What “spreading” actually is

Spreading is the process of taking a company’s raw financials (statements, tax returns, trial balances) and reformatting, normalizing, and aligning them into a consistent template across periods and entities. The output is a clean, comparable set of numbers plus key ratios you can use to judge performance, risk, and covenant compliance.

Think of it as turning messy, apples-and-oranges inputs into standardized, decision-ready financials.

Why people do it

1. Make different businesses comparable. Same chart of accounts, same ratios, same definitions.

2. See trends over time. Three to five years of history plus trailing twelve months or interim periods.

3. Calculate credit metrics like Debt Service Coverage Ratio and set covenants that actually mean something.

4. Feed models and credit memos with reliable, normalized data that committees can trust.

What gets spread

• Income Statement: Normalized revenue, COGS, operating expenses, EBITDA and EBIT, interest, taxes, net income.

• Balance Sheet: Standardized current and non-current splits. Debt broken out by type. Equity components.

• Cash Flow: Either indirect GAAP or a lending style like Uniform Credit Analysis to get Cash Flow Available for Debt Service.

• Debt Schedule: Principal and interest by facility, revolver usage, maturities, covenant tests.

• Affiliates and Guarantors: Consolidated or global cash flow view where needed for multi-entity structures.

Common normalizations and reclasses

• Remove non-recurring items. One-off gains or losses, litigation settlements, disaster relief, anything that distorts the underlying operating performance.

• Owner and related-party adjustments. Above-market compensation, personal expenses run through the business, related-party rents that need to be normalized to market rates.

• Convert cash-basis tax returns to accrual, especially for SMEs where tax filings are the only available financials.

• Lease treatment. Ensure operating versus finance leases are handled consistently, especially post-ASC 842.

• Reclassify interest income and expense, other income, extraordinary lines so they appear in the right places.

• Align seasonality and partial periods. Trailing twelve month calculations or annualizations, clearly labeled so reviewers know what they are looking at.

• Consolidate affiliates or attribute intercompany eliminations when dealing with related entities.

Ratios and metrics typically produced

• Liquidity: Current ratio, quick ratio, accounts receivable days, accounts payable days, inventory turns, cash conversion cycle.

• Leverage: Total Debt to EBITDA, Net Debt to EBITDA, Debt to Equity.

• Coverage: Debt Service Coverage Ratio (Cash Flow Available for Debt Service divided by debt service), Fixed Charge Coverage Ratio, Interest Coverage.

• Profitability: Gross margin, EBITDA margin, return on assets, return on equity.

• Common-size: Income statement as percent of sales, balance sheet as percent of assets, plus trend analysis to spot changes.

Workflow at a glance

1. Collect: Audited statements, reviews, management-prepared financials, tax returns, debt agreements.

2. Map: Tie every source account to your template’s chart of accounts.

3. Normalize: Do the adjustments listed above. Document each one.

4. Reconcile: Check subtotals, tie to source statements, explain variances.

5. Analyze: Compute ratios, trends, covenant tests, sensitivities.

6. Output: Tables and charts for the credit memo or investment committee.

Deliverables you usually see

Three to five years of historical financials plus interim or trailing twelve months in a standard template.

Adjustment log showing what changed and why.

Ratio deck, covenant tests, and a global cash flow if multiple entities or personal guarantees are involved.

Brief commentary on what is improving, what is deteriorating, what is one-off versus structural.

Pitfalls to avoid

Mixing cash-basis tax numbers with accrual profit and loss without adjusting. This creates phantom trends.

Using inconsistent periods like thirteen-month versus twelve-month without trailing twelve month normalization.

Double-counting debt service by counting both cash flow statement outflows and the debt schedule.

Ignoring revolvers. Availability versus usage matters. Off-balance sheet obligations matter.

Treating owner perks and related-party transactions as operating expenses when they are not representative of third-party costs.

Not documenting adjustments. Hard to defend in committee or to auditors when you cannot explain your normalization choices.

Tools people typically use

Excel or Google Sheets templates. Most common and most flexible.

Credit packages like Abrigo, Sageworks, Moody’s RiskAnalyst, nCino.

Light ETL scripts for mapping charts of accounts when doing many spreads.

Quick example

Reported EBITDA: $3.0 million

Add back one-time storm repair: +$0.4 million

Normalize owner salary to market rate (reduce expense): +$0.2 million

Adjusted EBITDA: $3.6 million

Term debt service (principal plus interest): $2.4 million

DSCR: Adjusted Cash Flow Available for Debt Service divided by $2.4 million (after working capital and capex adjustments)

This is the kind of work that takes an experienced analyst 30 to 45 minutes per company per period. Multiply that across dozens or hundreds of portfolio companies every month and you see why SME Capital was spending 35 analyst hours per month just on spreading.

Why spreading is so hard

Here is the fundamental tension: spreading needs to be both consistent and flexible.

Consistent because you are comparing companies across time, across industries, across deal structures. If one analyst treats owner compensation one way and another treats it differently, your portfolio metrics are garbage.

Flexible because every business is different. Cash basis versus accrual. Consolidated versus standalone. Operating leases treated as debt or not. Related party transactions that need normalizing. One-off events that distort trends.

