# Case Study: TSS

**Total Specific Solutions leverages raia to accelerate M\&A due-diligence with an Agentic Workforce**

***

**Project Goals**

| **Objective**                                                           | **AI-Driven Approach with raia**                                                                                               |
| ----------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------ |
| Screen more acquisition targets without adding head-count               | Auto-extract 80 + KPIs from Confidential Information Memoranda and run a proprietary **Fit-Score™** model inside each AI agent |
| Cut first-pass memo drafting time from days to hours                    | Have the **Deal Screener Agent** assemble an executive summary, red-flag list and benchmark tables immediately after upload    |
| Provide on-demand answers across thousands of historical deals          | Use a vectorized knowledge base of 10 000 + memos, diligence packs and post-acquisition reviews, searchable via chat           |
| Create a learning loop between post-deal performance and new screenings | Deploy a **Post-Deal Insight Agent** that ingests monthly portfolio data, updates benchmarks and retrains scoring weights      |

***

#### 1 | Company & Challenge

Total Specific Solutions (TSS) acquires and grows vertical-market software businesses throughout Europe. Each potential deal arrives with hundreds of pages—financials, tech audits, HR reports—that analysts must parse quickly. As the pipeline widened, three pain points emerged:

1. **Document Overload** – Manual review could take several days per target.
2. **Metric Normalisation** – KPIs like NRR, gross retention and Rule-of-40 vary wildly across accounting structures.
3. **Knowledge Drift** – Lessons from prior deals were buried in nested folders and unavailable at the moment of analysis.

***

#### 2 | Why TSS Chose raia

* **High-capacity ingestion**—Launch Pad can bulk-vectorize thousands of PDFs and spreadsheets in minutes.
* **Custom code hooks**—TSS’s proprietary Fit-Score™ Python script plugs directly into each agent’s reasoning chain.
* **Granular security**—Project-level silos ensure live deal data is visible only to the assigned pod.
* **CoPilot feedback**—Analysts can up-vote, down-vote and edit AI outputs in real time, tightening accuracy before full rollout.

***

#### 3 | Solution Blueprint

| **Agent**               | **Role**                                                                                           | **Key Integrations**                   |
| ----------------------- | -------------------------------------------------------------------------------------------------- | -------------------------------------- |
| Deal Screener Agent     | Parse new CIMs, extract KPIs, calculate Fit-Score™, draft one-pager.                               | Vector KB, scoring script, email/Teams |
| Deep-Dive Analyst Agent | Answer ad-hoc questions (“Compare churn to peer set”), cite sources.                               | SharePoint, Snowflake data marts       |
| Post-Deal Insight Agent | Pull monthly portfolio metrics, flag variance vs. acquisition case, feed learnings back to the KB. | Power BI API, email digests            |

***

#### 4 | Project Roadmap

1. **Weeks 1-3** – Bulk-upload historical memos & post-mortems; tune scoring model on ten-year IRR data.
2. **Weeks 4-6** – Pilot in CoPilot with two sector teams; capture thumbs-up/down feedback.
3. **Week 8** – Firm-wide release of Deal Screener Agent; Deep-Dive Agent follows.
4. **Month 4** – Go-live for Post-Deal Insight Agent; establish continuous learning loop.

***

**Results coming soon** — the TSS team is currently in the pilot phase and will report quantitative outcomes after the first full quarter of live use.