Analysis
Buy Last Year's Model: The Quiet Discount Hiding in Plain Sight
The moment a new model launches, the old one gets nearly as good and a lot cheaper. Here's when that's the smartest buy in the store, and the few times it's a trap.
Every year, on a schedule you could set a calendar by, a company announces the new version of a product you already wanted. The reviews go up, the launch coverage rolls in, and the model that was the flagship a week ago quietly slides down the price list. Most buyers watch the new one. The disciplined buyer watches the old one, because that price cut is one of the most dependable discounts in retail, and it usually costs you almost nothing in performance.
The mechanism is simple and it works in your favor. A new model launches, and the previous generation has to make room. It doesn’t get worse the day the successor arrives; it does exactly what it did the week before, but now it wears a lower sticker because the store needs the shelf space and the maker needs the headline to be about the new thing. You are buying the same engineering, the same build, the same capability, at the price the market assigns to last year’s story.
Why the old model is usually the smart buy
Start with performance, because that’s where the fear lives. In mature product categories, the yearly update is incremental by design. The core problem was solved several generations ago, and what’s left is refinement: a faster chip you won’t feel in ordinary use, a marginally better sensor, a redesign of something that already worked. The gap between this year’s model and last year’s is real on a spec sheet and nearly invisible in your hands. When performance has plateaued, the newest version is mostly selling you recency.
Then there’s the part the launch coverage never mentions: the old model is a known quantity. It has been in the world long enough for its real flaws to surface, for the early firmware bugs to be patched, and for long-term owners to report what actually breaks and when. You are buying a product whose weaknesses are documented instead of one whose weaknesses are still being discovered by the people who bought it first. The new model is the experiment. The old one is the result.
Support and parts follow the same logic. A product that shipped a year ago still sits inside its support window, still has accessories on the shelf, still has replacement parts in the channel and a community of owners who have already solved the common problems. You get maturity and a discount at the same time, which is a rare combination in any market.
When paying for the new one is right
Pretending the newest model is never worth it would be its own mistake. There are cases where the premium buys something real, and skipping them to save money is a false economy.
The first is a genuine platform jump. Most yearly updates are refinements, but occasionally a category takes a real step: a new architecture, a shift in the underlying technology, a change that alters what the product can do rather than how fast it does it. When the difference is a platform and not a polish, last year’s model isn’t a discount, it’s a dead end. The tell is that the change enables something the old one simply cannot do, not that it does the same thing slightly better.
The second is support that’s ending. A price cut on hardware whose software updates or security patches are about to stop is not a bargain; it’s a countdown. This matters most for anything connected to the internet or holding your data, where an unsupported product becomes a liability the moment the updates dry up. Check the support timeline before the price, because a cheap device you have to replace in a year was never cheap.
The third is a safety or reliability fix. When the new model exists specifically to correct a flaw that hurt people or failed early, the older version’s documented weakness is exactly the thing you’d be buying. This is the reverse of the usual case: here the known flaw is a reason to avoid the old one, not a reassurance. Anything protective, load-bearing, or built around a recall deserves the current version.
The decision rule
You don’t need to track product cycles like a trader. You need one question, asked at the moment the new model lands: is the difference between the two a refinement or a jump?
Pull up the actual changes between last year’s model and this year’s, and read them as a list. If every item is a spec bump, a new finish, or a feature you’ll never invoke, buy the old one and pocket the launch-day discount. If a single item on that list is a platform change that unlocks something new, an end to the support you’ll rely on, or a fix to a genuine safety or reliability problem, pay for the current model and treat the premium as insurance.
That’s the whole method. The default is last year’s model, because in most categories the update is incremental and the discount is immediate and the reliability is proven. The exceptions are specific and knowable, and they announce themselves plainly if you read the changelog instead of the marketing. Buy the old one on purpose, check the three exceptions before you do, and you’ll spend less for something you understand better. It’s the same discipline that runs through every one of our reviews: buy the product, not the story printed on the box.
Real Buyer Experiences is reader-supported and independent. We synthesize independent testing and owner reports into one honest verdict - we never trade a recommendation for payment. How we review →
