I own eight dresses. The average price is $79. The most expensive is $145. None of them cost more than $150, and three of them are, by my standard, the best dresses I have ever owned.

This is not how I expected my thirties to go. I grew up believing, by my mother's standard and by every fashion magazine of the 1990s, that a real dress cost at least $300. A "good" dress cost $500. A dress you could wear to a wedding cost $1,000. The dresses I actually wear — to dinner, to the park, to my partner's office holiday party — cost less than a hundred dollars each. The math, by every measure, surprised me.

Below is the list, in order of how often I wear each one. I have changed my mind about several things I used to believe about dress shopping. I have opinions on what to skip. There is a brief history of how Aritzia, Realisation Par, and Toast got from small Vancouver and Sydney shops to the international dress market. And there is, at the end, an honest summary — because anyone who tells you a good dress has to cost $300 has not, in fact, looked at the rack at H&M recently.

The list

1. Linen-blend midi, Uniqlo, $49 (2023). The single most-worn dress I own. I have worn it, by my estimate, 90 times since I bought it. It has been washed, on average, once every two weeks. The linen has softened. The seams are, by my partner's description, "still perfect." The fit is, by Uniqlo's standard, slightly loose. I sized down. This is, by my arithmetic, the single best value in my closet. Cost per wear: $0.54.

2. Wrap midi, Mango, $79 (2024). Bought on a clearance rack in February 2024. The fabric is, by every measurement, indistinguishable from dresses I have tried at three times the price. The wrap ties are, by my standard, the only flattering waist-tie I have ever owned. I wear it, on average, twice a month. It has been washed twelve times. The color has not faded. Cost per wear, so far: $3.30.

3. Crewneck cotton, Everlane, $98 (2023). Bought at full price, which I almost never do. The cotton is heavy — 280 grams per square meter, which is, by Everlane's standard, "the heaviest we make." The cut is a relaxed A-line. I wear it, on average, once a week in the warmer months. It has held up, by every measurement, better than the $300 cotton dress I owned before this one. Cost per wear, over three summers: $0.94.

4. Slip dress, Aritzia, $98 (2023). Bought at Aritzia's flagship in SoHo. Aritzia was, in 2026, founded in Vancouver in 1984 by Brian Hill — the brand is, by my standard, the most underrated dress maker in North America. The slip is silk-blend. The straps do not fall down. The hem is, by my partner's description, "exactly mid-calf." I wear it, on average, twice a month, year-round, with a turtleneck underneath in winter. Cost per wear, so far: $1.81.

5. Cotton midi, H&M, $39 (2024). Bought online, returned to the store, bought again in a different color. The cotton is, by every measurement, the best cotton H&M has ever sold. I am, by 2026 standards, suspicious of H&M's sustainability claims. I am also, by my standard, willing to overlook those claims when the dress fits this well. I wear it, on average, three times a month in summer. Cost per wear, so far: $1.30.

6. Shirt dress, Target (A New Day line), $35 (2025). The cheapest dress on this list. I bought it on a Sunday at the Target on Flatbush Avenue, after a doctor's appointment. I have worn it, by my estimate, 25 times since. It has been washed nine times. The buttons have not fallen off. The collar has not warped. Cost per wear: $1.40. Target's A New Day line is, by my standard, the single most underrated dress brand in 2026.

7. Long-sleeve dress, COS, $89 on sale (2024, originally $129). The COS discount rack is, by my standard, the single best place to buy a dress. I bought this at 31% off, in October, at the COS store in SoHo. The fabric is a heavyweight jersey that I cannot, by any measurement, replicate at this price elsewhere. The sleeves are slightly long. I rolled them. The dress is, by my partner's description, "the one I keep asking you to wear." Cost per wear: $4.45.

8. Silk midi, Realisation Par, $145 (2024). The only dress on this list over $100. I bought it at the Realisation Par sample sale in Brooklyn, marked down from $185. Realisation Par was founded in 2012 in Sydney, Australia, by Alexandra Spencer and Teale Talbot. The brand is, by every measure, the Australian Ganni — printed silk dresses that photograph well and cost, in their own country, about a third of what they cost in the United States. The silk is real. The print is, by my standard, the only print I have ever owned that I did not get sick of. Cost per wear, so far: $7.25.

Total spend: $632. Total cost per wear, weighted across all eight dresses: $1.40. Average price per dress: $79.

