Trade Shield Models- Payment Risk and Default Risk -How our predictive models work

Created by Mmalethabo Mashigo, Modified on Mon, 8 Jun at 10:04 PM by Amy Sara Price

Trade Shield  |  Knowledge Centre CREDIT INTELLIGENCE

Trade Shield Models: Payment Risk & Default Risk

How our predictive models work — and how to use them to make smarter credit decisions

8 min read    6 sections

Trade Shield uses two predictive models to assess every customer: Payment Risk and Default Risk. These models don't just collect data — they weigh multiple signals against each other to build an accurate picture of how a customer is likely to behave. Together, they power the credit limit recommendations you see in the platform.

1

What is Payment Risk?

Payment Risk is the likelihood that a customer will pay you late or miss a scheduled payment. It focuses on how the customer currently trades with you — their behaviour, consistency, and ability to pay within your agreed terms.

This score is also used to calculate the Potential Credit Capacity (before risk adjustment) — the maximum amount a customer could reasonably spend with you based on their payment behaviour and available capacity across other suppliers.

What the model considers

✓  Age of the company

✓  Number of directors, their age & tenure

✓  Location and industry of the business

✓  Economic conditions

✓  Number of judgements against the business

✓  How the customer pays you and other suppliers

✓  Historic ageing data and payment patterns

✓  Statutory data and negative credit events

✓  Recency of default events

✓  Size of the credit history

Payment Risk Bands

BandWhat it means
LowConsistently pays on time
ReducedSmall chance of late payments
MediumSome late payment history
ElevatedFrequent late payments
HighHigh chance of late payments
Very HighPersistent late payments and serious concerns

ⓘ Note on New Customers: If a new customer has no prior Trade Shield payment history, the platform will show Payment Risk as Unknown and the recommendation will be based on Default Risk only. Once trading data builds up, Payment Risk will populate automatically.

2

What is Default Risk?

Default Risk is the likelihood that a customer will fall into protracted or full default on a future payment. It takes a 6–12 month forward-looking view, assessing the customer's obligations across all suppliers — not just you.

This replaces the need for traditional trade references by providing a data-driven view of how the customer treats their entire credit footprint — not just their relationship with your business.

Key difference from Payment Risk: Default Risk considers the same model inputs as Payment Risk, with one important distinction — it focuses specifically on how the customer treats their obligations with all suppliers, whereas Payment Risk focuses on how they trade with you.

Default Risk Bands

BandWhat it means
LowVery unlikely to default
ReducedLow chance of default
MediumSome indicators of default risk
ElevatedNotable warning signs
HighStrong risk indicators
Very HighHigh probability of default

ⓘ Blue-chip customers: Larger companies may show a higher Payment Risk not because of financial difficulty, but because of internal AP processing delays. Always consider the customer's invoice approval and treasury process before acting on the score alone.

3

Reading the Risk Scores in Trade Shield

Both scores appear side-by-side in the Risk Scores section of the Customer 360 view. Each panel shows the current band, when the score was last calculated, and a colour-coded scale so you can see where your customer sits at a glance.

CUSTOMER 360 → RISK SCORES PANEL

app.tradeshield.ai / customer-360

Risk Scores

A look at this company's likelihood of paying within terms as well as their likelihood of protracted default. We consider their payment behaviour with you and other organisations and is based on the most recent recommendation data.

PAYMENT RISK
17 days ago

Medium

Inoxico

LOWREDMED ↓ELEVHIGHV.HI
DEFAULT RISK
17 days ago

Low

Inoxico

LOW ↓REDMEDELEVHIGHV.HI

Disclaimer

Trade Shield does not accept any liability as to the condition, quality, or performance of any recommendations or analytics provided. This recommendation is calculated for the legal entity and not just this single account within your book.

ⓘ Tip: Scores are recalculated daily, but a limit recommendation only changes when triggered by a data event — such as a customer exceeding their limit, a change in payment behaviour, or a business status update.

4

The Credit Strategy Matrix

Once both scores are produced, the Credit Strategy Matrix tells you how to respond — whether to Grow, Maintain, or Reduce your exposure to that customer.

