Industries / Consumer goods / Textiles and apparel / Footwear
Soles
Sole making is the choice of materials for the parts under the foot and the moulding, foaming or printing that shapes them: the outsole that meets the ground, the midsole foam that cushions each step and gives some energy back, and any stiff plate inside.
7
Milestones achieved, from 1844 to 2020.
3
To watch for: announced, not yet confirmed outside.
3
Open deliverables. Nobody has shipped them.
The road
AchievedWatch forOpen
Achieved
1844
Rubber that holds its shape
Before 1844, rubber went stiff and cracked in the cold, was too soft at room temperature to keep its shape, and stuck to everything it touched. Charles Goodyear's US patent 3,633, granted on 15 June 1844, covers mixing sulphur and white lead into rubber and heating it at a controlled temperature. The process, later called vulcanisation, left rubber less sticky and steadier as temperatures changed. In Connecticut, Goodyear's home state, rubber firms using the process multiplied. The United States Rubber Company merged nine of the leading makers in Naugatuck in 1892, and in 1916 it created Keds, its brand of canvas shoe with a rubber sole, when it brought together most rubber-soled shoe making.
Credit
Charles Goodyear; the United States Rubber Company
1937
A lug sole for climbing
In 1935 a party of climbers on Punta Rasica in the Val Bregaglia was caught by a sudden blizzard, and frostbite hit six of them, by the Fondazione Pirelli's account. Vitale Bramani, a mountaineer, then set out to make one boot that would work on both rock and snow. With Franco Brambilla of Pirelli he worked on a new vulcanised rubber compound and a tread based on Pirelli's tread patterns. In 1937 the sole came out of Pirelli's laboratory, was patented, and became known as the carroarmato, the "tank" tread, for its deep lugs (raised blocks of rubber). The Italian team that climbed K2 in 1954 wore boots with Vibram Carrarmato soles, by Vibram's own history.
Credit
Vitale Bramani; Pirelli (Franco Brambilla)
1974
The waffle sole
In 1971 Bill Bowerman, the University of Oregon track coach and a Nike co-founder, poured liquid urethane into his family's waffle iron while trying to make a better sole. That November, Oregon's cross-country runners won the national (NCAA) title in early versions, by Nike's account. His patent, filed on 30 August 1972 and granted as US 3,793,750 on 26 February 1974 to BRS Inc. (Blue Ribbon Sports, later Nike), claims a sole with "short multi-sided polygon shaped studs" whose flat sides give grip. It describes the shoe for football on artificial turf. Nike's history says waffle-soled trainers followed, with orders above 100,000 units by 1975.
LessonA sole patented for artificial turf found its market under runners.
Credit
Bill Bowerman; BRS Inc
2013
Foam made of beads
EVA (ethylene-vinyl acetate) foam had been the most widely used midsole material for four decades, by a 2024 paper in Footwear Science. BASF foamed granules of its thermoplastic polyurethane (TPU), a rubbery plastic, into beads ten times their starting volume, each holding tiny closed gas bubbles. It announced the material in July 2013 as Infinergy, "the world's first expanded thermoplastic polyurethane". adidas uses about 2,500 of the beads in each midsole and sold the foam as Boost in its Energy Boost shoe from 27 February 2013. adidas said tests by its own Innovation Team showed "the highest energy return in the running industry". The energy and cold-weather figures given at launch were adidas's and BASF's own.
Credit
BASF; adidas
2017
Foam and a carbon plate
Nike's prototype racing shoe paired ZoomX, a foam made with PEBA (polyether block amide), with a full-length carbon-fibre plate curved at the forefoot and embedded in the midsole, on a 31 mm heel. Wouter Hoogkamer, Rodger Kram and colleagues at the University of Colorado Boulder tested it on 18 runners (Sports Medicine, online November 2017) and found it cut the energy cost of running by 4% on average against the Nike Zoom Streak 6 and the adidas adizero Adios BOOST 2. That study was paid for by a contract from Nike; two authors were Nike employees, and Kram was a paid consultant to Nike. Kyle Barnes of Grand Valley State University and Andrew Kilding then tested the shoe sold as the Nike Vaporfly 4% on 24 runners, in a study funded by the university (December 2018), and found running economy, the energy a runner needs to hold a pace, 4.2% better than in an adidas marathon shoe and 2.6% better than in track spikes.
LessonThe gain wears off: in a 2023 study funded by Spain's science ministry, 22 trained runners tested a PEBA shoe and an EVA shoe new and after 450 km of wear, and after 450 km the two had similar running economy (Rodrigo-Carranza, Hoogkamer and colleagues).
