us: 8 known success criteria, 2 blind criteria, 6 months, independently judged. We didn't define what winning looked like — the client did. That clarity made the work sharper. We built to the criteria, not to a demo.
Implement Rapidly
Where the competing firms spent months in planning, we had early iterations running in month 1. A lean team, allocated only to what the problem required — not padded to meet an engagement minimum. The 24-agent orchestration was built iteratively, validated against the criteria at each stage.
Normalize the New Workflow
The solution was designed to run continuously — not as a project deliverable, but as operational infrastructure. Content scoring, summarization, and distribution became automated workflows. The linked taxonomy became the backbone of enterprise search.
Track and Expand
The ELT selected our implementation as the long-standing solution. We transferred full knowledge to their internal team and provided ongoing support. The CFO noted we delivered for a fraction of the cost of the competing proposals. We were awarded the CIO Excellence Award. CommonLogic was formally founded.
Learn more about the POINT framework → How We Deliver
Outcomes
Flying colors. CIO Excellence Award. CommonLogic.
The ELT evaluated all three submissions against the criteria. Our multi-agent workflow outperformed both competing offerings on every published criterion. On the two blind criteria the client withheld, we scored 76% — criteria none of us had seen.
The CFO noted publicly that we delivered the solution for a fraction of the cost of the competing proposals — because we allocated only the resources the problem required, not a headcount minimum designed to justify a large firm's day rate.
8 known criteria
2 blind criteria
criteria we'd never seen
Multi-agent workflow selected as long-standing implementation
The ELT chose our solution over both Google and Microsoft Consulting Services for permanent deployment — not a pilot, not a proof of concept.
Delivered for a fraction of the cost
The CFO cited cost efficiency as a standout differentiator. A lean, purpose-sized team versus firms adhering to engagement minimums that had nothing to do with the problem.
Full knowledge transfer to internal team
We didn't hand over a black box. The internal team received complete documentation, architecture walkthroughs, and ongoing support to own and extend the system independently.
CIO Excellence Award
Awarded by the client ELT in recognition of outstanding contribution to the organization. The highest recognition the engagement could produce.
This engagement founded CommonLogic
Winning this bid — against two of the largest technology firms in the world, with a lean team and a measurable-outcome approach — directly led to the formal founding of CommonLogic.ai.
Complex problem.
Need a team that won't over-engineer it?
We allocate only what your problem requires. We start building before the planning deck is finished. And we measure everything — because that's how you prove it worked.
Talk to us