What I have changed my mind about

A few beliefs I held in 2008, when I bought my first "real" dress, that the last eighteen years have killed.

I used to think expensive dresses lasted longer. The $300 Vince dress I bought in 2014 lasted four years. The $49 Uniqlo dress I bought in 2023 has lasted three years and is, by my standard, in better condition. The fabrics at the price point I am now buying — Uniqlo's linen, Everlane's heavyweight cotton, Realisation Par's silk — are, by every measurement, equivalent to the fabrics at three times the price. The construction is, in most cases, equivalent. The $300 difference is, by my standard, paying for the brand name and the marketing, not the dress.

I used to think dresses were for special occasions. I owned, between 2008 and 2023, approximately thirty dresses. About twenty of them were, by my estimate, for "special occasions" — weddings, holidays, the kind of events where you wear a dress once and then feel, by my own description, "guilty about the cost every time I look at it in the closet." The eight dresses I own now are, by every measure, for daily life. The dinner dress. The park dress. The doctor's appointment dress. The wedding dress, when there is a wedding, is rented from Rent the Runway ($30-$80 per rental, dry cleaning included). The closet is, by my standard, no longer a museum of missed occasions.

I used to think trends required investment. The "expensive" trend pieces I bought — the 2017 Off-White industrial belt ($450), the 2019 Ganni tiger-print dress ($390), the 2021 The Row bare sandals ($890) — were all, by every measure, dated within three years. The "cheap" trend pieces I bought at H&M and Mango were also dated within three years. The math, by every measure, is identical. But the loss on the cheap pieces was, by my standard, easier to absorb. A $39 dress I wore 30 times cost me $1.30 per wear. A $390 dress I wore 30 times cost me $13 per wear. The trend cycle does not, in either case, care about my cost per wear.

I used to think "investment dressing" meant buying less. The capsule wardrobe movement, which I followed closely between 2018 and 2022, told me to buy fewer, more expensive pieces. I bought, in 2020, a $450 Toteme slip dress. I wore it four times. The capsule principle, by my standard, was correct — I should buy fewer dresses. The implementation was wrong — buying fewer, more expensive dresses did not, by any measure, increase the cost per wear of my wardrobe. It increased it, because the expensive dresses were, by my standard, too precious to wear often.

I used to think I needed variety in silhouette. I owned, between 2008 and 2023, bodycon dresses, A-line dresses, shirt dresses, slip dresses, wrap dresses, maxi dresses, mini dresses, and a $400 structured Oscar de la Renta dress I wore once. The eight dresses I own now are, by my arithmetic, in three silhouettes: midi, slip, and shirt dress. I wear all three silhouettes on rotation. The variety is, by my partner's description, "more than enough." The Oscar de la Renta dress is, in 2026, in a bag in the back of my closet. I have not opened the bag in three years.

What to skip when buying dresses

If I were advising a friend who wanted to build a small, well-priced dress collection in 2026, I would tell them to skip these categories at the start.

The "special occasion" dress. The sequined floor-length gown. The beaded cocktail dress. The "I will wear this to my cousin's wedding in October" dress. These dresses are, by every measure, bought with a specific date in mind, worn once, and then kept for ten years in a garment bag. If you need a special occasion dress, rent one. The rental market — Rent the Runway, Nuuly, ByRotation — is, by 2026 standards, the only honest way to buy a special occasion dress.

Anything over $150 you have not tried on. I have, in my twenties, bought dresses online for $200 to $600 without trying them on. About 40% were returned. About 30% were kept and never worn. About 30% were kept and worn once. The cost of the mistakes, by my standard, was enormous. Try the dress on. If you cannot try it on, order from a place with free returns. Skip the gamble.

Anything in a "limited edition" or "final sale" tag. The urgency marketing on dresses is, by every measure, real. The "only 3 left!" banner. The "final sale — no returns!" tag. These are, by every measure, designed to make you skip the reflection phase. Buy the dress when you can return it. Skip the urgency.

Anything dry-clean only. I do not dry-clean. The eight dresses on this list are, every one of them, machine-washable. A dress that requires dry cleaning is, by my standard, a dress I will wear half as often. The dry-cleaning cost ($15-$25 per dress) is, by my arithmetic, an additional $60-$100 per year I would rather not spend.

Anything in a print that requires "building an outfit around." I owned, between 2015 and 2020, three printed dresses that required specific shoes, specific bags, and specific earrings to "work." I wore each of them, on average, four times. The eight dresses I own now go with, by my standard, every shoe and bag I own. A dress that limits your other choices is, by every measure, a dress that gets worn less.

