I often think about this precise scenario. There are so many cases and times where a big company shuts something down that would be perfectly viable at a different scale. My previous company got acquired by Verizon and then shut down because it didn't scale to a $1B business within five years (or something like that). I'm not claiming it was a viable business in the first place (it probably wasn't) but now we'll never know. And for this story, there's a thousand others. I understand that sometimes the small company becomes intertwined with the big company to a detrimental extent, but there are also plenty of cases where a separation is still very well possible. The separation doesn't even need to be the business whole, but it could be a constellation of IP and resources, for which someone creates a new business entity. I wish that more execs created that possibility, instead of just writing off a financial loss and calling it a day. While the write-off makes business sense to the bigger company, and there may not a huge business justification for spinning off instead of shutting down, there's also just something to doing the right thing in terms of providing continuity to the users of your product, and giving engineers a chance to double down on a product development that they may have faith in.
From the company’s perspective it’s a straightforward calculation.
Considering that this will only apply to failing products, the price they can expect to get for it is negligible, more so considering the buyer is going to be a handful of regular employees rather than a big company. So selling vs killing makes no material difference to their bottom line.
On the other side, there’s a mountain of due diligence they have to do to make sure that the decision is sensible. Is the product tied to the company’s brand at all? Will it reflect badly on them if the new owners decide to take it in some different direction? Or can it be used to compete with the company in the future? Does it have user data? Are they in accordance with the TOS and a thousand data laws when selling it to someone else?
So from the business side the sensible answer is usually to just kill the product.
Will it reflect badly on them if the new owners decide to take it in some different direction?
I think there's still a flip side to these decisions. Google has an established reputation as a product killer, I'm sure the effect of that won't bring them down but it's there and I doubt handing a product off to someone else would either.
> Considering that this will only apply to failing products
Companies like Google often shut down products that would be called incredibly successful if they were a startup or small business. In $BigCorp, the yardstick is their main moneymaker, and if you are less profitable and don't fit in the larger strategy you can get cut simply to help the company keep focus on what's important.
On the other side, there are so many companies that would never really make it without a lifeline that extends into the hundreds of millions. Even after IPO, there are so many companies losing heaps of money. So maybe Google just doesn't want to be losing $100M a year for 10 years.
I worked for a tiny, independent project for a giant corporation for a couple of years. It absolutely would have worked as a 2 or 3 person company but it was definitely the wrong place for a dozen engineers as part of a giant corporation. They shut it down, but now I'm wondering whether they would have entertained a small offer to buy it. Probably not, but I guess I'll never know.
There are similar examples within companies. For instance, I led the establishment and growth of a ~50m cyber security practice for a US vendor in a region. Ended up ~22% EBIT. Unfortunately the software business is >65% so my business was diluting regional performance. I got the call to slow down growth so I left. Ultimately they shut down the services altogether and gave it to channel. I offered to buy the services business but they weren't interested.
As a viable language-learning app (on the freemium tier), it's mixed and getting worse. I'm a 3.5-year Duolingo user.
See the 2/2023 HN post "How Duolingo reignited user growth" [0] describing how they/ the former CPO aggressively traded off reducing the utility for actually acquiring language vs too much gamification (not all users want 'Candy Crush' levels of gamification), user growth, leaderboards, challenges, cutesy animations, inventing silly revenue streams like 'streak protect' (all prior to the 7/2021 IPO, ticker 'DUOL').
Duolingo locked the Chinese course mid-2022 [1], that means no more fixes ever, no corrections, the broken or missing audio (and answers, and hints) stay broken or missing, all discussions (users helping users, handy links to third-party sites) are frozen. For no good reason. They don't give a hoot for gamifying valuable user contributions which add/improve course content/fix mistakes. If any of you were contemplating paying for DL Chinese before they locked it on the charitable basis that they'd fix the broken bits someday after they'd banked the IPO loot, well obviously not now. (Here's e.g. Langoly's list of "The 11 Best Apps to Learn Chinese (Mandarin) Fluently" [2])
By comparison the Spanish course is alive and being improved; they added Stories.
TL;DR: optimizing the product to juice up metrics like DAU prior to the IPO is good for valuations, but is not a great long-term approach to a language learning app and its community.
I agree that there can sometimes be a tradeoff between retention and learning. But having been on the inside the overwhelming impression I got was that Duolingo really does care about teaching effectively. They have a whole Learning Area with multiple teams dedicated to this and I believe the CEO and exec team really care about it.
Obviously Duolingo isn't perfect but they're very aware that there's still lots to improve. And sure, maybe their approach to gamification isn't for everyone, and that's fine, since there are many people who do like it!
No, I'm talking about the tradeoff between monetization/gamification vs learning. Too much of the former is degrading the latter and this has been getting worse since they switched to juice their metrics (CURR, DAU, $/user) for the IPO, as Mazal posted.
User Kortaggio [0] talks about how Duolingo's claim “an average of 34 hours of Duolingo are equivalent to a full university semester of language education” is misleading because it's based on their long-discontinued SRS algorithm. Sounds like false advertising.
The gamification features are one-size-fits-all (whether you're age 13 or 30 or 65, occasional or dedicated user) and default-on: interstitial animations, success videos, chests of gems, then yet more 5/15/30-second ads... I challenge you to measure the % of time within a 10-minute segment that is actually spent on learning. Or at minimum, they could show me the cutesy stuff while I'm doing the language task, Super-Puzzle-Fighter style.
I'm a 3.5-year user (>1200 days), I clearly don't need somersaulting owls and faux-challenges to keep me going and I passed that point 1190 days ago, if not always.
Many (adult) DL users write about how the ever-increasing pushiness with leaderboards, quests, streaks ('create false urgency' per Candy Crush which the former CPO cites as a model) turn them off. Should have different use-modes for different segments of user.
I believe DL care about teaching effectively only up to a limit, and that limit is anything that might reduce their metrics, as a freemium app.
They don't care about teaching Chinese effectively, not since back in 2021 if not earlier, they simply gave up. This is quite sad and weird for the world's second-largest language. There isn't "still lots to improve" on the Chinese course, they simply gave up and locked the forums permanently so that now not even well-intentioned volunteer users can fix that one.
Also, DL committing intentional trust and GDPR violations without any online opt-out doesn't sound ethical "Facebook Still Tracks People on Yelp, Duolingo, Indeed" [1]. Obviously tracking without user consent is for monetization of ads. Like I said, monetization trumps other considerations at DL.
(PS I signed up for your ReadLang, but for Tagalog it's pretty beta. Good luck with its second incarnation).
(Update: China banned for-profit foreign tutoring 7/2021. But that still doesn't justify why DL locked their Chinese course, leaves lots of other countries where people want to learn Chinese.)