Industries / Retail, hospitality and personal services / Dry cleaning and laundromats
Pickup and delivery
Pickup and delivery is the work of collecting customers' dirty clothes and bringing them back clean, by van, courier, locker or drop box.
5
Milestones achieved, from 2004 to 2024.
2
To watch for: announced, not yet confirmed outside.
2
Open deliverables. Nobody has shipped them.
The road
AchievedWatch forOpen
Achieved
2004
Japan registers pickup-only firms
An amendment to Japan's Cleaning Business Act (Act No. 33 of 2004) requires anyone who runs a business collecting and returning laundry without opening a cleaning shop to notify the prefectural governor in advance of its methods and number of workers. Firms already working that way were given three months to file. The act also asks operators to explain how they will treat the laundry and requires them to tell customers where to send complaints. Japan's health ministry now lists such shopless services among the rivals of neighbourhood cleaners, and advises cleaners to add services such as storage and 24-hour drop-off lockers. The shops themselves are thinning out: the ministry counted 159,801 licensed cleaning premises in fiscal 2001, 113,953 of them counters that only take in and hand back clothes, and 64,305 at the end of March 2026, 2,691 fewer than a year earlier.
Credit
the Japanese Diet; the Ministry of Health, Labour and Welfare
2014
Lockers open all night
Tide Dry Cleaners, franchised by Agile Pursuits Franchising, a subsidiary of Procter & Gamble, opened its 20th US store in Charlotte, North Carolina, in March 2014. Its stores offered drive-through service and 24-hour drop-off and pickup through drop boxes, lockers and automated kiosks, so customers did not have to reach a counter during opening hours.
Credit
Procter & Gamble; Agile Pursuits Franchising
2015
China's laundry by the bag
Edaixi ("e-washing"), which grew out of the laundry franchise Rongchang, collected laundry booked through its app or WeChat at a set time for a flat 99 yuan (about $16) a bag and returned it within 72 hours. In August 2015 it raised $100 million in a round led by Baidu, saying it had more than 5 million users in 16 cities and over 100,000 orders a day; TechNode reported these as the company's figures. Part of the money was to pay for discounts to customers. By November 2016, amid a funding shortage for Chinese start-ups and rumours of layoffs, it had moved to a partner model that needed fewer assets of its own and narrowed its focus from household services in general to laundry, chief executive Lu Wenyong told TechNode.
LessonGrowth bought with customer discounts lasted only as long as investors paid for it.
Credit
Edaixi, chaired by Zhang Rongyao; Lu Wenyong
2016
One-hour pickup fails
Washio, a start-up launched in 2013 in Santa Monica, California, sent drivers to collect laundry and dry cleaning ordered by app, and in 2015 began promising pickup within an hour and clothes back within 24 hours. It served customers in Boston, Los Angeles, Chicago, the San Francisco Bay Area, New York and Washington, and raised $16.82 million by CrunchBase's count, reported by TechCrunch, before it stopped taking orders on 29 August 2016. Another 2013 entrant, Prim, had closed in January 2014. In October 2016 Rinse, a rival whose drivers collected only between 8 and 10 in the evening, bought Washio's customer lists; its chief executive, Ajay Prakash, said the on-demand model "isn't the most efficient or economical way" to clean clothes.
LessonLaundry apps sold convenience at a premium for a service people buy far less often than rides, Quartz noted, and scheduled windows that let one driver collect many orders in a trip outlasted pickup on demand.
Credit
Washio, led by Jordan Metzner; Rinse
2024
Laundry done in homes
Poplin, a US app formerly called SudShare, passes orders to independent contractors who collect the laundry, wash and dry it in their own machines and bring it back folded, for $1 a pound in most cities. Reviewed, a product-testing site, tested the service in March 2024 and reported that its basic cover pays up to $35 a garment and $200 an order for loss or damage. In September 2026 Axios reported the company's figures: more than 500 US cities, and workers who keep 75% of the price plus tips and pay for their own utilities, transport and machine wear.
