Root cause analysis: what it is, when you really need it, and when it is overkill

A professional with two team members, a fifteen-person machine shop and a hundred-employee company instinctively choose the second, and sometimes they are right: a full analysis is estimated at between 20 and 90 person-hours [3], too many for a half-hour problem.

Root cause analysis (RCA) is the family of methods that works back from the symptom to the cause that, once corrected, keeps the problem from coming back [1][2].

In Italy, 80.9% of companies with at least 3 employees are under personal or family control [7]: the person being called to account is often a long-standing team member.

Root cause analysis is not a technique to apply to every problem: it is the decision to dig down to the cause that, if corrected, prevents the problem from returning — and you make it only when the cost of digging is lower than the cost of the problem that keeps repeating.

This article explains what it is, when you need it and when it is overkill, how it differs from the 5 whys, and walks through a complete case with a one-page template.

From "it happened again" to root cause analysis: what it is and how it differs

Reissuing the wrong invoice, or understanding why the invoice goes out wrong every week: which of the two actually closes the problem?

Only the second, and the quality vocabulary gives them two different names: the first is a correction, the second a corrective action [6].

Without a method, a recurring problem gets the same response every time: you fix the effect, look for whoever made the mistake, and close the meeting with "let's be more careful."

Root cause analysis — analysis of root causes or of deep causes, as the Italian document that codified it calls it [2] — is the family of methods that travels the opposite way: from the symptom to the cause that, if corrected, prevents a recurrence [1][2].

This section sets out the definition, distinguishes immediate cause, root cause and contributing factor, and separates RCA from its close neighbors: correction, business diagnosis, audit and risk analysis.

The signal is a single sentence: "it happened again."

RCA identifies what happened, how and why; the root causes it looks for are specific, reasonably identifiable and under management's control [1].

If the cause lies outside your control — the customer, the market — it is not a useful root cause, it is a constraint.

In the Prodability glossary, root cause analysis is the family of techniques that carry out the analysis phase of problem solving: here the focus is on how to choose among those techniques.

In the invoice case, the document's three levels of cause [2]:

• Immediate cause: the price on the invoice differs from the quote.
• Root cause: the change agreed verbally on the shop floor does not reach the accounting office in writing.
• Contributing factor: the price list is out of date; on its own it does not generate the error, but it makes it more likely.

A correction eliminates the detected nonconformity; a corrective action eliminates the cause so that it does not recur [6].

Business diagnosis and audits compare the company with predefined criteria; RCA starts from a specific event and looks for the reason behind a deviation that has already happened.

Risk analysis asks what could go wrong and how serious it would be; RCA was born instead as a reactive analysis tool [2] and asks what went wrong and why.

Reissuing the invoice is a correction; understanding why it goes out wrong and acting there is a corrective action: RCA is the stretch of road between the two.

Deciding whether to dig: when root cause analysis is needed and when it is overkill

A twenty-hour analysis of a problem that costs ten minutes a week, or no analysis at all of a complaint that comes back every month: which of the two mistakes costs more?

Both, which is why the Italian document that codified the method recommends stratifying events by frequency and severity before opening an analysis [2]: a full RCA is estimated at between 20 and 90 person-hours [3].

Digging down to the root cause is a decision, not a reflex: you make it when the problem keeps coming back and its cost exceeds the cost of the analysis.

For companies with a certified quality system — in Italy there were 101,426 active ISO 9001 certificates at the end of 2024 [8] — determining the causes of a nonconformity is a requirement [5]; for everyone else it is a choice of method.

This section gives the three signals that a problem deserves digging, the three cases in which a correction is enough, and a two-variable grid for deciding in five minutes.

The 20-90 hour estimate [3] is an expert estimate and comes from the US healthcare literature, where the method is mandatory.

In a fifteen-person machine shop, twenty person-hours are two and a half days of a manager's time; reissuing an invoice costs twenty minutes.

Three signals tell you the problem deserves digging:

• Recurrence: the same effect at least twice in the same period; symptoms that repeat are the first signal.
• Measurable cumulative cost: hours of rework, credit notes, postponed deliveries, lost customers.
• Non-obvious cause: different people point to different causes, or the "obvious" cause has already been corrected and the problem came back.

