Use caseRetail banking

Complaints traced to their cause

1,284 complaints closed in Q3, all on time, and 27 percent filed under Other. One failure hides 208 times in there. Proving it took nights.

Every closed complaint gets a cause from your own listTwo handlers check a sample before anything is reportedThe fix goes to the process owner, with a date

Actions on the quarter
The cause on each complaint

One cause on your list fits · the reasoning sits on the case

The assistantCause on complaint CPL-26-30871

Card dispute refund delayed. The customer was promised 10 days and waited six weeks for $389.

source · customer message of 08/14, answer of 08/20
01

The friction

The whole thing in one move

27 percent. That is the share of last quarter's 1,284 closed complaints filed under Other. Every one was answered, most on time. The customer's words sit in the messages, nobody has the evenings to read thirteen hundred of them, and the report to the examiners counts complaints without saying why customers wrote. The quarter is handled case by case and never read as a whole.

The team writes down the causes it believes in, and the AI classifies every closed complaint against them, with its reasoning on the case. Handlers check a sample, the process owner confirms the cause, and the head of complaints signs a report that names it and a plan that fixes it.

Closed complaints, Q3 2026Case management
One cause under three labels. The old field scattered the refund delay across Cards, Payments and Other. The team's cause list gathers it.
Complaint CPL-26-30871AI console
Two fields on every closed case. The cause from the team's list and one sentence on why, here for Laura Mitchell's card dispute.

Dashboards, cause lists, sample reviews: a complaints team knows each of these. Run in this order, they explain a quarter: the causes are written by people, applied to every case, checked on a sample, confirmed at the source, then reported and fixed. The nine steps below go from Dana's dashboard on October 5 to the January figures that show the fix held.

02

The quarter as a whole

Each complaint was handled in its own case. Dana now reads all of them, through causes her team wrote.

Step 1Reporting

The quarter's dashboard, and the size of Other

On October 5, Dana Whitfield, head of complaints, opens her home page for the third quarter. Her reports count the closed complaints across channels, the share answered within your deadlines and the share upheld. In the old category field, 27 percent sit in Other.

Complaints, Q3 2026Case management
Each tile is a report Dana built on the case fields.
The deadline is under control, the cause is notThe team answers on time. What the quarter cannot say yet is why three hundred fifty customers wrote under Other.
Step 2AI console

The team writes its own list of causes

Dana and two senior handlers write twelve causes in plain language, each with a short description and an example: what they see in the messages week after week. She sets them up as a business task in the console, with the instructions, the model and a confidence threshold under which a person reads the case.

Task: classify the complaint causeAI console
The instructions change without a release. The cause list is the team's.
The causes are the team's wordsThe AI applies a list that handlers wrote from experience. Nobody has to accept a grouping invented by a model.
Step 3AI check

Every closed complaint gets its cause

With 1,284 cases, the task runs asynchronously. For each one, the AI reads the customer's messages and the answer sent, and writes two fields: the cause and the reasoning in plain words. 61 complaints fall under the threshold and go to handlers; 23 fit no cause.

Confirmed: 1,200Needs a person: 61Outside the list: 23
Causes, Q3 2026Case management
The largest cause was spread across Cards, Payments and Other in the old field.
Other stops being the biggest lineEvery complaint carries a cause and a sentence saying why. The ones that fit nowhere are counted, and read by a person.
03

The cause is checked with people

Before anything is reported, handlers check the classification and the process owner confirms the cause.

Step 4Assistant

Ask the quarter a plain question

Dana asks the repository in plain words for the upheld card dispute complaints where the answer gave no reason for the delay. The question becomes a query; the result is a table she can export to Excel. Clicking a row opens the case, and she reads three in full.

Ask the repositoryAI console
The answer names its source, and only covers cases Dana is allowed to read.
The evidence is a table of real casesThe manager reads real cases behind the count, in the time it takes to ask.
Step 5Workflow

Two handlers check a sample

Dana sends Malik and Jenna, two handlers, a sample of 40 cases in the largest cause, as review tasks. They confirm 36. Three belong to a merchant who refused the dispute, and one is a fraud claim. They move the four, and the cause settles at 208.

Sample reviewCase management
Each correction carries the handler's name and stays in the case history.
The report rests on checked figuresThe agreement rate is written down before the number goes to the examiners, so the manager can defend it.
Step 6Workflow

Find the stage where the wait starts

Dana meets Nathan Ford, who runs card operations. The card dispute process runs on the workflow engine, so each case shows when it reached each stage. The wait sits after the merchant's acceptance: accepted disputes wait for a weekly batch, and a manual check on amounts above $300 adds two more weeks. Nathan confirms the cause.

Card dispute workflowCase management
The delays come from the case histories. Why the batch exists is Nathan's to explain.
Ask it of your own quarter: how many complaints does a bank answer one by one, for a delay that sits in one stage of one process?
The cause is confirmed where it startsThe person who runs the process agrees with the reading. The fix can now be his, with a date.
04

Reported, then fixed upstream

The cause goes into the report to the examiners, and the fix goes to the person who runs the process.