The situation gets worse when you factor in document variety. Audited statements, reviews, compilations, management-prepared financials, tax returns, trial balances. Each has different levels of detail, different formats, different reliability. A good spreading process handles all of them and produces comparable output.

Then add the time pressure. Deals move fast. Underwriters need numbers today, not next week. Portfolio managers need monthly updates on hundreds of loans. The manual approach does not scale.

We will dive deeper into how AI solves these specific challenges in an upcoming post on using AI in your spreading flow. For now, let’s establish what spreading actually involves.

Why we built an agentic solution from the ground up

After seeing what spreading actually required, we knew rule-based automation would not cut it. The variability was too high. The judgment calls too nuanced. The document formats too inconsistent.

We needed a system that could reason about financial statements the way an analyst does. Not just extract numbers, but understand context. Recognize when something looks off. Know when to apply an adjustment and when to leave things alone. Adapt to different industries, different accounting treatments, different deal structures.

That is why we built an agentic, specification-driven approach.

The SME Capital breakthrough

When SME Capital came to us, they had the exact problem we had studied for years. A growing portfolio of hundreds of loans between £0.5 million and £5 million. Every month, covenant monitoring and financial spreading ate up 35 analyst hours. Skilled people doing repetitive work instead of building relationships or closing deals.

We deployed our agentic spreading system. It learned their normalization rules, their covenant definitions, their portfolio quirks. It handled different statement formats automatically. It extracted the key metrics, calculated covenant ratios, flagged potential breaches.

The result: 35 hours down to 5 hours. A 95% reduction.

Monthly covenant monitoring that used to take a week now takes less than a day. With better accuracy and deeper analysis. Thirty hours freed up every month, like adding a full analyst without hiring anyone. Those hours went straight to revenue generation with a 3x increase in loan origination capacity.

The system hits 98% accuracy extracting financial metrics. Errors that led to missed covenant breaches are basically gone. Covenant breaches get flagged in hours, not weeks. Management sees portfolio health in real time, not quarterly.

Why flexibility and specification-driven behavior matter

Every lender has different requirements. Different covenant structures. Different normalization policies. Different risk appetites.

Our agentic approach lets lenders specify exactly what they need. Define your EBITDA adjustments. Set your covenant thresholds. Specify how you want related-party transactions handled. The system adapts to your specifications, not the other way around.

This is critical because spreading is not a commodity. The way you normalize financials reflects your credit philosophy. A system that forces you into rigid templates forces you to abandon the judgment that makes your underwriting defensible.

SME Capital needed to handle businesses with £250,000-plus earnings across different sectors. Manufacturing, distribution, services, each with different normalization needs. Our specification-driven agents handled that variety without requiring separate workflows or manual intervention.

Long-running agentic behavior as a key capability

Spreading is not a single API call. It is a process that unfolds over minutes or hours as the system works through documents, reconciles inconsistencies, applies adjustments, validates results.

Our agents run as long as they need to. They iterate on extractions when confidence is low. They cross-reference multiple documents to resolve conflicts. They apply learned normalization rules and flag cases that need human review.

This long-running behavior is what makes the system feel like working with a junior analyst who handles the tedious parts while escalating the judgment calls. You get the speed of automation with the flexibility of human reasoning.

For SME Capital, this meant the system could handle their entire portfolio refresh each month without babysitting. Analysts log in at month-end, review the automated spreads, approve the ones that look clean, dig into the flagged cases. What used to consume their entire first week of the month now fits into a morning.

Before and after

Before: Hours per spread multiplied by portfolio size equals bottleneck. Growth limited by analyst capacity. Covenant breaches discovered weeks late. Portfolio analytics always out of date.

After: Automated spreading that scales with portfolio size. Analysts freed up for relationship management and deal origination. Real-time covenant monitoring. Portfolio health visible at a glance.

The shift from manual to automated spreading is like going from sending individual emails to using a CRM. The work still happens, but the leverage changes completely.

Just as our work on institutional memory for lending and cross-sell automation shows, the unlock is not just speed. It is the ability to do things that were previously impossible at scale.

When spreading takes 30 minutes per company, you only spread when you have to. When it takes 2 minutes and happens automatically, you spread everything. You spot trends earlier. You catch problems before they become crises. You have the data to build predictive models.

What we are excited about for the future

Financial spreading is just the starting point. Once you have clean, normalized financials flowing automatically, the next layer becomes possible.

Predictive covenant breach modeling. If EBITDA is trending down and debt service is fixed, you can forecast when covenants will break before it happens.

Automated health scores. Portfolio-wide risk rankings that update monthly without analyst intervention.

Real-time stress testing. What happens to your portfolio if interest rates jump 200 basis points or if a sector has a down quarter?

Early warning systems that flag deteriorating borrowers before they miss a payment.

All of this depends on having reliable, normalized financial data at scale. That is what automated spreading unlocks.

We are building the infrastructure for a new generation of portfolio management. Where lenders spend their time on judgment and relationships, not data entry and spreadsheet reconciliation. Where risk monitoring is proactive, not reactive. Where growth is not limited by analyst capacity.

The journey from testing every spreading solution on the market to building our own taught us what actually matters. Not features. Not integrations. Not fancy dashboards.

What matters is whether the system does the work correctly, handles the variety that real lending presents, and scales without breaking.

We built that system. The results at SME Capital and our other lending partners prove it works. And we are just getting started.