Payment ↓ / Default →LowReducedMediumElevatedHighVery High
LowGrowGrowGrowGrowMaintainReduce
ReducedGrowGrowGrowGrowMaintainReduce
MediumGrowGrowGrowMaintainMaintainReduce
ElevatedGrowGrowMaintainMaintainReduceReduce
HighMaintainMaintainMaintainReduceReduceReduce
Very HighMaintainMaintainReduceReduceReduceReduce
Grow — actively increase exposure
Maintain — hold current position
Reduce — decrease exposure

ⓘ How to read this matrix: Find your customer's Payment Risk on the left (rows), then cross-reference with their Default Risk across the top (columns). The intersection tells you the recommended credit strategy.

5

How Risk Scores Become a Credit Limit

Once the models produce a Payment Risk and Default Risk band, those scores are applied to the Risk Adjustment Framework (RAM). The order always matters: first the model tells you what the risk is — then the RAM tells you what to do about it.

The 5-step calculation

1

Check Business Status

If the business is deregistered or invalid, a zero recommendation is issued immediately.

2

Determine Maximum Limit (Affordability)

Calculated from turnover data (financials, BEE certificates, or bank statements) multiplied by a turnover percentage — representing how much a customer can reasonably spend with you.

3

Apply Risk Scores via the RAM

The RAM takes both scores and translates them into a percentage of the customer's requested or maximum limit that you should actually offer. Low-risk customers receive a higher percentage; high-risk customers receive a lower — or even negative — adjustment.

4

Apply Knockout Rules

The system checks bank statement standing and business status. All rules must pass for the recommendation to proceed.

5

Generate the Recommended Limit

The final rand amount is produced. For example: Current Limit R80,000 + 50% RAM adjustment = Recommended Limit R120,000.

RECOMMENDATION PANEL — CUSTOMER 360

app.tradeshield.ai / customer-360 / Company 147
C

Company 147

Account: 100147  |  Tag: Critical Customer

Review Limit
▲ RECOMMENDATION 17 days ago

▲ R120,000

GROW LIMIT

Current Limit

R80,000

Current Exposure

R107,695

How we calculated this limit

Default Risk

Reduced

Payment Risk

Unknown

% of Max Limit

70%

Maximum Limit (based on turnover x 3%)

Turnover Estimate

R5,800,000

Max Limit

R170,000

Min Limit

R10,000

Knockout Check

✓ Bank Statements: Good Standing ✓ Business Status: In Business

Configuring the RAM for your credit strategy

The RAM can be configured differently depending on your credit appetite. Three models are available:

Conservative

Minimal exposure, tight limits

Intermediate

Balanced approach

Aggressive

Growth-focused, higher tolerance

Two separate frameworks can run simultaneously — for example, a more aggressive framework for new buyers while keeping a conservative one for existing accounts. Contact your account manager to adjust.

ⓘ Tip: The audit trail in the Recommendation panel shows exactly how a limit was derived — which data sources were used, the risk bands applied, and the resulting percentage. Use this to explain a recommendation to a customer or internally.

6

Reviewing & Acting on Recommendations

You are never locked in to a system recommendation. Trade Shield gives you the tools to review, override, or act on a recommendation using the built-in workflows.

✓ Accept the recommendation

Use the Review Limit workflow in the Customer 360 to apply the recommended limit directly. This triggers the standard approval workflow.

↺ Override with new data

Request updated financials via the platform, or send a credit application to the customer directly to gather fresh information before deciding.

⚠ Important: Recommendations are calculated for the legal entity — not just a single account in your book. A customer with multiple accounts will have the same risk score applied across all of them.

ⓘ Want to learn more? Watch the full training video Behind the Limit: How We Calculate Credit Risk for a walkthrough of the calculation process with live examples.

Need more help?

The support team is ready to assist

Was this article helpful?

That’s Great!

Thank you for your feedback

Sorry! We couldn't be helpful

Thank you for your feedback

Let us know how can we improve this article!

Select at least one of the reasons
CAPTCHA verification is required.

Feedback sent

We appreciate your effort and will try to fix the article