In three cases a correction is enough:

• Isolated event with an evident cause: a correction and one line in the nonconformity log; there is nothing to dig into.
• One-hour problem with a known fix: a correction and a date to see whether it comes back; the analysis would cost more than the problem.
• Problem already being measured: you wait for the data; without numbers the analysis starts from opinions.

The document's two-variable matrix [2] becomes a recurrence × impact grid: how many times in the last quarter, how many hours or how much money each time it returns.

scroll the table →

The grid answers the opening question: the twenty-hour analysis sits in the low-low quadrant, where it is not needed; the monthly complaint sits in the high-high quadrant, where not opening one costs you every month.

Family or technique: the difference between root cause analysis and the 5 whys (and Ishikawa)

Five "whys" in a row on a sheet of paper, or a fishbone with six branches: is that already a root cause analysis, or just one piece of it?

It is one piece: the Italian document that codified the method lists the 5 whys and the cause-and-effect diagram among the techniques of the same family [2], and the literature describes RCA as a four-phase process in which identifying the cause is the third phase [1].

The confusion arises because the techniques have their own names and the family has another.

The 5 whys digs deep into a single causal chain, the Ishikawa diagram opens up the families of causes in breadth: they are two techniques of the analysis phase, not two alternatives to RCA.

This section places the three levels — the processes that host the analysis, the family, the techniques — and puts in a table which technique to choose based on the shape of the problem, with links to the dedicated pages.

The difference between root cause analysis and the 5 whys is one of level: RCA is the family of methods, the process that works back from the symptom to the cause; the 5 whys is one of its techniques, like the Ishikawa diagram.

The three levels:

• Complete processes: PDCA, A3 report and 8D, each with a cause analysis phase.
• Family: root cause analysis, the phase in which causes and contributing factors are determined within each process.
• Techniques: 5 whys and the Ishikawa diagram, plus the timeline, barrier analysis and fault tree [2].

The Kepner-Tregoe method belongs to the same family: before looking for the cause, it bounds the problem with an "is / is not" template and derives hypotheses from the differences between the two columns.

As a process, RCA has four phases — data collection, causal factor charting, root cause identification, recommendations and implementation [1] — which the six steps in the next section spell out in detail.

You choose the technique by the shape of the problem:

scroll the table →

The criterion fits in one line: you choose the technique by the shape of the problem, and the process by severity and scope, using the comparison table from PDCA to root cause analysis.

Five "whys" on a sheet of paper remain one piece: RCA starts earlier, with the data, and ends later, with the verification.

A complete root cause analysis example: from symptom to root cause in six steps

Six credit notes a month for invoice errors: is it a problem with the accounting clerk or with the company?

The answer changes with the method: without analysis, the blame falls on the last person who touched the document; with six steps, the cause moves upstream, to where the price is decided verbally on the shop floor — and a root cause, by definition, is under the company's control, not a single person's [1].

The case is hypothetical but built on a common situation: a fifteen-person contract machine shop, family-controlled like 74.5% of small Italian companies [7], where the business owner agrees on changes and prices directly on the shop floor.

The symptom is measurable — six credit notes out of about one hundred twenty invoices a month — and the table shows the path step by step: definition, data, sequence, contributing factors, causal chain, verified root cause.

The same path works for a practice with two team members and for a hundred-employee company: the documents on the table change, the questions do not.

Credit notes are, first of all, customer complaints.

scroll the table →

The leap is between the apparent cause and the root cause: "the clerk makes mistakes" is the immediate cause, "changes do not reach the office in writing" is the root cause, "the price list is out of date" is a contributing factor.

The five notes on quantity and billing details stay out: an RCA does not have to explain every case, it has to explain the part that carries the weight.

The answer to the opening question is "the company": the problem lay in a missing handoff, and the person who "made mistakes" is the first to benefit from the form.

The one-page sheet as a root cause analysis template: the fields and how to fill them in

One page filled in within an hour, or a ten-page dossier written over a week: which of the two gets reread when the problem comes back?

The page, which is why the A3 report requires a single sheet; but a template is only as good as its last field, the verification one, and it is precisely the check on effectiveness that is most often missing in practice [3].

The template is not a form to be filed away: it is the structure that forces you to write the problem with a number, the causes with evidence and the action with a name and a date.

The one-page RCA template follows the same six steps as the machine shop case and adds two fields that make them verifiable: who does what by when, and what gets measured sixty days later.

