Predictive Risk: Which Swiss Companies Are Most Likely to Dissolve?

Every year, thousands of Swiss companies enter liquidation. For banks, insurers, and trade-credit providers, spotting these failures early is the difference between a managed write-down and an unexpected loss. The VynCo predictive risk model scores dissolution probability for all 780,000+ companies in the Swiss commercial registry, using 10 interpretable signals derived entirely from public filings.

Quick Start

One call gives you a risk score and a full signal breakdown.

from vynco import VynCo

client = VynCo(api_key="your-key")
risk = client.ai.predictive_risk(uid="CHE-123.456.789")
print(f"Risk score: {risk.score:.2f} ({risk.risk_level})")
for signal in risk.signals:
    print(f"  {signal.name}: {signal.value} (weight: {signal.weight}%)")

The score ranges from 0.0 (minimal risk) to 1.0 (high dissolution probability). The risk_level field maps this to low, medium, high, or critical for easier triage.

The 10 Signals

Every score is decomposed into exactly 10 interpretable signals. No black boxes.

The weights sum to 100% and are fixed across all companies, making scores directly comparable.

Comparative Analysis

You can score an entire portfolio in a single batch call.

uids = ["CHE-100.000.001", "CHE-100.000.002", "CHE-100.000.003"]
results = client.ai.predictive_risk_batch(uids=uids)

for r in sorted(results, key=lambda x: x.score, reverse=True):
    print(f"{r.company_name}: {r.score:.2f} ({r.risk_level})")

This makes it easy to rank counterparties by risk and focus your review on the companies that need it most.

Monitoring Over Time

Risk scores update as new SOGC publications appear. Set up a watchlist to track score changes.

watchlist = client.watchlist.get()
for company in watchlist.companies:
    risk = client.ai.predictive_risk(uid=company.uid)
    if risk.risk_level in ("high", "critical"):
        print(f"ALERT: {company.name} risk is {risk.risk_level} ({risk.score:.2f})")

What Drives Real-World Risk?

Looking at the signal weights, a few patterns stand out:

• Liquidation status alone accounts for 19%. This might seem obvious, but many companies linger in liquidation for years. The model captures this.
• Capital decreases (15%) and trajectory (10%) together represent a quarter of the score. A company reducing capital is statistically far more likely to dissolve than one holding steady.
• Auditor signals (11% + 8% + 5% = 24%) collectively form the second-largest risk factor. Changing auditors frequently, losing an auditor, or having a lower-tier auditor all correlate with higher dissolution rates.
• Officer turnover (11%) captures governance instability. Rapid board changes often precede dissolution.

Use Cases

• Credit risk: Score counterparties before extending trade credit or loans
• Portfolio monitoring: Flag deteriorating companies in a managed portfolio
• Compliance screening: Identify high-risk entities in KYC/AML workflows
• Insurance underwriting: Assess policyholder viability for trade credit insurance
• Supply chain: Monitor supplier health to avoid disruption

---

Ready to score your portfolio? Start your free trial and run predictive risk on up to 100 companies with your initial credits.