On a Sunday in October it becomes obvious that the portfolio is far too concentrated, and you write down five steps. Cut the smallcap fund from ₹6,00,000 to ₹2,00,000. Put the ₹4,00,000 into a short-duration debt fund. Raise the monthly SIP from ₹15,000 to ₹22,000. Add a nominee to the demat account. Close the dormant trading account with the second broker. That evening you do the first one. The redemption takes a few working days to reach the bank, by which time it is Wednesday and work is difficult, and the ₹4,00,000 sits in the savings account. It is still sitting there in March. What the household owns in March is not the old portfolio and it is not the new one. It is a third arrangement that nobody designed, that nobody holds an opinion about, and that will not be reviewed by anybody — because in everyone’s head the plan is in progress, and in-progress is the one state that never gets graded.
The renovation was decided in one evening and started on a Saturday. The old tiles came off, the new ones went onto one wall, and then the mason’s other job started and he stopped answering. The kitchen is now worse than it was before anybody touched it: no old tiles, no new ones, and a family cooking around a half-wall for five months. Nobody chose that kitchen. It is not a design, it is a stopping point.
A rebalancing halted after step one is the kitchen with one wall tiled — with one difference that matters. The family sees the kitchen every day and is irritated by it every day. Nobody sees the portfolio, so nothing pushes the plan back to the top of anybody’s list.
Partial completion is not random
The comfortable reading of the story above is that the household ran out of time, and that the fix is discipline. That reading is wrong and it guarantees a repeat, because the steps that get done and the steps that do not are not drawn at random from the list. The ones that survive are quick, pleasant and reversible. The ones that stall crystallise a tax, admit a mistake, need a form, or cannot be undone. And the value of a plan almost always sits in the second group — which means a half-finished plan is reliably the wrong half, not an average half.
| The step | What doing it actually takes | How often it gets done |
|---|---|---|
| Buy the new fund or the new stock | Two minutes in an app that is already open, and it feels like progress | Almost always, and usually first |
| Raise a SIP amount | A few taps. The pain is spread across future months and never appears as one event | Usually |
| Sell the thing being replaced | A tax event, possibly an [[exit load]], and a loss to admit out loud — plus a decision that cannot be un-taken | Often not |
| Add or update a nominee | A form, a login that has expired, sometimes a signature and a witness | Rarely — and the delay costs nothing at all until the day it costs everything |
| Cancel the dormant account or the unwanted policy | A telephone call, a hold queue, and somebody whose job is to talk you out of it | Rarely |
| Write the plan down in the first place | Twenty minutes, and the discomfort of being specific about amounts and dates | Almost never, which is why nobody can say afterwards which steps were skipped |
Sequencing rules that survive being abandoned
- Do the irreversible, unpleasant step first, while it is still voluntary. It is the one that will not survive being postponed, and doing it first also makes the rest of the plan self-funding rather than savings-funded.
- One absolute exception: never cancel a protection you still need before its replacement is in force. Cover is the case where the uncomfortable stopping point is not a spur to finish but a period of genuine exposure, and where the replacement may be refused. Order for discomfort everywhere else; order for continuity of cover wherever cover is involved.
- Make every stopping point a state you would accept living in for a year, because that is roughly how long "in progress" actually lasts once the first weekend has passed.
- Give every step its own date rather than a number in a list. A numbered list quietly implies that step three cannot begin until step two is finished, which turns one stalled step into a stalled plan. Five dates are five independent commitments and fail one at a time.
- Write down where you stopped. The five-month gap is almost never caused by forgetting the plan. It is caused by not remembering which parts of it happened, which makes picking it up again a research task rather than a two-minute one.
- Review the plan as one decision, on the last date, not the first. Otherwise the first completed step is experienced as success and the plan is filed as done.
Fourteen months to a school admission payment
You have decided to move ₹5,00,000 out of a mid-cap fund and into a short-duration debt fund, because a school admission payment falls due in fourteen months. It is Monday morning. Which order do you do it in?
Why is the half of a plan that gets done so rarely a random half?
Ravivaar ko plan bana: teen mehnge largecap fund (₹9,00,000) band karke ek sasta index fund. Usi raat pehla kadam ho gaya — ₹6,00,000 index fund mein daal diye, savings account se, kyunki redemption ke paise aane mein do-teen din lagte hain. Doosra kadam aaj tak nahi hua: ek fund mein faayda hai toh tax lagega, doosre mein nuksaan hai toh maanna padega ki chunav galat tha. Nateeja: equity ₹9,00,000 se ₹15,00,000, holding 3 se 4, savings ₹6,00,000 khaali — aur jo problem thi woh jyon ki tyon. Ghar mein sab kehte hain "hum index fund mein shift kar rahe hain", aur yahi line kisi ko dobara dekhne nahi deti. Jo kadam aasaan, achha lagne wala aur wapas-badla-ja-sakne wala hai wahi hota hai; jo tax kaatta hai ya galti manwata hai woh nahi. Isliye pehle bechna, phir khareedna — tab agar ruke toh savings mein pada bada sa paisa khud chubhega. Ek hi apwad: insurance kabhi naya chalu hone se pehle purana band mat karna.
- Every multi-step plan has intermediate states you will actually own, often for months, and nobody designs them.
- Partial completion is not random — pleasant and reversible steps survive, and the plan’s value is usually in the others.
- Buy-then-sell turns a swap into an addition; sell-then-buy leaves a stopping point you cannot ignore.
- The one exception is protection: never cancel cover before its replacement is in force.
- Give each step its own date, and write down where you stopped.
Mark it done to track your progress through the curriculum.