How To Choose Dinner Faster — Stop the 6pm Spiral

Dinner is uniquely positioned to be the most expensive recurring decision of the day. Three reasons compound:

• It's made late. By 5-6pm, both household members are decision-depleted from their working day. Quality of decision-making degrades sharply across the afternoon (the parole-board research is the classic illustration).
• It's multi-stakeholder. "I don't mind, what do you want?" multiplied by however many people are eating. Each person is partially deferring to the others, so no decision emerges.
• The option space is unbounded. "What's for dinner?" has thousands of valid answers — from restaurant out, to takeaway in, to grocery-shop-and-cook, to leftover-eat. The lack of a default makes every option require evaluation.

Fix any one of those three and the time-cost drops. Fix all three and the decision becomes routine. This is the goal.

The most underrated household productivity move: theme the days of the week. Examples that actually work:

• Monday — pasta night. Whatever pasta. Decision space reduced to choosing the sauce.
• Tuesday — taco / Mexican night. Same logic. Tacos, fajitas, quesadillas, rice bowls.
• Wednesday — recipe box night. Whatever Gousto / HelloFresh sent. Pre-decided by the box.
• Thursday — leftovers night. Forces use of what's in the fridge before it spoils.
• Friday — fakeaway night. Pizza, curry, or burger — but cooked at home. Or takeaway if you've had a week.
• Saturday — date / try-new-thing night. The one night with full decision freedom.
• Sunday — roast / slow-cook. The decision is "roast what?" not "what to cook?"

The framework reduces the 7-day "what shall we eat?" question to "what sauce for the pasta tonight?" — a 30-second decision instead of a 20-minute one. Studies on themed-day routines show they stick once a household runs them for 3+ weeks.

For households resistant to themed days, the alternative is constrained randomness. Critical word: constrained. A picker that randomises from 1,000 dishes is just deferred decision-making — you still have to evaluate the output. A picker that randomises from 12 hand-picked "dishes we like" is genuinely useful.

Our What Should We Eat? tool defaults to a curated list but accepts custom inputs. Paste in your household's 15 favourite dinners once, save the URL, spin every night the decision feels heavy.

The discipline that makes this work: no re-rolls. If the picker says "spaghetti carbonara" and you've "got the feeling for tacos," you've just reintroduced choosing. The whole point is the pre-commitment.

Recipe boxes (Gousto, HelloFresh) solve the dinner decision in a different way: they decide for you, 3-5 days a week, and post pre-portioned ingredients with the recipe card. The decision space collapses to "open box, follow card."

The trade-off is cost: £5-£7 per portion vs £3-£4 for supermarket cooking. The premium pays for: the planning (no shopping list), pre-portioning (no half-used ingredients in the fridge), and recipe variety (you cook things you wouldn't have chosen). Whether it's worth it depends on what your decision-time is worth — for households where 6pm is consistently a friction point, £15-£25 extra a week is often the best money spent on quality-of-life improvement.

Most boxes do heavy first-month discounts (£20-£40 off your first 4 boxes). Worth trying for a month even if you're uncertain. See our easy-dinner guide for the recipe-box analysis in more depth.

If you're sticking with the "we'll figure it out tonight" approach, the only structural lever is the option space. Two practical moves:

• Stock for 5 specific meals. Buy ingredients with intent. Most households who say "we have nothing in the fridge" actually have ingredients for any number of dinners — they just don't have ingredients for a specific dinner. Plan 5 specific dinners per shop and the "what shall we eat?" question has a defined answer set.
• Pre-decide "the easy 3." Three dinners you can make in <20 minutes from whatever's in the fridge: pasta + tinned tomatoes; eggs + toast; stir-fry. If none of the planned dinners appeal, fall back to one of these. The fallback removes the "we have to decide from scratch" stress.

Apps that suggest dinners (Mealime, Whisk, Yummly) help, but only if you constrain them heavily. Default settings give you 100,000+ recipe options. Strip down to:

• Dietary requirements (vegetarian, gluten-free, etc.).
• Time budget (under 30 minutes).
• Skill level (easy / no-special-equipment).
• Ingredient blacklist (no cilantro for the divisive household).

The narrower the filter, the more useful the output. Wide-open meal-planning apps are decision-replacement; narrow filters are decision-acceleration. Different products. Both legitimate. Pick which problem you have.

The phrase that kills more households' dinners than any other. The pathology:

1. Person A asks: "What shall we eat?"
2. Person B defers: "I don't mind, what do you want?"
3. Person A reciprocates: "Anything's fine, you choose."
4. 20 minutes later: still no dinner. Both are now hungrier and more irritable. Quality of the eventual decision drops.

The fix is decision-authority allocation. Pick one of these protocols and stick to it:

• Alternating choice: Mondays / Wednesdays / Fridays you decide; Tuesdays / Thursdays / Saturdays your partner does. No conferring, single-stakeholder decision per night.
• Veto, don't propose: One person proposes. The other can veto with a counter-proposal. No "I don't know" replies allowed.
• Constraint-from-other: One says the cuisine ("something Italian"). The other picks the specific dish.

Honest position: takeaway 1-2 nights a week is fine for most households. The trap isn't takeaway itself, it's the slide from "occasional treat" to "always". Set a max — 2 nights a week, say — and the maths and the health both work.

The right nights to take takeaway:

• The week's hardest day. Spend the £20 to reclaim the hour.
• When you've worked out hard or you're ill. Energy to cook isn't there.
• When you genuinely want a specific cuisine that's hard to cook (pho, Korean fried chicken, dim sum). Don't fight it.
• Friday — symbolic week-end marker. Cultural rhythm.

Just Eat for breadth of local options (often the only option in smaller towns); Deliveroo for restaurant-grade kitchens. Both have monthly subscriptions worth it if you order 4+ times a month.

Putting it all together, the working toolkit:

1. One pre-decided weekly schedule. Theme each day. Reduces decisions to "what version of pasta" instead of "what shall we eat?"
2. One random picker for the "I don't know" nights. Bookmark What Should We Eat? with your custom list of 15 favourites.
3. One recipe box subscription, on/off as needed. Subscribe for a month when life gets heavy, pause when you've got time and headspace.
4. One takeaway app for the days that warrant it. Just Eat or Deliveroo. Set personal max of 1-2 nights a week.
5. One "fallback 3" — meals you can always make. Eggs, pasta with tomatoes, stir-fry. When all else fails, default to one of these.

Five elements. Set up over a weekend. Pays back compound interest in reclaimed evening hours.