Credit
Nike; Hoogkamer, Kram and colleagues; Barnes and Kilding
2017
Printed lattice midsoles
Carbon, a Silicon Valley company, makes parts by Digital Light Synthesis: light projected from below a vat turns liquid resin into a solid as the part is drawn upward. In April 2017 adidas showed Futurecraft 4D, a shoe whose lattice midsole is made this way. It planned to sell 5,000 pairs that autumn and to reach more than 100,000 pairs by the end of 2018, and its 2019 annual report says it produced more than 100,000 pairs in 2018, its own figure. Carbon says adidas tried more than 50 lattices before settling on one, and that a midsole whose properties vary across the part cannot be injection- or compression-moulded in one piece.
Credit
Carbon; adidas
2020
Limits on sole thickness
On 31 January 2020 World Athletics, the sport's governing body, set new shoe rules for elite competition: the sole may be no thicker than 40 mm (30 mm in a spiked shoe with an extra plate), the shoe may hold no more than one rigid embedded plate or blade, and from 30 April 2020 a shoe must have been on open sale for four months before it can be used in competition. It cited "independent research" indicating that the new soles may give a performance advantage. In December 2021 it set a single limit of 20 mm for track and field shoes from 1 November 2024, giving makers time because they had invested in spikes with soles of 20 to 25 mm, and it left energy return uncapped.
LessonThe rules capped thickness and plates, not how much energy a foam gives back, and shoes kept improving inside them: in April 2026 Sabastian Sawe won the London Marathon in 1:59:30, which Dezeen calls "the first-ever official sub-two-hour time", in a 97 g adidas shoe with a 39 mm heel.
Credit
World Athletics
Watch for
A 97-gram marathon shoe
adidas says its Adizero Adios Pro Evo 3 weighs 97 g, has a 39 mm heel of a new foam, Lightstrike Pro Evo, which it calls 50% lighter than its earlier versions, and improves running economy by about 1.6%. The testers it names are its own labs in Herzogenaurach and athletes' camps in Kenya and Ethiopia, and a search found no test by an outside lab.
Achieved whena lab with no stake in adidas publishes a running-economy test of the Evo 3 against its predecessor or another racing shoe.
Claimed by
adidas, April 2026
Outsole wear that biodegrades
The materials company KUORI says its BIOWA 80A, a thermoplastic elastomer (a rubbery plastic that can be melted and moulded) filled with olive pits and walnut shells and made for outsoles, is designed so that the particles worn off it can biologically degrade. It says it has moved to prototypes and pilot series of outsoles. The interview names no test standard, time frame or tester.
Achieved whena lab with no stake in KUORI collects particles worn from soles of the material and reports, by a named standard test, how much of them breaks down, how fast and in what conditions.
Claimed by
KUORI, in an interview with Fashion for Good, February 2026
Bio-based soles benchmarked
In September 2025 Fashion for Good, a group that tests new materials with fashion brands, started The Next Stride with adidas, Target and Zalando: a 12-month project to "test and validate bio-based polymers" from Algenesis Labs, Balena, Evoco, KUORI and Yulex as sole materials, against fossil-based EVA, PU, TPU, rubber and TPR, with prototypes tested with SATRA, a footwear testing organisation. Fashion for Good says soles make up at least 40% of a shoe's mass. No results had been found published by November 2026.
Achieved whenthe benchmark results are published with SATRA's test data, and a partner brand sells a shoe with one of the soles.
Claimed by
Fashion for Good, September 2025
Open
01
Sole wear per kilometre
A public table of how many milligrams of outsole at least 30 shoe models (running, walking and work shoes) lose per kilometre in real use. Volunteers wear three pairs of each model for 300 km of logged walking or running, and each shoe is cleaned, dried and weighed every 50 km. The same outsole rubber also goes through the ISO 4649 rotating-drum abrasion test, to see whether the lab test predicts wear on the road.
The briefCredit · unclaimed02
Whole-shoe foam fatigue
A four-station machine that presses a foot-shaped form heel to toe into whole racing shoes at running force and stride rate, recording each shoe's stiffness and the share of energy it gives back every 10,000 steps, out to 200,000 steps. Test ten current racing shoes with PEBA, TPU and EVA-blend foams, two pairs each, and check three of the models against pairs worn 450 km by runners.
