Root cause analysis for complaints in five questions

A faster way into root cause analysis

Root cause analysis for complaints does not need a workshop to start. Five questions, answered from the reviews themselves, usually narrow the cause enough to act or to know what to check next. They work because most recurring complaints have a pattern in place and time that points at the cause.

The questions are deliberately simple. The discipline is in answering each with evidence, such as counts, dates and quotes, rather than opinions.

1. What exactly is the complaint?

State it in the customer's terms and as narrowly as the reviews allow. "Service" is too broad; "waiting more than 20 minutes to pay on weekend evenings" is a complaint you can investigate. Read twenty reviews and write one sentence.

2. Where does it happen, and where doesn't it?

List the stores, cities, formats, channels or app versions where it appears, and those where it does not. The contrast is the strongest clue. A complaint present in mall stores but not high-street ones points to something mall stores share.

3. When does it happen?

Times of day, days of the week, seasons, events. Reviews often say. A complaint that only appears on weekend evenings points to staffing or capacity at those times.

4. Since when?

Find the week the complaint started or jumped. Compare it with the dates of changes: new systems, policies, suppliers, staff, store refits, app releases. Most recurring complaints begin with a change.

5. Why there, then?

Now ask what is different about those places at those times since that date. The answer is usually concrete: one counter instead of two, a new approval rule, a different contractor, a shorter app timeout. That is the likely cause.

Test before you fix

Check the likely cause against the places where the complaint does not happen. If stores with two counters at weekends have no queue complaints, the counter explanation holds. If they have the same complaints, look again.

An example

What: exchanges refused for resized rings. Where: 12 Mumbai stores, not Pune or Bengaluru. When: any day. Since when: the first week of July. Why there, then: a Mumbai-only policy update requiring head-office approval for resized items. Test: Pune stores without the rule have no such complaints. Fix: delegate approval to store managers; check in two months.

Common mistakes

• Starting with opinions instead of reviews.
• Stopping at a person rather than a process.
• Skipping the where-it-doesn't-happen question.
• Never checking after the fix.

Common questions

Checklist

• Complaint stated narrowly.
• Where and where not.
• When.
• Since when, matched with changes.
• Cause tested against contrasts.

In short

Ask what, where and where not, when, since when, and why there and then. Answer each from the reviews, test the cause against the contrasts, then fix and measure.

Doing this with Falcon

Falcon labels each topic in scope as needing attention, emerging or a strength. Each insight names the issue, why it matters and a next step that names the taxonomy owner; actions can be tracked to done, and because reviews keep being classified, the same signals show whether the issue's negative share falls afterwards.