Skip to content
clientclue

How it works

Four stages, every week

Importyour file

  • Message Import
  • CSV
  • Excel
  • Paste

Drop in a CSV or Excel file, or paste messages. ClientClue shows what it will drop and what it will hide before anything is read.

Readevery message

  • Detail Hiding
  • Pidgin Reader
  • Hausa
  • Yoruba
  • Igbo

ClientClue AI reads every message and notes its language, its feeling and its kind: problem, question, request, praise or leaving.

Groupwhat repeats

  • Issue Finder
  • Rising Alerts
  • New this week
  • Leaving Reasons

Messages that mean the same thing become one issue with a plain name. Rising and New this week problems get a flag with their numbers.

Reportevery week

  • Weekly Report
  • Ask ClientClue
  • PDF
  • Share link

Each team picks its report day, and a week then runs from that day. When the week ends, your team gets one report: the top issues, what is rising, why customers leave, and quotes checked word for word.

The ten steps

From a file to a report, one step at a time

Code does the counting. ClientClue AI does the reading. Every number comes from code, and every quote is checked word for word before you see it.

  • 1. Read the file

    ClientClue opens your CSV or Excel file and finds the message column, the dates and the ratings.Message Import. Code does this

  • 2. Match the columns

    Headers are matched by name first. ClientClue AI suggests the rest from header names alone. You confirm.Message Import. Code first, then ClientClue AI

  • 3. Clean

    Empty rows, duplicates, agent replies, automatic replies and greetings are dropped. One-word messages such as Slow stay.Code does this

  • 4. Hide personal details

    Names, phones, emails, account, BVN, NIN and card numbers become labels such as [phone] before ClientClue AI reads anything.Detail Hiding. Code does this

  • 5. Find the issues

    ClientClue AI proposes 6 to 12 areas and up to 40 issues from your messages, named in plain words.Issue Finder. ClientClue AI does this

  • 6. Read every message

    ClientClue AI reads each message in batches of about 50: language, feeling, kind, issue, money taken, and the quote that says it best.Pidgin Reader. ClientClue AI does this

  • 7. Check the quotes

    Code looks for every quote word for word in its message. A quote that does not match is dropped.Code does this

  • 8. Count

    Code counts messages by Lagos week, sets the Rising and New this week flags and counts the reasons customers give for leaving.Rising Alerts and Leaving Reasons. Code does this

  • 9. Write the report

    Code picks the issues, numbers and quotes. ClientClue AI writes one plain line per issue, and code checks every number in it.Weekly Report. Code first, then ClientClue AI

  • 10. Ask ClientClue

    ClientClue AI turns your question into filters. Code counts the matching messages. ClientClue AI writes the answer with its quotes.Ask ClientClue. ClientClue AI first, then code

What you can upload

Four kinds of file

  • Support tickets

    Columns it understands: message or description, subject, date, channel, customer reference, status.A help desk export or a spreadsheet your team keeps

  • Cancellations

    Columns it understands: reason, date, amount, product or plan, customer reference.A cancellation form, a call log or a leavers sheet

  • App-store reviews

    Columns it understands: review text, rating, date, channel. A column holding your own replies is left out.A reviews export from your app store

  • Other feedback

    Columns it understands: comment, score, date, channel, product. Survey answers and WhatsApp notes fit here.Surveys, forms and anything else with customer words in it

CSV or Excel, or paste. You do not connect a help desk. Most teams export a spreadsheet, and that is enough to start.

Prepare your file

What you get

The weekly report, part by part

Seven parts

  • The week at a glance: messages read, the negative share, issues found, Rising and New this week, and customers talking about leaving
  • Top issues, ranked by negative messages, each with its count and its change on last week
  • Rising and New this week, each with its numbers, for example 19 this week and 12 a week before
  • Two or three quotes for each issue, from different customers, checked word for word
  • One plain line for each issue and a Check this first line, with every number in them checked against the counts
  • Leaving reasons, each with its count and a quote, and the naira at stake when your file has an amount
  • A PDF, a CSV, a link anyone can open and a summary to paste into WhatsApp
Read your weekly report

See it onyour own file.

Get startedBook a call