How Big Is Big Enough?

Thursday, Nov 5, 2026. 9:47 PM

Anika is on her couch with a bowl of cereal she is calling dinner. Her roommate's door is shut. The company group chat, #ops-random, usually carries memes about the warehouse printer. Tonight it carries a screenshot of a calendar invite: "Company Update, Monday, 10:00 a.m. Attendance required." Under it, forty-one replies. One says: my last startup did this the week before. Another: don't panic, panic later.

She opens her Ledger app. Then she closes it. Then she opens it again.

Sunday, mid-afternoon. She lets herself into her mother's house in Fairfax with the screenshot still open on her phone, and Bayes, the orange cat, watches her cross the kitchen like a customs officer.

The Question

Anika: Okay, so I have a group chat that is either nothing or a funeral, and I have $0 in my buffer. First deposit fires after my next check. Everyone says three to six months. Why three to six? Who decided?

Revathi: Nobody decided. It got repeated until it looked like a measurement.

Anika: Be so for real, what are the odds I need it?

Revathi: Good, Annie. That is two questions, and the slogan answers neither. How likely is it that you lose the job? And if you do, how long until the next paycheck? Pass me that pencil. Call it.

Revathi's Frame

Revathi is at the stove, onions going soft in the pan, a pot of dal behind them. She talks without turning around.

Revathi: Think of the weather. A forecast of rain doesn't tell you whether to carry an umbrella. You need the chance of rain, and how much you'd mind getting wet. A buffer is the same. One number is the chance the storm comes. The other is how long it lasts. Three months of cash is an umbrella that covers a short shower.

Anika: So the buffer isn't for the job loss. It's for the length of the search.

Revathi: Right. A layoff you can recover from in two months is an inconvenience. One that runs nine months is a different event. You are not buying protection against the first part. You are pricing the tail of the second.

Anika: Say it in odds.

Revathi: That's my line.

Anika: I'm borrowing it. I'll bring it back.

The Napkin

Revathi writes the assumptions first, then slides the napkin across the yellow table.

Assumptions:

• Anika's chance of losing this job in the next 12 months: 12%.
• If she is laid off, how long the search runs (a guess, not a statistic): 55% find work within 3 months; 30% take 4 to 6 months; 10% take 7 to 9; 5% take longer than 9.
• Her essentials, rent to the bare minimum, in the table below.

Essential, per month	Amount
Her half of rent	$1,650
Utilities and internet	$140
Groceries	$400
Car insurance	$150
Gas	$110
Phone	$60
Student loan minimums (rough)	$260
Renters insurance	$15
Household basics	$70
Prescriptions and copays	$45
Essentials total	$2,900

Survival number = monthly essentials x months.

Months covered	Survival number	Months of saving at $500	Chance a search outlasts it	Chance per year you are laid off and still short
3	$8,700	17.4	45%	5.4%
6	$17,400	34.8	15%	1.8%
9	$26,100	52.2	5%	0.6%

Revathi: Read the last column like this. Twelve percent times 45 percent is 5.4. With no buffer at all, every layoff leaves you short: 12 percent. Three months cuts that by 6.6 points. Six cuts 3.6 more. Nine cuts 1.2 more. Each extra $8,700 buys less. And if the 12 is really 24, the whole column doubles and the order does not change.

Anika: Translate it to rent.

Revathi: Three months is a bit over five months of your half. Six is about ten and a half.

Anika: Okay, that actually makes sense, I hate it.

Revathi: Your spending averages $4,400. Essentials are $2,900. That gap is $1,500 of things you would cut in a bad month and be annoyed about. The survival number covers the floor, not the life.

Anika: Wait. Health insurance. My premium comes out of my paycheck now. If I'm laid off, I pay it myself.

Revathi stops stirring. She takes the pencil back and writes a star at the bottom of the napkin.

Revathi: Which means the real essentials are higher. How much higher?

Anika: I have no idea.

Revathi: Then the table has a star on it. It also assumes no severance and no benefits. Both would shrink the need. That's a second star. Honest forecasts come with stars.

The Disagreement

Anika: Here's my problem. At $500 a month, six months is almost three years of saving. Nine is more than four. By then I'm either rich or I've moved. Meanwhile I have loans. This is overkill for someone with a bedroom at her mom's house twenty minutes away.

Revathi: That is not nothing. A room at my house is an asset. I did not put it in the model.

Anika: Because you were embarrassed to price it.

Revathi: Because I don't want you to need it. But fine, it's real. It lowers the number you'd need, and it has a cost: you'd be sleeping in the room you left. Some people would spend savings to avoid that. Put a star on it, not a subtraction.

Anika: So the model says nine months and my gut says one.

Revathi: The model says what the odds say, and the odds are a guess. My search numbers come from how I picture a search. You just finished one. Maybe mine are too kind.

Anika: Mine ran most of the summer. That's one data point, so I'm not touching your 55%. But it's not nothing. Okay, one more thing your napkin missed. Every layoff story in my feed ends with someone's login dying mid-sentence. That's my feed, not data, and it only shows the loud ones. But a startup can shut off your access before you finish reading the email. The 12% says nothing about how fast it happens.

Revathi: That is a real correction. Warning time matters, and I have no number for it. Neither do you. Star it.

Anika: And you don't have one for your own, do you? The forty thousand.

The kitchen gets quiet. Bayes, the orange cat, steps off the windowsill and onto a chair, unmoved by evidence.

Revathi: No. That figure was never sized by any odds. I picked a number that felt like safety and left it there. Pass the rice.

Anika lets it go. She also writes it down, with a timestamp.

Monday Move

Under 30 minutes. Do this once; revisit it each time your life changes.

1. Open your last three months of bank statements. Circle only the things you would pay even if you had no income: rent, utilities, groceries, insurance, phone, loan minimums, medications.
2. Add them. That is your monthly essentials.
3. Multiply by 3, 6 and 9. Those are your survival numbers.
4. Find one thing you left out (health coverage after a job loss is a common one) and write it next to the total with a star. [CHECK CURRENT: unemployment benefit rules and amounts for your state]
5. Write one sentence: "My first target is ___ months, and I will review it on ___." This is a forecast about your own life, not advice. Your numbers will differ.

The Ledger

On the drive home, parked outside her building, Anika adds two entries before she can talk herself out of it.

#	Prediction	Probability	Resolves
4	Her company announces layoffs of any kind	Anika 30%, Revathi 25%	Jan 31, 2027
5	Her own role is cut (laid off, not quit)	Anika 12%	Nov 1, 2027

Prediction 5 sits inside her existing 30% for "not still at this job" on Nov 1, 2027. The other 18 points are for quitting or moving on.

She also logs her decision: first target $1,000 (about a third of one month of essentials, 0.34), then $8,700 as the three-month number. She will review the longer target in six months, using whatever the Monday all-hands turns into.

Process note: if the all-hands is routine, she will not call the prediction wrong. A 30% event not happening is a 30% event not happening.

At 11:52 p.m., #ops-random lights up once more: does anyone know if there will be bagels? Anika types nothing. She sets her phone face down.