Why auto recycling is stuck in the Stone Age

Watch the video: https://www.youtube.com/watch?v=SNwmg7Tsj7o

This read is fairly long, but I promise it's worth it if you touch auto recycling in any way.

Think back to the best car you ever broke. You know the one. Maybe it came in for scrap at £350 and made you £3,000. Maybe it was a write-off you picked up at Copart for £1,000 that turned into £5,000.

Why was that car so good?

Walk through it. The parts on it were in huge demand at exactly the right time. They sold for high prices, and there were loads of buyers queuing up for them. Whatever got smashed wasn't where the money was, so the damage didn't get in the way. And the money came in fast, which tells you something else: you probably didn't have any of those parts on the shelf already.

If you're honest, you got lucky.

Now imagine you knew all of that before you bought every car. The demand, the prices, the buyers, the damage and your own stock. Everything that told you, after the fact, that your best car ever was the winner of all winners.

That's what POXA does. In seconds, for as little as 50p, it tells you whether a car is a winner or just like every other car, and checking it against your own stock is coming next. Take the MINI above, front end smashed in. POXA's UK sales data puts the main parts that should sell in the next 6 months at about £2,300, assuming your shelves are empty.15 Whether that's a winner depends on what you pay for it, and now you'd know before you bid.

Most of the trade still finds its winners the Stone Age way: by luck, and after the fact.

Our mission: to bring what drivers, bodyshops and insurers actually need into the moment a yard buys a car, so more of every car gets reused, yards earn more, repairs cost less and CO2 comes down.

How yards work today

Today most yards buy cars on a feeling. Will the parts sell or won't they? Then they dismantle on the same feeling: I think this part sells, that one doesn't. Even the biggest salvage firms say there's no fixed answer. When the competition regulator reviewed a big salvage merger in 2022, the companies involved said a salvage car's value for green parts is "subjective", because it depends on which cars need repairing and what stock is already on the shelf.1 In other words, a car's value turns on demand and stock, and most yards can't check either quickly.

The problem with buying on a feeling is that you can throw away cars that actually carry a ton of value, or miss parts in the dismantling process because who knew that part off that car sold well. A 2026 study of UK accident repair calls this "a structural paradox". The market often has a fair idea which parts will be needed, but cars that could yield reusable parts can still get crushed or exported, "even where downstream demand is evident".2

Why we think this is the biggest problem in the industry

In 2024 alone, 562,185 vehicles were written off in the UK, more than one a minute.3 The same year, UK motor insurers paid a record £7.7bn for vehicle repairs,4 and the average repair on a petrol or diesel car cost 42% more in 2023 than in 2019.5 e2e, a salvage and green parts business, estimates that around £2.4bn of insurers' 2023 repair bill went on replacement parts, and that less than 2% of those parts came from recycled stock.6 A separate survey of UK bodyshops, counting parts fitted, put green parts at just 4.1% in 2025.7

It isn't that nobody wants them. 91% of repairers fitted green parts last year,7 58% of drivers say they'd be likely to use a recycled part,8 and insurers are asking for them. The problem is the one insurers and bodyshops keep giving: the stock isn't there. 42% of repairers say poor availability of green parts is a problem,7 and in 2021 one UK recycler could fill fewer than half of the green part requests from its main insurer.9

That supply problem starts at the yard. If yards knew which cars and which parts the market is crying out for, they would buy those cars and pull those parts.

What solving it would mean

Yards get a much higher yield per car. In a 10-car pilot, a yard we work with that runs Pinnacle yard software made £300 more per car using POXA. Bodyshops have an easier time buying parts, and insurers save a lot of money on repair bills, all while CO2 comes down.

For the insurer, a green part is far cheaper. One of the UK's largest recyclers sells them for up to 75% less than the new list price.10 For the planet, a UK study of a typical four-part repair (front bumper, headlight, bonnet and front door) found that the repair produced 283 to 309 kg CO2e with new parts, depending on where they were made, but only 132 kg with UK green parts. That's a saving of roughly 150 to 177 kg CO2e on one repair, about half.11 A US study found that making one new engine produces around 700 kg of CO2 for a family car, and about 1,000 kg for a pickup truck's V8.12

Now look at Sweden. Swedish insurers sell their write-offs straight to dismantlers they have agreements with, and expect workshops to use those parts.13 Official figures suggest Swedish yards reuse around 175 to 210 kg of parts by weight from every scrapped car, against about 28 kg in the UK's last return (2018).14 That gap is the size of the opportunity for UK yards.

In our opinion, connecting customer demand to the decision to buy a car is the biggest problem to solve in salvage, globally, and we will solve it. This isn't one yard taking business from another. It grows every yard, lowers insurance costs and helps the planet at the same time.

Who is solving what

Most of the auto recycling journey is already being built. Online marketplaces, and platforms like Partly and e2e's PartsMarket, are working on how bodyshops and insurers find and buy parts. Yard management systems like Eladene and Wilma help yards run their stock, and some now offer pricing and profit tools too. But in our view nobody, in the UK or anywhere else, has cracked the supply of cars yet: knowing which cars to buy because their parts are already in demand, fast and cheaply enough to run on every car. Which is ironic, because in this trade you make your money when you buy the car. A car with low demand won't make money, however good your dismantling, your stock management or your sales. We are going to crack it.