This section describes the eight fields one by one, with the typical sentence to write in each and the mistake that empties it.

The eight fields:

1. Problem — one sentence with a number and a period; mistake: "invoices often go out wrong."
2. Data collected — source, period, classification of occurrences; mistake: opinions gathered in a meeting.
3. Sequence — who, what, when, from the first step to the symptom; mistake: starting from the last step.
4. Contributing factors — by family, each with its evidence; mistake: a list without proof.
5. Causal chain — the cascading whys, three to five levels, down to a cause statement; mistake: stopping at the first plausible level.
6. Root cause — one sentence with the double test: if corrected, does it prevent recurrence? is it under the company's control? [1][2]; mistake: writing down a culprit.
7. Action on the cause — specific, feasible, with an owner and a date, tested before being extended [2]; mistake: "more attention," "training."
8. Verification — indicator and date for the fourth question, "has the risk of recurrence been reduced?" [3]; mistake: empty field.

The template is not an A3 report: the A3 communicates a cross-functional problem to the people who decide, while the template covers only the analysis phase on one page and can go into the "cause analysis" field of an A3.

For certified companies, the completed template is the evidence on causes and effectiveness required by clause 10.2 [5].

The page gets reread because it goes into the log with field 8 filled in: verification is what sets it apart from meeting minutes.

The mistakes that turn a deep-cause analysis into a ritual (and how to avoid them)

An analysis that ends with "let's be more careful" and one that ends with a new form signed on the shop floor: which of the two found a cause?

Only the second: the literature that has studied the method where it has been mandatory for twenty years flags as a recurring flaw the fallback on weak remedies, such as reminders, instead of acting on latent causes [4].

The mistakes are not in the technique but in how it is used, and they are the same in a hundred-employee company and in a three-person practice.

This section puts six of them in a table, from stopping at the apparent cause to searching for "the" single cause, and for each one shows how to recognize it and the move that neutralizes it.

The last one is the most expensive: closing the analysis without checking, sixty days later, whether the problem has come back.

The observations [3][4] come from the healthcare literature, where the method is mandatory, and serve as a methodological warning.

The form signed on the shop floor is proof that the analysis touched a handoff and not a person; "let's be more careful" is proof that it stopped one level too early.

Limits and conditions of applicability

The estimate of 20-90 person-hours per analysis [3] and the critical literature on RCA [4] come from the healthcare sector, where the method is mandatory and repeated.

The figure is an expert estimate, not a measured figure for Italian companies: in a business the cost depends on which quadrant of the grid the problem falls into.

The flaws reported — a single cause, weak remedies — serve as a methodological warning, not as a measure of how often they occur in companies.

The machine shop case is hypothetical and for teaching purposes: the numbers illustrate the method, they do not estimate results, and in a real case the classification of the notes and the cause may be different.

The method assumes a problem that has already happened and been documented: without data on recurrence the grid cannot be filled in, and the analysis starts from opinions.

FAQ

What does root cause analysis mean?

Analysis of root causes or analysis of deep causes: that is the name used by the Italian document that codified the method [2].

It refers to the family of methods that works back from the symptom to the cause that, if corrected, prevents the problem from returning.

What is the difference between root cause analysis and the 5 whys?

Root cause analysis is the family; the 5 whys is one of its techniques, like the Ishikawa diagram.

The 5 whys digs into a linear causal chain; RCA adds the data before and the verification after, and the technique is described in the glossary.

How long does a root cause analysis take?

It depends on the quadrant of the recurrence × impact grid: a light technique can be wrapped up in half an hour, while a full six-step analysis takes a few days of work spread over two or three weeks, plus the check at sixty days.

The estimate of 20-90 person-hours [3] refers to settings where the method is mandatory.

Do you need software to do a root cause analysis?

No: a one-page template, filled in by hand or in a spreadsheet, contains the eight fields you need.

As a generic example, with no endorsement or recommendation, a spreadsheet in Microsoft Excel, Google Sheets or LibreOffice Calc is enough for the log and for classifying the data.

Practical summary

Before opening an analysis, place the problem in the recurrence × impact grid: full digging goes to the high-high quadrant, a light technique to high-low, containment to low-high, and a correction with one line in the log to low-low.