Credit
Poplin
Watch for
One app, sixteen countries
Laundryheap, a London company founded by Deyan Dimitrov that collects clothes and returns them within 24 hours, says it now works in 28 cities across 16 countries after launches in Bogotá, Mexico City, Kuala Lumpur and Edinburgh. It says it has served more than 400,000 customers, processed over 110 million items, grown 700% since 2020 and made seven acquisitions in three years, among them France's Lavoir Moderne and Singapore's Oppa Laundry.
Achieved whenits audited accounts, which a UK company files at Companies House, confirm the volumes and show whether the business makes a profit.
Claimed by
Laundryheap, February 2026
Cleaners with no staff
Tide Services, which franchises Tide Cleaners and Tide Laundromat, says it is rolling out automated dry-cleaning stores that customers can use 24 hours a day with no staff on site, including a new format in Minneapolis, and that its garment counts are rising while the industry's are falling.
Achieved whenan outside party publishes the staffless stores' rates of lost and damaged garments and complaints, and garment counts that compare Tide's stores with the industry's.
Claimed by
Andrew Gibson, chief executive of Tide Services, for the third quarter of 2026
Open
01
Identical loads, many orders
A published tally of how pickup laundry services treat identical loads. Make 10 matching loads, each with marked test items: a white shirt with a pen mark, a pair of socks, a wool sweater labelled for a cold wash, a dark T-shirt. Send them through repeated orders with three kinds of service, in at least two cities: app services that use cleaning plants, home-based gig services, and laundromat wash-and-fold with delivery. Record lost items, damage, shrinkage, stains left, and promised against actual return times.
The briefCredit · unclaimed02
Pickup cost per order
An open model that plans routes for a laundry pickup service on a real street map and reports the cost and vehicle kilometres per order for three designs, pickup within an hour, a fixed evening window and building lockers emptied once a day, at several order densities. Each order takes two trips, out dirty and back clean, which parcel routing studies do not model.
The briefCredit · unclaimed
Open 01 · the brief
Identical loads, many orders
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A published tally of how pickup laundry services treat identical loads. Make 10 matching loads, each with marked test items: a white shirt with a pen mark, a pair of socks, a wool sweater labelled for a cold wash, a dark T-shirt. Send them through repeated orders with three kinds of service, in at least two cities: app services that use cleaning plants, home-based gig services, and laundromat wash-and-fold with delivery. Record lost items, damage, shrinkage, stains left, and promised against actual return times.
Why it is still open
The published tests found are single orders. Time Out New York ranked five services in November 2015 on one test each, by cleanliness, speed and price, and Reviewed tested one Poplin order in March 2024. A search found no multi-order tally of lost or damaged items, and the services' cover terms, such as Poplin's $35 a garment, give no loss rates.
Built with what exists
a kitchen scale and tape measure for weight and shrinkage; a phone on a fixed stand with a grey card for before-and-after photos; an open image model, fine-tuned on one graphics chip, that flags colour change and stains (see the Note "One GPU to fine-tune"); a shared spreadsheet.
First step
Make three identical loads and send one through each kind of service in one city.
Done looks like
At least 90 orders across the three kinds of service and two cities, with lost items, damaged items and late returns per 100 orders for each service, and all photos published.
Open 02 · the brief
Pickup cost per order
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
An open model that plans routes for a laundry pickup service on a real street map and reports the cost and vehicle kilometres per order for three designs, pickup within an hour, a fixed evening window and building lockers emptied once a day, at several order densities. Each order takes two trips, out dirty and back clean, which parcel routing studies do not model.
Why it is still open
The 2013 to 2016 failures turned on these costs, but the figures stay private. Rinse's chief executive said Washio's one-hour model was not economical without giving numbers. A search for a laundry pickup routing model found none; the nearest work is general, such as a 2024 review of route planning for same-day delivery, and a GitHub search for "laundry routing" found nothing related.
Built with what exists
OpenStreetMap street data through OSMnx; the OSRM router; Google OR-Tools, which solves routing with pickups, deliveries and time windows; public wage and vehicle cost figures.
First step
Model an 8-to-10 pm evening window against one-hour pickup for 200 made-up orders a day in one city.
Done looks like
Published code and tables for three cities at five order densities, plus one operator's anonymised route log for a week run through the model, with the model's vehicle kilometres within 15% of the log's.