The three challenges

So why hasn't anyone solved this before? We think there are three reasons: technology and data, cost, and speed. Each one is hard on its own. Together they're a mountain.

Put simply, for a product like this to work it needs to be very fast (under 5 seconds), very accurate (taking in the damage, the OEM part numbers, the prices and the demand) and so cheap you can run it on every car quote or Copart listing your heart desires.

1. Technology and data

Valuing damaged cars is a hard game with lots of complex, ever-changing targets. Cars that get broken for parts are usually damaged, and the damage is different every time. You need to know which OEM part numbers are on that specific car, and the going rate, not for a new part but for a used one, which changes all the time with supply and demand.

This year three big things came together for POXA. We finally had enough data to match part numbers from what actually came off cars of the same spec, to price them from real UK sales and, in testing, to read damage from photos.

Here's where it gets messy. To value one car properly you need four things, and none of them lives in one place.

Part numbers. First you need to know which OEM part numbers are on that exact car. Manufacturer catalogues are built for ordering new parts, one VIN at a time. They don't tell you which of those parts a breaker can pull and sell. The same model can carry different numbers depending on engine, spec and build date, and numbers get replaced over the years. So you have to learn, car by car, which part numbers actually come off which cars, and there are hundreds of thousands of those numbers.

Prices. Then you need the going rate for each of those parts second-hand. There is no price list for used parts. The real prices are buried in individual sales spread across thousands of sellers, with typos, cut-off titles, wrong part numbers and kits mixed in with single parts. Every sale has to be cleaned up, matched to the right part and the right number, and the one-off prices left out. And because used prices move with supply and demand, you have to keep doing it.

Demand. Finally you need to know how fast each part sells. How often has it sold in the last 12 months? How many are already for sale? Will it go next week or sit on a shelf for a year? Researchers studying UK accident repair found that demand data already exists, in insurer estimating systems, repair quotes and parts sales, but it is rarely joined up in a way that shapes which cars get dismantled.2

Stock. And then there are your own shelves. A car is only worth what its parts are worth to you. If you've already got two of every headlight, door and tailgate it carries sitting in the racking, it's worth a lot less to you than to a yard with none of them. So you need to cross-reference every part number on the car against your own stock, which lives in your yard system or online store rather than in anyone's sales data.

Getting each of these right is hard. The real challenge is connecting them.

That's why doing it by hand fails. On some cars, 100 of a possible 300 breaker parts could sell. To find those 100 you have to search all 300, and that assumes you didn't miss anything. Done properly, that's three to four hours per car. Done quickly, it's 30 minutes to an hour, and you miss at least half. You can build up knowledge over the years, but nobody can carry all of that in their head. POXA can. It's built on more than 1.6 million UK used part sales recorded since January 2025, and every valuation uses the most recent 12 months of sales for that car's parts.

2. Cost

Believe it or not, this was the hardest challenge, and really what took POXA two and a half years to figure out. Knowing what a car is worth in parts only helps if a search costs pence, so you can run it on every car before you buy. It's no good at £5 a search.

There are tools that get part of the way. Some UK yard systems can list the part numbers on a car with prices, and some now promise sales predictions too. In the US there are bidding tools built around a yard's own sales history and stock. What we haven't seen in the UK is one that shows the demand for every part on the car, from sales right across the UK rather than one yard's history, including the parts that will never sell.

And if every search costs pounds rather than pence, good luck running 100 auction cars a week, or plugging it into an online scrap car quoting system doing 1,000 quotes a week. That's exactly where you need it. A search needs to cost pence so you can run it on everything, and I mean everything: every car through your online scrap car portal, every car on a scrap car comparison site, and as many cars as you like at Copart, Raw2K, IAA or Salvage Market.

The idea is to bring the cost down so far that you always know which cars are in high demand, and for which parts. That way, cars don't hit the scrap heap too early while they still carry valuable parts that drivers, bodyshops and insurers need. It also helps you stock a variety of parts rather than the same ones over and over.

3. Speed

It also needs to be fast. Bidding in an auction lasts 30 seconds to a minute, and someone on a scrap car website isn't waiting 30 seconds for a price. So in 2 to 5 seconds you need to find the parts and part numbers on that specific car, value every one and check the demand for each. You need to review the damage or photos of it, take the damaged parts out of the valuation and cross-reference everything with your current stock. Then you need a valuation, a demand score and your stock crossover back, so you know what to pay for that car, if anything. That's a lot to happen in a few seconds, but that's how fast it needs to be.

Where we are today

Hopefully you can now see why nobody has solved this properly, and why it's taken us two and a half years to climb the mountain. We've now launched the core of it. Enter a reg or VIN and POXA finds the parts on that car, prices each one from real UK sales, gives every part a demand score from 1 to 10 and takes off the damage you mark.

Next, POXA will check every valuation against your own stock, from your online store or your yard system such as Pinnacle, and read damage straight from photos. Our API is in early access. It's designed to let you build POXA into your own flows, like quoting tools or scrap car websites, so tell us what you'd like to build.

