S25 Sales Mystery: When Numbers Don't Add Up
During the lead-up to the launch of the Galaxy S25 series of phones, there was a lot said about how little Samsung’s next line of phones had changed compared to last year. The design was largely untouched, the specs only saw typical yearly upgrades, and in the case of the S25 Ultra, some features were even removed. The S25 Ultra would launch with a diminished version of Samsung’s famous S Pen. One that lacked the Bluetooth functions Samsung claimed virtually no one was using. The primary features in question were Air Actions and the ability to use the Pen as a remote shutter for the camera.
Samsung Galaxy February Sales Trends– Hana Securities
— Jukanlosreve (@Jukanlosreve) March 28, 2025
Samsung Electronics recorded sales of 19.03 million units, reflecting a 3% decrease compared to the same month last year but a 15% increase from the previous month.
• Sales declined year-over-year in all regions except…
Nonetheless, the S25 phones all reviewed well and while I did criticize the Ultra for being a bit too iterative, my overall review was positive. The phones are rock solid, if a bit boring. The general consensus from the internet tech world was that these boring phones with removed S Pen features and far too much reliance on AI features would surely sell poorly.. But did they?
On February 28th, ETNews reported that the S25 Ultra had set a new sales record. It was the fastest Samsung phone ever to 1 million sales in South Korea. Taking on 21 days to achieve 1 million units sold means the S25 was off to a surprisingly great start that even the much lauded Note 10 could not match. But that’s just one region and just a three week span.
At the end of March we got much more information that paints a slightly confusing picture. Twitter user @Jukanlosreve posted a synopsis of some data from Hana Securities, an investment firm in Seoul, South Korea.
For the month of February, Samsung apparently posted sales totaling 19.03m units which was a 3% decrease year-over-year, but a 15% increase from January. That sounds middling at best, bad at worst and a far cry from the record settings sales from South Korea.
However, he goes on to state that the Galaxy S25 series sold a total of 4.56m units in the first month. That’s over three times what the S24 series did a year prior. Here’s the direct breakdown of the numbers by model.
S25 Ultra 2.55m
S25 Plus 840K
S25 1.17M
S24 Ultra 830K
S24 Plus 270K
S24 320K
Here’s where things get really confusing. In the tweet, Jukanlosreve says that if you break down the sales by day after the pre-order window closed, the S24 series sold better. As this was hard for me to believe based on the numbers, I decided to try and do the math. Maybe the S25 sold really well in pre-orders, but tanked after launch? That is precisely what other publications had taken from this data, but let’s see if the numbers make sense.
Jukanlosreve claimed that after pre-orders closed, the S25 sold 180K units per day and the S24 averaged 240K per day. Keep in mind, the sales totals we have (4.56M for the S25 and 1.42M for the S24) are pre-orders plus sales after launch. Let’s get the dates together so that we know how many days we’re looking at.
Galaxy S25: Pre-order period ended January 25th, 2025 and official release was on January 31st, 2025.
Galaxy S24: Pre-orders closed February 3rd and were released on February 7th, 2025.
So, after the pre-orders closed the S25 had a full 28 days at 180K units per day and the the S24 had 21 days at 240K per day. Here’s the problem: if this were accurate, both lines would have sold over 5 million units after pre-orders closed… How can the post pre-order first month numbers be higher than the total first month sales? Unless I’m missing something, there is a serious error in this latest numbers. There is simply no way that in their first months the S25 sold 4.56M units while the S24 sold 1.42m units and yet the S24 sold more per day.
Daily totals post pre-order would have to be no more than 162K per day for the S25 and 64K for S24.
Even turning to ChatGPT for a different perspective, these numbers just don’t add up. It calculated “first month” as a 30 days span, not just the month of February. The conclusion? There’s no way the S24 averaged 240K units per day.