Step 7Assistant

The report narrative starts from the reviewed figures

Dana fills your quarterly report with the counts from her reports. She asks for a draft of the narrative on the main causes; every figure in it cites the report or the sample it comes from. She rewrites it, Ryan Dupree in compliance reviews it, and she signs.

Quarterly complaints reportWorkspace
The draft gives the structure and the figures. The words and the signature are Dana's.
The report says why customers wroteThe examiners read a cause, its size and its fix, with every figure traceable to the cases behind it.
Step 8Workflow

Each fix becomes a task with an owner

Nathan commits to two changes in card operations: a daily refund run for accepted disputes, and the manual check raised to $1,500. Dana adds hers: the answer draft's instructions now ask it to explain the refund timing, a change made in the console without a release. Compliance and Dana decide the redress for customers who paid overdraft fees while waiting.

Action planCase management
Each action has an owner, a date and an SLA timer.
The fix has a name and a dateThe cause does not stop at a slide. It goes to the person who runs the process, and the plan is followed like any other work.
Step 9Reporting

The next run shows whether it worked

In January, the same classification runs on the complaints closed since the change. Card dispute complaints drop from 69 a month in the third quarter to 18 in January. Dana closes the action with the report attached.

Card dispute complaintsCase management
The same task, the same cause list, the same report. Only the month changes.
The proof comes from the same readingThe cause that was counted is counted again after the fix, the same way. The examiners see the drop and where it came from.
05

The possible results

Twelve causes, written by Dana's team, are the only labels in play. Each closed complaint comes back confirmed, sent to a handler, or flagged as fitting none, and each answer starts different work.

Confirmed

One cause fits

The cause and a sentence of reasoning are written on the case, above your confidence threshold.

Then: the case counts in the cause report, and a sample is checked by handlers.
Needs a person

The AI is not sure enough

Two causes could fit, or the messages are thin. The case goes under the threshold.

Then: a handler reads it and chooses the cause, with their name on it.
Outside the list

No cause on the list fits

A complaint about something the team has not named yet. The AI says so and gives its reasoning.

Then: the head of complaints reads them and decides whether a new cause belongs on the list.
What the AI does not do

It answers no customer. It decides no redress. It signs no report. Across the nine steps on this page, the cause list, the sample check, the report and the fix are decided by people, each with their name on the case.

The AI classifies, searches, summarizes and drafts. The control is in the list the team wrote and the threshold set in the console, so the reading holds when the quarter is two thousand complaints long.

06

In numbers

Counts from one retail quarter of 1,284 closed complaints. With a shorter cause list or a higher threshold, more cases would go to handlers.

1,284 casesclassified on the team's own cause list, each with its reasoning.
40 caseschecked by two handlers before the largest cause went into the report.
0 releasesto change what the answer draft tells customers about refund timing.

Where these numbers come from. They are counted on the scenario described on this page, with example data. They are not a customer result.

Why these numbers hold

Zero nights reading filesProductivityEvery closed complaint is read against the cause list. The manager starts from counts and real cases.
Zero unchecked figuresReporting reliabilityA sample is reviewed and the agreement rate recorded before a number goes to the examiners.
Decision supportPeople decideThe causes, the report and the fixes belong to named people. The AI reads and drafts.
07

The people involved

Complaints belong to the business, get read by compliance and run on IT's systems. Each of the three has to like what they see here.

Business operations

Heads of complaints and process owners

You know the same failure repeats, and proving it takes nights.

  • Every closed complaint carries a cause from your list.
  • A plain question returns the cases behind a count.
  • The stage where the delay starts shows in the case histories.
  • Each fix is a task with its owner and its date.
Risk and compliance

Compliance officers

Examiners get a count of complaints and very little about why.

  • The report names causes, with figures traced to the cases.
  • A sample is checked by handlers before reporting.
  • The AI answers, redresses and signs nothing.
  • Every classification keeps its reasoning and its date.
IT and architecture

CIOs and architects

A new category still costs a change request on the reporting tool.

  • The cause list and the instructions change in the console.
  • Classification runs asynchronously over a whole quarter.
  • Answers only cover cases the user may read.
  • Model chosen per task, hosted where your data rules say.

Where to start

Take the failure your handlers mention most, write it down with one example, and classify last quarter's closed complaints against it. The count that comes back is your first argument.

Your complaint cases, reports and workflows stay as they are. The classification writes two fields onto cases you already have.

See it live

See it on your own complaints

Bring a few hundred anonymized closed complaints and the causes your team believes in. We will classify them with you and see which cause tops the list.

The screens on this page are drawn to show one step at a time, with example data.

Uxopian Software helps organizations put AI to work across the systems they already run, without giving up control of their data, their infrastructure or their critical processes.