We haven't won yet. We need more yards using it, more data to make the models better and more integrations to make it easier to plug in. But we're bringing this to the world so that yards make more money, bodyshops and insurers buy more recycled parts, and we cut CO2 in a meaningful way.

Sources

  1. Competition and Markets Authority, Copart / Hills Motors phase 1 decision (ME/7010/22), January 2023, para 72(a). This is the companies' own submission, not a CMA finding. https://assets.publishing.service.gov.uk/media/63b4004fd3bf7f36af590df1/Copart_Hills_-_Phase_1_decision_-_Non-confidential_publication.pdf
  2. Pang, Sun, Geens and Park, "Orchestrating Circularity in UK Accident Repair", Business Strategy and the Environment, July 2026, section 4.5.3 (open access, a case study of one large UK recycler). https://doi.org/10.1002/bse.71259
  3. DVLA figures obtained under FOI by Allegiant Finance Services, March 2025. https://pressat.co.uk/releases/a-car-written-off-every-minute-dvla-foi-data-unveils-a-46-surge-in-vehicle-write-offs-as-cmc-warns-of-historic-vehicle-valuation-under-settlements-0d5698624488e0a734bab838d10fad06/
  4. Association of British Insurers, "Motor claims hit record £11.7 billion in 2024", February 2025. https://www.abi.org.uk/media-hub/news-post/motor-claims-hit-record-117-billion-in-2024
  5. Financial Conduct Authority, Motor Insurance Claims Analysis, July 2025. Average across 11 firms, £2,350 in 2019 to £3,344 in 2023, petrol and diesel cars. https://www.fca.org.uk/publication/multi-firm-reviews/motor-insurance-claims-analysis-multi-firm-review-2025.pdf
  6. Eddie Longworth, CEO of e2e, Insurance Times, November 2024. Sponsored column, 2023 figures, an estimate. https://www.insurancetimes.co.uk/expert-views/living-the-green-transformation-e2e/1453591.article
  7. ABP Club, 2025 State of the Body Repair Industry Report, as reported by Insurance Edge, December 2025. https://insurance-edge.net/2025/12/15/silverlake-welcomes-abp-club-report-on-bodyshop-repairs/
  8. Aviva, "Almost three in five motorists likely to use recycled car parts", September 2023. https://www.aviva.com/newsroom/news-releases/2023/09/almost-three-in-five-motorists-likely-to-use-recycled-car-parts/
  9. Insurance Times, September 2021. Hills Salvage filled 44.58% of green part requests on its Ageas contract over 90 days. https://www.insurancetimes.co.uk/news-analysis/green-parts-strike-environmental-and-economic-gold-so-why-are-insurers-holding-back/1438934.article
  10. LKQ SYNETIQ, parts breaking page. https://www.synetiq.co.uk/parts-breaking/
  11. VRA Certification, "Repair, Reuse or Replace" (Oakdene Hollins with SYNETIQ, Allianz and Metsims), August 2024, Table 2. Commissioned by the recyclers' trade body. 131.8 kg with UK green parts against 282.6 kg (new, made in Europe) or 309.0 kg (new, made in China). https://www.vracertification.org.uk/wp-content/uploads/2024/09/ODH-report-19-08-24-EU.pdf
  12. Automotive Recyclers Association / WPI, carbon study executive summary, 2025. Paid for by the US recyclers' trade body, and counts only the steel, aluminium and copper. About 711 kg for a 2015 Toyota Camry engine and 1,002 kg for a 2015 Ford F-150 engine, on the study's conservative basis. https://www.a-r-a.org/wp-content/uploads/2025/05/executive_summary_of_wpi_carbon_credit_study.pdf
  13. Konkurrensverket (Swedish Competition Authority), "Makten över bilen", report 2019:2. https://www.konkurrensverket.se/contentassets/1e84ffaec8994ce8b180df7ca562ae62/rapport_2019-2-makten-over-bilen.pdf
  14. Our sums from official returns. Sweden: Eurostat, 23,582 t reused from 134,177 cars in 2023 (176 kg) and 39,577 t from 186,370 in 2019 (212 kg). UK: OPSS / Defra return for 2018, the latest UK figure, 39,883 t reused from 1,406,975 cars (28 kg). Both figures are by weight and include tyres and batteries. Countries measure reuse differently, and the UK return is marked draft and may undercount parts sold, so treat the gap as a rough guide. https://ec.europa.eu/eurostat/databrowser/view/env_waselvt/default/table and https://assets.publishing.service.gov.uk/media/5fe1e48bd3bf7f089f999887/elv-waste-statistics-2018.ods
  15. POXA UK sold parts data for a first-generation MINI Cooper S hatch (1.6 supercharged, 6-speed manual), sales recorded in the 12 months to 11 October 2026. Main parts only (13 part types, 18 pieces), with the front-end parts taken off, parts worth under £40 and parts unlikely to sell within 6 months left out, and nothing already in stock. About £2,300 at typical prices, or about £1,470 if every part went at the cheap end of what buyers paid.

Try POXA free