Fictional scenario. Synthetic data. Your notes stay on this browser; do not enter real customer information.
Thirty prompts, one a day, each with one idea, one thing to do today on your own data or with your own colleagues, and one thing to read. Week one is about seeing clearly and writing it down. Nothing here needs a tool you do not already have.
Day 01 — Write down what you saw.
The habit everything else depends on is the plain observation, written before the explanation. "Forty-one refunds on Friday between 18:00 and 19:00" is an observation. "Someone is gaming the refund queue" is not yet anything.
Today: open whichever queue you worked yesterday and pick one item you decided on. Write three lines about it in a notebook, physical or otherwise: what you saw, what you decided, what it would cost if you were wrong. No adjectives.
Read: the Volume 01 lesson Evidence before action, which is the whole of this week in nine minutes.
Day 02 — Separate the inference.
Every note has two kinds of sentence in it. The ones that describe what is in the data, and the ones that describe what you think it means. Both are necessary. Mixing them is the commonest way a case note misleads the next reader, usually you, three weeks later.
Today: take yesterday's three lines and mark each sentence O or I in the margin. If you cannot decide, it is an inference wearing an observation's clothes. Rewrite it so that it is honest about which it is.
Read: Write the note for the person who reopens it. covers the layout we use.
Day 03 — Read one decline export.
A decline is a message from an issuer, and most desks never read it. The reason codes behind "declined" separate cards that do not exist from cards whose issuer said no from cards whose holder has no money, and those three things have three different meanings for the pattern you are looking at.
Today: export one day of declines for your busiest merchant or product and count them by reason. Write down the top three reasons and their share. If you do not know what one of them means, ask someone today, and write down the answer.
Read: What a decline is trying to tell you.
Day 04 — Keep an alternative alive.
An inference without an alternative is a conclusion in disguise. The desk's discipline is to write, under every "this looks like X", the innocent thing that would also look like X, and then say which field would tell them apart.
Today: for the item you noted on day one, write the honest alternative. A marketing send. A courier failure. A shared household. Then write the single column of the export you would check to rule it in or out. Check it, if you can.
Read: the Volume 01 case False positive, where the alternative turned out to be the answer.
Day 05 — Write the note for the person who reopens it.
A case note is not a diary. It is written for someone who has never seen the case, has ten minutes, and needs to know what was observed, what was decided, who decided it, and when it should be looked at again. the person who reopens it is most often the author, who has forgotten.
Today: find a case you closed more than a month ago and read the note cold. Time how long it takes to understand what happened. If it is more than five minutes, rewrite it in the four-part form: observed, inferred, decided, review.
Read: look up cohort, because the review date usually belongs to one.
Day 06 — Who saw it first.
Most incidents have a first observer, and it is rarely the fraud desk. It is the support agent who noticed the complaints, the finance person who noticed the refunds, or the merchant who rang. The desk that knows who saw it first can ask them next time.
Today: pick the last incident your team handled and write down who noticed it, how long before the desk knew, and what they did with it. Then ask that person one question: what would have made them tell you sooner?
Read: The first fifteen minutes.
Day 07 — Read the week back.
Six days of notes is enough to see your own habits. Which sentences keep turning up. Which alternative you never write. Which cost you never price.
Today: read the week's notes in one sitting and write one sentence about what you noticed about yourself as a writer of notes. That sentence is the first entry in a second notebook, the one about how you work rather than what you worked on. Keep it short. You will add to it on day fourteen.
Read: the Volume 01 decision study The quiet queue, for a week-end exercise in deciding with incomplete notes.
Fictional learning material. Any numerical examples are synthetic. Use only authorised, appropriately minimised records for workplace exercises; do not upload customer information to this learning site.