Five weekends. One family project. How to pick.

The list on the fridge has been the same list since June. Gutters. The garage door that catches. The upstairs bathroom fan that sounds like a coffee grinder. The dead spot in the yard where the sprinkler head sank. The window in the kids’ room that will not stay open on its own. The list has not gotten shorter over the summer. It has gotten one item longer, because in August the smoke detector in the hallway started chirping at 2 a.m. every third night.

The kids went back to school on the seventeenth of August. It is now the first Tuesday after Labor Day. You and your spouse have — for the first time since May — actual Saturdays. Which means the fridge list is on the clock, and the clock runs out at Thanksgiving.

This post is about that clock, why it is shorter than it looks in 2026, and how to pick the one thing on the list that will actually get done before the house is decorated for Christmas.

Why the fall window is narrower this year

Three things about the 2026 fall — none of them subtle if you have been trying to schedule anything.

Contractors are booking further out than they were. The Bureau of Labor Statistics’ latest occupational outlook for construction trades shows the same story it has shown for the last three years: retirements outpacing new entrants in every skilled trade, with roofers, HVAC technicians, and electricians at the top of the tightening curve. In practical terms: the general contractor who could put you on the calendar in three weeks in 2022 is putting you on the calendar in eight weeks in 2026, and the eight weeks he quotes you this Tuesday runs out on the first full weekend in November. If you call Halloween, he is booking into January.

Hurricane season peaks right now. NOAA’s National Hurricane Center puts peak Atlantic activity in August through October, and the 2026 season outlook called for above-normal activity again, on top of a decade of above-normal seasons. If you are anywhere from the Carolinas to the Gulf — or, increasingly, anywhere in the Northeast where the remnants make landfall — the first three weekends on the list below are the ones a named storm can eat. Any project that requires the roof open or the yard torn up is riskier per calendar week in September and October than it is in April.

The Q4 spending crush is closer than parents remember. By the second week of November, most families’ discretionary cash is already committed — the National Retail Federation’s holiday spending survey has run at or above $900 per household in average holiday outlays for four straight years, and that is before travel, tuition second-half payments, or the January insurance renewal. A $6,000 furnace tune-up-turned-replacement in early October is a different family financial event than the same $6,000 hitting on December 5. The dollars are the same. The room to absorb them is not.

Put together, that is why the fridge list needs a decision this week — not because the items are more urgent than they were in June, but because the calendar is more expensive than it was in June.

The five weekends, on the actual calendar

There are eleven weekends between Labor Day and Thanksgiving. These are the five a family with school-age kids can realistically spend on the house — every other one starting the twelfth of September, because the weekends in between are already spoken for by the games, the birthday parties, and the one Saturday a month where nobody schedules anything at all. The last two, mid-to-late November, are already gone to half-light and holiday prep. Here they are, with the constraints attached, on the 2026 calendar.

Weekend 1 (September 12–13). First full weekend after Labor Day. School routines have set. Weather is still permissive for exterior work almost everywhere. Contractors are back from summer vacation and are quoting their fall books.

Weekend 2 (September 26–27). Peak hurricane window on the Atlantic. Ragweed pollen peak in the Midwest. Youth-sports schedules are fully underway; most families have one or two committed days.

Weekend 3 (October 10–11). Columbus Day / Indigenous Peoples’ Day long weekend for most school districts. Historically the best-attended fall weekend for family projects; also the last one where daylight extends past 6:30 p.m. in most of the country.

Weekend 4 (October 24–25). Last full weekend before Halloween. Contractors are starting to close their fall books. Exterior painting temperature margins get tight north of I-70.

Weekend 5 (November 7–8). After the time change. Daylight ends at 5 p.m. Most contractors will not start a new exterior job this weekend unless it is small and rain-day-proof. This is realistically your last window for anything that touches roof, siding, or paint.

Thanksgiving is November 26. Anything not finished by November 8 either finishes indoors in half-light or waits until March.

Five weekends. That is the actual window. Not “September through November” — eleven weekends on the calendar, five you can actually spend. Five.

The framework: pick the one that fails hardest if you skip it

Here is the family version of the framework we use on every project decision. It is the same framework as the roof/kitchen/HVAC sequencing post, applied to a shorter calendar and a smaller list.

Ask three questions about every item on the fridge list. Score each 1–3.

1. What does it cost if I do nothing between now and April?

• A chirping smoke detector costs a battery and thirty seconds — score it 1.
• A garage door with a bad opener costs an inconvenience per day — score it 1.
• A gutter run that is already sagging costs a fascia board and possibly a ceiling stain if the November rains find it — score it 3.
• A furnace that had a hard start last February costs — potentially — a January no-heat call at $1,800 for weekend emergency service, or a full replacement forced by the failure — score it 3.

