You will be able to map a task step by step and remove the steps that exist only because of habit.
Mei Ling runs the front desk at a tuition centre in Tampines. You met her in lesson 1.1. When a parent fills in the enquiry form on the centre's website, she opens the notification email and copies the parent's name and number into the enquiry sheet. She then copies them again into the class register "just in case", checks the timetable and writes a reply. Last, she writes the same details on a sticky note for the teacher who will call back. Each enquiry takes her about six minutes. That figure, like the others about her below, is an example.
After scoring the task in lesson 1.1, her first instinct was to hand all of it to an automation tool. Most people feel the same pull. Before you build anything, look closely at the steps, because an automation copies a process exactly and does not ask why a step exists.
If a process has a pointless step, the automation runs it every time. Because it now happens out of sight, the step also becomes harder to notice. Every step in an automation is a place where something can fail. Take a step that writes to the class register. It needs a connection to that file, a mapping of the right columns, and a format the file accepts. If someone changes the register's columns, the flow breaks. An automated messy process runs faster and fails in more places. When you cut a step before building, you remove one possible failure permanently.
The method for finding those steps is simple and slightly tedious, which is why most people skip it.
Write the task out as a numbered list. Each step is a verb-and-object step, an action and its target, short and literal: "copy name to enquiry sheet", "check timetable", "send reply", "write sticky note". If you can't describe a step in a few words, it is probably two steps, so split it.
Mei Ling wrote her list like this:
Step 1: Open notification email. Step 2: Copy name and number to enquiry sheet. Step 3: Copy name and number to class register. Step 4: Check timetable for the child's level. Step 5: Write reply to parent. Step 6: Write sticky note for teacher. Step 7: Hand sticky note to teacher.
Next, go down the list and ask three things about each step: who uses its result, what would go wrong if it disappeared tomorrow, and whether the same information is already kept somewhere else.
Step 3 failed straight away. The class register is for enrolled students, and most enquiries never enrol. The "just in case" copy was a habit left over from before the enquiry sheet existed. Steps 6 and 7 failed as well, because the teacher could read the enquiry sheet directly if it had a column saying who should call back. Cutting steps 3, 6 and 7 left four steps. One of those, opening the email, goes away once the form writes to the sheet by itself. Her form also asked for the parent's occupation, which nobody had read in two years.
Almost every office has the same three kinds of waste: duplicate data entry, unused fields, and approval steps nobody reads.
Duplicate data entry means the same name, number or amount is typed into two or three places. Usually each place belongs to a different person or was set up at a different time. Mei Ling's copy into the class register is an example. One record that everyone reads is better than three copies that slowly drift apart.
Unused fields are the things forms and sheets collect that nobody looks at, like the parent's occupation on Mei Ling's form. Every extra field is one more thing to map, check and protect. If it holds personal data, it is also one more thing you are responsible for keeping safe.
Approval steps nobody reads are sign-offs or copy-ins that are always given or never acted on. A manager copied on every reply "for visibility" is one. A form sign-off that is always given is another. If the approver has never once said no, the step is not doing any checking. Keep approvals where they actually catch mistakes. Lesson 1.3, "Know which work should stay manual", looks at where a human check matters, and module 7 shows how to build a human check that people actually use.
Be careful with steps that look pointless. Some exist for a reason nobody wrote down, such as an audit, a legal record or a colleague's monthly report. Before you remove a step that involves someone else, ask them what they use it for. A two-minute conversation now saves an awkward one later.
After the cuts, Mei Ling timed her new routine: copy details into the enquiry sheet, check the timetable, send the reply, tick a column for the teacher. It took about three minutes instead of six. Lesson 1.1 used an example figure of fifteen enquiries a week. At that rate, 15 enquiries x 3 minutes saved = 45 minutes saved every week, with no software involved.
Sometimes a shorter manual process is the whole fix, and this is a common result. If the remaining steps take only a few minutes a week, deciding not to automate at all is a good outcome. If the task is still worth automating, the cleaner list means fewer connections, fewer fields to map and fewer places to fail.
You already have two candidates, the tasks you circled at the end of lesson 1.1. Pick the one you are more likely to build. In the activity below, write its steps out in full as a numbered list, with a verb and an object for each, and split any step that takes more than a few words. Ask the three questions of every step. Look especially for duplicate data entry, unused fields and approval steps nobody reads, and check with anyone else involved before you drop a step that affects them. Cross out every step you could drop with no software at all. Then time or judge the routine that is left and decide whether it is still worth automating. Not automating is an acceptable answer. Keep the crossed-out version, because it becomes the starting point for your diagram in module 2.
Write out the steps of one of your top two tasks and cross out every step you could remove without any software.
Junxiong-WFG Organisation is an authorised representative of AIA Financial Advisers Private Limited (Reg. No. 201715016G).