The steady progression of new people - some created every day, new homeowners and what have you was enough to give a vibrant business, and ongoing sales.
What changed was the deal.
Pre 1980s most companies tried to balance the needs of all - staff, customers, local area, and shareholders. Post 1980s only the shareholders count, all else is secondary. Products and offerings are made and subscribed accordingly. The few folk still making stuff to last seem to have an anachronistic worldview to go with it. Perhaps still care about craftsmanship, repairability, and reputation, rather than built in obsolescence, subscriptions and DRM'd parts. Now they can sell you the thing 15 times, perhaps 12 of those without good cause apart from greed.
Not exactly "farm tractors that last forever".
Not too different with Hoover vacuums in the era of the Hoover Senior and Junior -- they stuck around, being repaired and reconditioned easily for decades. Had Hoover or Dyson not gone with cheap throwaway plastic for everything when cyclones meant capability improved in post 1980/1990 models we might have those lasting 30+ years easily too. As is they self destruct in a decade or less as more and more bits break. The (cheap plastic) spares that are available are priced such that you're heavily incentivised to replace. One or two commercial brands still make products that might last a while...
Most of the current kitchen appliance brands are mainly now just worthless logos, usually owned en masse in a larger group, probably after a leveraged hostile buyout. Now made in the same single factory and generate churn with new models every year - for the sake of difference not because there is any improvement, just deliberately changed from the previous. Spare parts might be just filter, lamp or element, often deliberately unique to model, and will be available at vastly inflated price for limited time only.
No end of other examples from hand tools, garage equipment, and home goods through to the most complex products.
I think you have a very strong set of rose-tinted glasses. Planned obsolescence was a strategy long before 1980.
I could compare and contrast in just about every sector I can think of. Sure there were some abuses, notably some of the car makers come to mind. The needs of the customers -- beyond making the initial sale -- were far higher up the priority list than today.
I can't help but think they regret making these things that durable.
If you sell software, do you let the old stuff atrophy and become insecure? If not, you are spending salaries on maintenance without any associated revenue. As time goes on, the maintenance costs increase without bringing in another cent.
The shining example for actually doing it right is Jetbrains. There is a schedule for new releases they keep to. You get loyalty discounts for long term subscriptions and ending a subscription entitles you to a version that is ~1 year old in perpetuity.
Plus the prices are (IMHO) entirely reasonable.
Almost no one else does this. Other companies will, say, take the MSRP of the software package, divide it by 12 and give you that as the monthly price with a minimum contract period of 12 months. They'll also restrict you to choose platform when you purchase and may limit the number of installs rather than concurrent users.
No direct affiliation, but I've shared drinks with the devs and am a very pleased user.
This might be bad for a tractor manufacturer; but an everlasting tractor would be great for society, right?
The incentive now is to innovate in order to create a better more efficient product. E.g.; a more fuel efficient tractor which increases productivity - then you'll get another sale.
Deliberately hobbling your product so that it needs repair or replacement disincentives innovation and is a form of rent-seeking https://www.investopedia.com/terms/r/rentseeking.asp and as such should be treated as a market failure. (I.e., legislated against)