The 3s are the projects the calendar is actually asking you about. The 1s are not fall projects; they are winter-inside projects, or fifteen-minute-Saturday projects, or hire-a-handyman-in-February projects.

2. Is this project weather-dependent, and does the weather run out this fall?

• Exterior paint north of the Mason–Dixon line: weather runs out mid-October. Weekend 3 at the latest.
• Roof work: weather-dependent but rain-day-tolerant with a decent crew; runs to Weekend 4 in most of the country, Weekend 5 in the South.
• HVAC replacement: not weather-dependent for the install, but pre-winter parts availability tightens weekly through November. Ideally Weekends 1–3.
• Interior projects — bathroom fan, GFCIs, drywall — are not weather-dependent. They can wait for December or January and cost you nothing to defer.

The weather-dependent items either happen in the fall window or they roll to April. If they roll to April, they compete with the spring list, which is longer than the fall list.

3. Does this project cascade if it fails?

This is the question that separates a fridge-list nuisance from a fall-window emergency. A gutter that fails takes out the fascia, the soffit, and sometimes the corner of the living room ceiling. A roof leak takes out the attic insulation, the drywall, and the master bedroom carpet in a specific and predictable order. A furnace that fails on the coldest weekend of the year takes out frozen pipes if you are traveling, or a $500 hotel bill and two nights of grumpy children if you are home.

A garage door opener does not cascade. It just annoys you.

Rank the list by the sum of the three scores. The single highest-scoring item is your fall project. Not the top three. The top one.

Why one and not three

Every family has a version of the “we’ll knock out four things in October” conversation. It does not survive contact with the actual October.

The reason is not laziness or scheduling. The reason is that the coordination cost of a home project — finding the contractor, getting the quote, comparing the quote, opening the wall, discovering the surprise inside the wall, negotiating the change order, paying the deposit, being home for the walk-through, and cleaning up after — is not linear in the number of projects. It is roughly quadratic. Two projects is not twice one project. Two projects is three or four project-weeks of parent-hours, split across five weekends that also contain three soccer games, one flag-football tournament, one back-to-school night per kid, and a family birthday.

The family that picks one and finishes one in October has a house that is one thing better in November. The family that picks four and starts three has a house with three half-projects, three tarps, and one contractor who will not return calls.

Pick one. Finish one.

The worked example

The Kims are a family of four outside Cleveland. Two kids, second and fifth grade. Dad works hybrid; mom is a nurse on a rotating schedule. Their fridge list this September:

1. Chirping hallway smoke detector.
2. Bathroom fan that sounds like a coffee grinder.
3. Furnace made a screeching noise on its first cold night in February. Ran fine the rest of the winter.
4. Gutter on the north side sags visibly, especially after rain.
5. Kids’ bedroom window that will not stay open — sash cord broken.

They score, together, over coffee on the Saturday after Labor Day:

The gutter wins. Not because it is the most expensive project, but because it is the one where doing nothing between now and April costs the most, the weather runs out first, and the cascade if it fails is the worst.

They call three gutter contractors on Monday of Week 1. Two respond by Wednesday. They pick the one who can come the following Tuesday. New seamless aluminum, one full replacement run on the north side, fascia repair where needed, half-day job — $1,400. Done by Weekend 2.

The furnace, the second-highest-scoring item, is on the list for the following Saturday: they book an HVAC tune-up and diagnostic on Weekend 2, decide from the report whether Weekends 3 or 4 is a repair-or-replace weekend, and either way, it is on the calendar before Halloween.

The smoke detector, the fan, and the window sash go on a separate list — the “November-through-February indoor Saturday” list. They will get done because they are cheap and easy and do not compete with the calendar the fall is running.

The Kims did not do everything. They did the one thing that failed hardest if they didn’t, and they scheduled the second thing before the third contractor stopped answering the phone. Their fridge list in December has three items on it, all indoor, all deferrable, none cascading.

What to do this week

By Friday of this week, three things.

Write the actual list. Not on the fridge — on paper, at the kitchen table, both parents, twenty minutes. Every deferred item since March. Not editorialized. Not prioritized yet. Just the list.

Score it with the three questions above. Twenty more minutes. Do this once, together. The purpose is not to rank the items you already knew were important. The purpose is to catch the item that did not look important but scores an 8 because the cascade risk was hiding.

Call the contractor for the top-scoring item Monday morning. Not next Monday. This Monday. If the top-scoring item is the furnace or the roof or anything else weather- or capacity-constrained, you are calling into the eight-week window and every week you wait is a weekend you lose off the back of the calendar.

Everything else on the list either has a home in the December-through-February indoor-project slot, or it moves to the spring list and gets triaged there.

The five weekends will happen whether you use them or not. If you pick one project and finish it, the house is one thing better by Thanksgiving. If you pick four, the house is four things half-done, three tarps in the yard, and a family that will remember this fall as “the fall we tried to do the house.”

Pick one. That is the whole framework.