Causal Analysis – Not Just “What Changed?”, but Above All “Why?”

In the previous part of the financial analysis, we were able to determine what had changed in the company and how its financial structure had evolved. Causal analysis takes us one step further. Its purpose is to identify the factors that led to a particular result and determine how much each of them contributed to the final outcome.

This is particularly important from the perspective of a business owner or manager. Simply knowing that costs increased by PLN 200,000 or that the margin decreased by 3 percentage points does not tell us very much. The real question is:

WHY?

From the Result to Its Causes

Imagine a transport company.

Last year, its revenue amounted to PLN 10 million, while this year it reached PLN 12 million.

Horizontal analysis tells us:

Revenue increased by PLN 2 million, or 20%.

This is useful information, but it is far from sufficient for someone managing the company.

We want to know where the additional PLN 2 million came from. Did the company:

* handle more orders,

* increase the average price of its services,

* acquire more profitable customers,

* change the structure of its services,

* or did several factors work simultaneously?

This is where causal analysis begins.

⸻

Example 1. Why Did Revenue Increase?

Let us assume a simple model:

Revenue = number of orders × average order value

In the previous year:

1,000 orders × PLN 1,000 = PLN 1,000,000

In the current year:

1,200 orders × PLN 1,100 = PLN 1,320,000

Revenue therefore increased by:

PLN 1,320,000 – PLN 1,000,000 = PLN 320,000

But where exactly did this additional PLN 320,000 come from?

We can apply the successive substitution method.

First, we change only the number of orders:

1,200 × PLN 1,000 = PLN 1,200,000

Impact of the increased number of orders:

PLN 1,200,000 – PLN 1,000,000 = +PLN 200,000

Next, we introduce the new average order value:

1,200 × PLN 1,100 = PLN 1,320,000

Impact of the increase in the average order value:

PLN 1,320,000 – PLN 1,200,000 = +PLN 120,000

Therefore:

+PLN 200,000 + PLN 120,000 = +PLN 320,000

Only now can we make a much more valuable statement:

Revenue increased by PLN 320,000. PLN 200,000 of this increase resulted from a higher number of orders, while the remaining PLN 120,000 resulted from an increase in the average value of each order.

This is management information.

⸻

Example 2. Labour Costs Increased. Is That Bad?

Suppose a manufacturing company notices a significant increase in labour costs.

The first reaction might be simple:

“Employee costs are too high.”

However, causal analysis may completely change our assessment of the situation.

Labour costs can be simplified into the following relationship:

production volume × time required per unit × labour cost per hour

Assume that the company:

previously produced 1,000 units and now produces 1,050 units;

the time required per unit increased from 1.50 to approximately 1.57 hours;

while the average labour cost per hour decreased from PLN 8.20 to PLN 8.00.

We therefore have three factors operating simultaneously.

Higher production increases costs.

Higher labour intensity also increases costs.

A lower hourly labour rate reduces costs.

Simply stating:

“Labour costs increased”

could therefore lead to an incorrect conclusion.

Causal analysis shows which factor was responsible for the increase and which factor worked in the opposite direction.

⸻

Example 3. Sales Are Growing, but Profit Is Falling

This is one of the most interesting situations from a management perspective.

Assume:

Year 1:

sales – PLN 10 million

profit – PLN 1 million

Year 2:

sales – PLN 12 million

profit – PLN 700,000

Sales increased by 20%, while profit decreased by 30%.

At first glance, this seems contradictory: the company is selling more but earning less.

Causal analysis encourages us to look for the explanation.

It may turn out that sales growth was achieved by reducing prices. Fuel, energy, subcontractor or labour costs may have increased. The company may also have acquired a large customer generating substantial revenue but a very low margin.

In that case, the company’s problem is not a lack of sales.

The problem may be the quality of those sales.

This is a crucial distinction.

⸻

Example 4. Revenue Is Growing, but Cash Is Running Short

The company increased its revenue:

from PLN 10 million to PLN 13 million.

That represents growth of as much as 30%.

At the same time, trade receivables increased:

from PLN 2 million to PLN 4 million.

A manager should therefore ask:

Are we really selling more, or are we primarily extending more credit to our customers?

Possible causes include longer payment terms, a higher proportion of customers paying late, or rapid sales growth without adequate working capital financing.

A company can therefore demonstrate impressive revenue growth while simultaneously experiencing increasing cash flow problems.

⸻

Example 5. Customer Profitability

Causal analysis can also be used beyond traditional financial statements.

Suppose a customer generates PLN 2 million in annual revenue for the company.

They are one of the company’s largest customers.

Does that mean they are also one of the best?

Not necessarily.

We can analyse the costs associated with serving that customer: transportation, warehousing, additional administrative work, complaints, discounts, extended payment terms or the financing of receivables.

It may turn out that a customer generating PLN 2 million in sales contributes less profit to the company than a customer with turnover of PLN 800,000.

Causal analysis therefore changes the perspective:

instead of asking only “How much does this customer buy?”, we ask “What causes us to make or lose money on this customer?”

⸻

Methods of Causal Analysis

Financial analysis uses various methods to isolate the impact of individual factors. These include the successive substitution method, logarithmic method, functional method, residual method and partial differences method.

One of the most intuitive approaches is the successive substitution method.

If a result depends, for example, on three factors:

Result = A × B × C

we compare the values from two periods and then successively replace the old values with the new ones. This allows us to determine how the change in A, then B, and finally C affected the overall result.

The logarithmic method allows the impact of several variables in multiplicative relationships to be allocated without making the result dependent on the order in which substitutions are made. The functional method can likewise be used to attribute changes in the result to individual factors.

In management practice, however, the mathematical technique itself is not the most important issue.

What matters most is the interpretation of the result.

Causal Analysis as a Management Tool

A good manager should not stop at the statement:

“Costs increased.”

The next question should be:

Why did they increase?

It is not enough to say:

“Sales decreased.”

We need to know:

Did we sell less, or did we sell at lower prices? Which customer group was responsible for the decline? Which product? Which market?

Nor is it enough to say: