You will be able to check progress and correct course while leaving ownership with the other person.
Two weeks after handing Hui Min the monthly service report, Wei Ling opens the draft and sees problems. The figures for one large client look wrong, two charts are missing, and it's due to Daniel tomorrow. She knows exactly how to fix it. It would take her an hour tonight. She could quietly correct it, send it off, and spare everyone the awkward conversation.
That hour feels like a kindness. It is how delegation fails. Hui Min would learn that her work gets redone without her, Wei Ling would be back doing a task she handed over, and next month the same problems would appear.
Andy Grove put it bluntly in High Output Management: delegation without follow-through is abdication. Handing over a task does not end your responsibility for the result. You still own it to your own manager. What changes is how you exercise that responsibility, by checking and coaching instead of doing.
Grove compared it to quality checks in production. You monitor delegated work the way an inspector checks a process: more closely early on and when the person is new to the task, less closely as they show they can handle it. That ties directly to lesson 3.2, Match your involvement to their experience. Low experience means more checks, high experience means fewer.
The easiest way to follow up without seeming suspicious is to agree when you will check before the work begins. When the checkpoints are part of the plan, a check-in is just the next step. When they are not, every unannounced look at someone's work feels like a sign you don't trust them.
For the service report, a sensible set of checkpoints would have been these: a quick look at the data after Hui Min pulled it, in the first week; a review of the draft three working days before the deadline; and a final read on the day before. Agreeing those at the handover makes it easy to say "it's draft review day, can I see where you are?"
Wei Ling had agreed only one check: the final read, the night before. That left no room to fix anything properly, which is why she was tempted to fix it herself.
Put checkpoints early enough that there is time to change course. A check the day before a deadline can only catch a problem. It can't help anyone solve one.
When you find work drifting off track, start with questions. You are trying to understand what happened and help the person find the fix, not to show them that you would have done it better.
Useful questions include: "Talk me through where you are with it." "What have you tried so far?" "Which part was hardest?" "What options do you see?" "What would you do if I weren't here?"
When Wei Ling asked Hui Min about the client figures, the answer was simple. The client had changed its account codes mid-month, and the report merged the old and new codes incorrectly. Hui Min had noticed the totals looked odd but assumed she had misread something and didn't want to bother Wei Ling. That told Wei Ling two things: Hui Min could spot problems, and she didn't yet feel free to raise them. One question bought far more than an hour of fixing would have.
She then asked Hui Min what she would do. Hui Min suggested asking the client's finance contact to confirm the mapping. It was a better fix than Wei Ling's, who would have guessed.
Sometimes you have to take work back. The test is the cost of failure. If letting the person continue risks something serious, such as a client relationship, a legal or safety problem, or a commitment your company cannot walk back, step in.
If it is only that the work will be slower, rougher, or different from how you would have done it, leave it with them. That cost is the price of the person learning.
When you do take a task back, explain why, and make it clear whether it is about this situation or about them. "Daniel has moved the deadline to this afternoon, so I'm going to finish this one. Let's sit down next week and go through what happened, and you'll do next month's." That sentence keeps the ownership with Hui Min even while Wei Ling steps in.
What you should never do is take work back silently, by redoing it overnight and sending it without a word. The person will find out, and they will draw their own conclusion about whether you trust them.
Wei Ling did not redo the report. She moved the deadline to Daniel by one day, which he accepted without fuss, and Hui Min fixed it with the client's help. They also added two checkpoints for the following month. The next report arrived with one small error, which Hui Min flagged herself.
Questions only help if you have them ready when the drift happens, because the urge to fix it yourself arrives fast. Write yours now, before the next task goes off course.
Write three coaching questions you will ask the next time a delegated task drifts, instead of fixing it yourself.
Junxiong-WFG Organisation is an authorised representative of AIA Financial Advisers Private Limited (Reg. No. 201715016G).