Complaints root cause analysis: what the FCA expects from firms

Complaints root cause analysis (RCA) is how a firm finds and fixes the cause of complaints rather than only their symptoms, and it is required by DISP 1.3.3R. The rule says a respondent must put in place appropriate management controls and take reasonable steps to ensure that, in handling complaints, it identifies and remedies any recurring or systemic problems, for example by:

1. analysing the causes of individual complaints so as to identify root causes common to types of complaint;
2. considering whether such root causes may also affect other processes or products, including those not directly complained of; and
3. correcting, where reasonable to do so, such root causes.

The guidance that follows the rule (DISP 1.3.3B) lists the processes a firm may need, depending on the nature, scale and complexity of its business and the number of complaints it receives. They include management information on the causes of complaints, a process to identify and prioritise root causes, a process to consider whether they affect other processes or products, and regular reporting to senior personnel. The FCA's good and poor practice report reviewed 40 firms and found that some treated completing an RCA as the goal, rather than acting on what it showed.

What the FCA's report covers

The FCA's good and poor practice report on complaints and root cause analysis reviewed 40 firms. It summarises the requirement as: firms "should have appropriate management controls in place to analyse individual complaints and identify the root causes of systemic or recurring problems", and should take appropriate action where they find failings under the Consumer Duty. The sample baseline was at least 90 complaints to the Financial Ombudsman Service over the previous three years, and the information request went out in January 2024. The report records these examples of good practice:

• Most firms sampled had a framework for carrying out RCA.
• Action plans had owners and deadlines. Staff training or a process change was more likely to be taken forward when it was in the plan.
• Some firms reviewed whether their solutions worked afterwards.
• One firm measured the impact of an electronic verification change using customer feedback and take-up rates.
• One firm shared insights beyond the complaints team, with front-line agents and the risk team.
• One firm used the "Five Whys" technique to test its assumptions about root causes.

What the FCA found was lacking

The same report identifies the areas where firms fell short. The FCA's wording is worth reading closely:

1. Action after harm. "It was not always evident from the material that was reviewed that firms were taking action after identifying harm."
2. RCA as the goal. "It appeared that some firms saw the completion of RCA as the goal, rather than whether they needed to take further action."
3. Analysis too high-level. Some RCA processes were "high-level and didn't effectively identify trends and systemic issues".
4. Impact not checked. Many firms did not show whether a change was the right change, or whether it had the desired effect.
5. Training instead of other action. Some firms focused on staff feedback and training rather than other action, such as process changes or proactive redress.

The report also notes that complaints data was not granular enough to show outcomes for different customer groups, including customers with characteristics of vulnerability. Our Consumer Duty support covers how that data feeds outcome monitoring.

Do you need an RCA on every upheld complaint?

DISP 1.3.3R does not say that every upheld complaint needs its own formal RCA. It requires firms to analyse the causes of individual complaints so that common root causes are found. An upheld complaint is a useful prompt to look at its cause. A single error with a clear and tested fix may not need a formal analysis, while a cluster of complaints with the same cause probably does.

It helps to set a trigger in your procedures, for example when complaints share a cause or when a control has failed. Record why you decided for or against an analysis each time, so the decision is consistent.

The root cause versus the symptom

The FCA's understanding complaints root cause analysis page makes the same point with a case study. It says it is important for RCA to "distinguish between a symptom and the root cause of a complaint". In the case, a firm saw more calls about collections, a rise in accounts in arrears after a payment system update, and more social media complaints. The initial symptom was missed payments and fees. The root cause was a technical fault in the update, weak governance, and testing that was not enough. No senior manager had clear responsibility for the update.

The firm reassigned responsibility for critical systems updates, revised its testing and risk management framework, and checked a sample of standing order accounts to confirm the fix worked. It also reviewed every affected account, wrote to customers, refunded incorrect fees and amended credit reference agency files.

A useful test is to ask whether the same cause would produce a complaint next month. If it would, the fix needs to reach the process, not only the person who handled the complaint. Record the date of each impact check, so you can show that the change was tested and not assumed.

Complaints data is only useful for RCA if you can slice it. Where you can, record each complaint's likely cause, the product involved and whether the customer has any characteristics of vulnerability. That lets trends show up in the data rather than in individual files, and it addresses the granularity gap the report describes.

Practical steps

1. Set a documented trigger for a formal RCA, such as shared causes or a failed control.
2. Give each corrective action an owner and a deadline.
3. Check whether each change worked, using complaint volumes, customer feedback or a sample of cases.
4. Where a systemic problem is found, consider whether customers who never complained may also have been affected.
5. Record the decisions senior staff make on the findings.

How MEMA can help

We can set up an RCA framework with triggers, owners and impact checks, and test it against your complaints data. Our complaints root cause analysis explainer sets out the method, and our FCA complaints handling support covers the procedures behind it. For how complaints data feeds outcome monitoring, see our guide to Consumer Duty outcomes testing.