Anything over $200 from a brand you have not bought from before. The dress market is, by every measure, designed to encourage experimentation. New brands launch every month. The "buy once, cry once" mentality, when applied to a brand you have never tried, is, by my standard, a recipe for a $400 mistake. Buy from brands you have tried. Skip the discovery purchase.

Anything described as "investment" or "heirloom." A $400 dress is, by every measure, not an investment. An investment, in 2026, is something that appreciates in value or generates income. A dress does neither. The "investment" language is, by every measure, marketing. The "heirloom" language is, by my standard, an even more cynical version of the same pitch — the suggestion that your daughter will, in 30 years, want your Toteme trench. She will not. She will want her own.

A brief history

The cheap-but-good dress market is, by every measure, a 2010s invention. Until about 2010, the dress market was, in the United States, polarized: cheap fast fashion on one end (H&M, Forever 21, Charlotte Russe) and expensive designer on the other (Marc Jacobs, Tory Burch, Alice + Olivia). The middle — the $40-$150 dress market — was, by every measure, dominated by brands like Ann Taylor, J.Crew, and Banana Republic, none of which made, by my standard, very good dresses.

The 2010s changed this in three ways. The first was the rise of direct-to-consumer brands. Everlane, founded in 2010 in San Francisco, was, by every measure, the first DTC brand to make a $98 cotton dress that was, on a fabric basis, equivalent to a $300 designer dress. The transparency about cost — Everlane published, in 2014, a cost breakdown of a $98 cotton t-shirt showing $7 of materials and $91 of brand markup — was, by every measure, a turning point in how consumers thought about price.

The second was the international expansion of mid-priced brands. Aritzia, founded in 1984 in Vancouver by Brian Hill, expanded to the United States beginning in 2007 and by 2026 operated approximately 130 US stores. The Aritzia price point — $80 to $200 for dresses — was, by my standard, the sweet spot for the modern dress market. Mango, founded in 1984 in Barcelona, opened its first US store in 2006 and by 2026 operated approximately 140. COS, founded in 2007 as a H&M sister brand, expanded to the United States in 2011 and by 2026 operated approximately 60 US stores. Each of these brands brought European quality at non-European prices.

The third was the rise of Instagram-era brands. Realisation Par, founded in 2012 in Sydney by Alexandra Spencer and Teale Talbot, built its initial following on Instagram in 2014-2016, when its printed silk dresses were worn by every fashion influencer. The brand opened its first US store in Los Angeles in 2018 and expanded to Brooklyn in 2023. The Realisation Par sample sale is, by my standard, the only sample sale worth attending.

The 2020s cheap-but-good market is, by every measure, still evolving. The closure of Lord & Taylor (2021), the bankruptcy of J.Crew (2020, emerged 2021), and the slow decline of department store dress departments have pushed consumers toward both fast fashion (H&M, Target) and direct DTC brands. The 2026 dress market is, by every measurement, healthier for the consumer than the 2008 dress market. The dress is, by my standard, the single most democratized category of clothing.

The honest summary

A dress is, by every measure, a piece of clothing you wear because you do not want to wear trousers. The $79 average price I pay for a dress is, by my standard, the right price. Below $40, the fabrics are too cheap. Above $150, the cost-per-wear math stops working unless you wear the dress twice a week.

What works for me: machine-washable fabrics (cotton, linen, silk-blend), three silhouettes (midi, slip, shirt dress), neutral colors with one print, and buying on sale whenever possible. The eight-dress collection takes, by my estimate, about four cubic feet of closet space. The eight-dress collection goes, by my partner's description, with "every shoe you own." The eight-dress collection is, by every measurement, the only dress collection I have ever maintained for more than two years.

What does not work for me, and may not work for you: the idea that a dress has to be expensive to be good. The "investment dress" pitch is, by every measure, marketing. The "special occasion" pitch is, by my standard, a way to sell you a dress you will wear once. The "timeless" pitch is, by every fashion editor I have read, the most misleading pitch of all — no dress is timeless, because the body wearing the dress is not timeless, and the dress only looks right on the body that is currently wearing it.

Eighteen years and approximately forty dresses in, my dress collection is, by every measurement, the smallest and least expensive it has ever been. The dresses I own now are, by every measurement, the most worn. The math is, by my standard, the only thing that matters.