The briefCredit · unclaimed03
Bio-based soles, measured
A public register of at least 30 shoes sold with plant-based sole claims (sugarcane EVA, algae foam, corn- or castor-based materials). Cut the outsole and midsole of each apart, have two labs test each part for bio-based carbon by radiocarbon dating, and set the results beside the brand's claim.
The briefCredit · unclaimed
Open 01 · the brief
Sole wear per kilometre
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A public table of how many milligrams of outsole at least 30 shoe models (running, walking and work shoes) lose per kilometre in real use. Volunteers wear three pairs of each model for 300 km of logged walking or running, and each shoe is cleaned, dried and weighed every 50 km. The same outsole rubber also goes through the ISO 4649 rotating-drum abrasion test, to see whether the lab test predicts wear on the road.
Why it is still open
RIVM, the Dutch public health and environment institute, put microplastics from footwear in the Netherlands at 140 tonnes for 2022, but built its sole figure on assumptions "as no measured abrasion rates were found in literature", and asked for experiments. Fraunhofer UMSICHT estimated about 100 g of sole wear per pedestrian per year. A 2020 University of California, Santa Barbara master's project, with adidas as client, ran what it calls the first real-life wear test and estimated 1.4 to 40.0 g per person per year. A November 2025 study at International Christian University in Tokyo weighed 23 pairs from 12 makers, named only by letter, and found about 35 mg/km for a pair of running shoes. A search for wear rates by named model found none.
Built with what exists
A balance reading to 0.01 g; a drying cabinet; GPS-watch route logs exported from Strava or Garmin; phone photos or a low-cost 3D scanner for tread depth; a rubber lab's ISO 4649 abrader; a spreadsheet.
First step
Buy three pairs each of five best-selling running shoes, weigh them, and sign up 15 runners from one club for their first 50 km.
Done looks like
30 named models and 90 pairs, each with milligrams lost per kilometre and a range, the ISO 4649 result beside each, and every weighing and route log published under an open licence.
Open 02 · the brief
Whole-shoe foam fatigue
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A four-station machine that presses a foot-shaped form heel to toe into whole racing shoes at running force and stride rate, recording each shoe's stiffness and the share of energy it gives back every 10,000 steps, out to 200,000 steps. Test ten current racing shoes with PEBA, TPU and EVA-blend foams, two pairs each, and check three of the models against pairs worn 450 km by runners.
Why it is still open
Aimar and colleagues at Université Grenoble Alpes, funded by Decathlon and France's ANRT research agency, ran 200,000 compression cycles, which they reckon at 400 to 700 km of running, on 2 mm slices cut from the midsoles of four commercial shoes, not on whole shoes, and called for "more biofidelic fatigue tests", meaning tests closer to how a foot loads a shoe. Rodrigo-Carranza and colleagues measured runners in one PEBA and one EVA shoe, new and after 450 km. A search for whole-shoe energy return after controlled mileage, across brands, found nothing.
Built with what exists
Four linear actuators with load cells and displacement sensors; a foot-shaped loading form; a Renesas RZ/T2N chip, which Renesas says can control up to four motors from one chip (see the Note "Four axes, one chip"); a university materials tester to calibrate; Python.
First step
Build one station and run 20,000 cycles on a new pair and a well-worn pair of the same racing shoe.
Done looks like
Ten models, two pairs each, with stiffness and energy-return curves from 0 to 200,000 cycles, three models checked against worn pairs, and the raw data and rig drawings published.
Open 03 · the brief
Bio-based soles, measured
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A public register of at least 30 shoes sold with plant-based sole claims (sugarcane EVA, algae foam, corn- or castor-based materials). Cut the outsole and midsole of each apart, have two labs test each part for bio-based carbon by radiocarbon dating, and set the results beside the brand's claim.
Why it is still open
When Allbirds launched its sugarcane foam SweetFoam in 2018, a co-founder said "minor components" were still made from petroleum. Labs sell the test to brands: Beta Analytic validated Reebok's Cotton + Corn materials as 75% bio-based for the US Department of Agriculture's BioPreferred programme, and CTC, a French technical centre, tests footwear materials to EN 16640. A search found no independent survey of soles on sale tested against their claims.
Built with what exists
Commercial radiocarbon labs such as Beta Analytic and CTC; a band saw; a web agent such as Browser Use 1.0 to collect each brand's claim from its product pages (see the Note "Deleting 80% to start over"); a spreadsheet.
First step
List 50 shoes sold with bio-based sole claims and get quotes from two labs to test five of them.
Done looks like
30 shoes from at least ten brands, outsole and midsole tested separately at two labs, and every lab report published beside the claim, with the gap between them.