The first step writes the problem down with a number and a period.

The second collects data from the documents and classifies it, without opinions.

The third reconstructs the sequence from the first step to the symptom.

The fourth lists the contributing factors, each with its own evidence.

The fifth digs into the causal chain on the heaviest factor, down to a cause statement.

The sixth verifies the cause with the double test — it prevents recurrence, it is under the company's control — and closes it with an action that has an owner, a date and an indicator to reread sixty days later.

Conclusion

Root cause analysis is not a technique, it is a choice: to dig down to the cause that, once corrected, prevents recurrence — and to do it only where the repeating problem costs more than the digging.

The machine shop case applies to a practice with two team members as much as to a hundred-employee company: the documents on the table change, the six questions do not.

Which process to choose for a serious or cross-functional problem is in the comparison table of the guide to business problem solving: https://blog.prodability.com/problem-solving-aziendale/

When the causes are many and disorganized, the first step of the digging is the Ishikawa diagram: https://blog.prodability.com/diagramma-di-ishikawa/

When effects and causes get mixed up in the same sentence, the preliminary breakdown is done with the problem tree.

A company that opens a template the second time a problem comes back, and closes it only after verification, builds up in a year an archive of its own recurring causes.

Meetings about "who did it" get shorter, credit notes stop being a fixed line item, and the business owner goes back to the shop floor to make decisions, not to fix things.

Sources and references

[1] Rooney, J. J. and Vanden Heuvel, L. N., "Root Cause Analysis for Beginners", Quality Progress, vol. 37, no. 7, pp. 45-53, ASQ, July 2004. Available at: https://asq.org/quality-progress/articles/root-cause-analysis-for-beginners?id=0228b91456514ba490c89979b577abb4

[2] Ministero della Salute – Direzione Generale della Programmazione sanitaria, "Metodi di analisi per la gestione del rischio clinico — Root Cause Analysis (RCA), Analisi delle Cause Profonde", Volume I, October 1, 2009 (revised edition December 2010). Available at: https://www.salute.gov.it/new/it/pubblicazione/metodi-di-analisi-la-gestione-del-rischio-clinico-root-cause-analysis-rca-analisi/ — PDF: https://www.salute.gov.it/new/sites/default/files/imported/C_17_pubblicazioni_1103_allegato.pdf

[3] Wu, A. W., Lipshutz, A. K. M. and Pronovost, P. J., "Effectiveness and Efficiency of Root Cause Analysis in Medicine", JAMA, vol. 299, no. 6, pp. 685-687, February 13, 2008. DOI 10.1001/jama.299.6.685. Available at: https://pubmed.ncbi.nlm.nih.gov/18270357/

[4] Peerally, M. F., Carr, S., Waring, J. and Dixon-Woods, M., "The problem with root cause analysis", BMJ Quality & Safety, vol. 26, no. 5, pp. 417-422, 2017. DOI 10.1136/bmjqs-2016-005511 (open access, PMC5530340). Available at: https://europepmc.org/article/MED/27340202

[5] UNI, "UNI EN ISO 9001:2015+A1:2024 — Sistemi di gestione per la qualità — Requisiti", Ente Italiano di Normazione, in force since October 16, 2024. Available at: https://store.uni.com/uni-en-iso-9001-2015-a1-2024

[6] UNI, "UNI EN ISO 9000:2026 — Sistemi di gestione per la qualità — Fondamenti e vocabolario", Ente Italiano di Normazione, in force since May 27, 2026. Available at: https://store.uni.com/uni-en-iso-9000-2026

[7] ISTAT, "Censimento permanente delle imprese 2023: primi risultati", press release, November 14, 2023 (reference year 2022; about 280,000 responding companies, representative of 1,021,618 companies with at least 3 employees). Available at: https://www.istat.it/comunicato-stampa/censimento-permanente-delle-imprese-2023-primi-risultati/ — PDF: https://www.istat.it/it/files/2023/11/REPORTCensimprese.pdf

[8] Accredia, "ISO Survey 2024: le imprese accelerano su sostenibilità e digitalizzazione", AccrediaHub, 2025 (ISO data as of December 31, 2024). Available at: https://www.accredia.it/comunicazione/accrediahub/iso-survey-2024-le-imprese-accelerano-su-sostenibilita-e-digitalizzazione/