Apple has often said they will do products where they can do something better (and different) that everyone else, but if there is someone doing something very well then they are not as likely to do that. ( Apple buys screens and RAM from others. Apple doesn't do printers. etc. )
A router doesn't really have to "integrate better". It needs to do the job :
1. general internet security for EVERTHING on the local area network.
2. the 802.11__ standards signals.
3. the modem WAN interface ( keep up. > 1Tb/s )
Whose in home LAN now is 100% Apple products. If don't probably interact with the modem then whole thing is pointless (and the modem is a 100% non Apple device). There are also TVs, refrigrators , smart plugs , doorbell cameras , etc etc. The diversity of stuff on a local LAN network is substantially different now than 20 years ago.
The software that runs on the network on a variety of devices is better now. AirPrint (wire print ) isn't pointing at an Apple printer and/or a AEexpress attached printer. Stream audio. ... pretty device independent standard these days.
Protocols like Threads/Matter ... again device independent.
Even if narrow it down to just router administration. Administer router from an iOS/iPad app.... who doesn't do that at this point? Likely in a much more inclusive manner also because likely have an Android app also.
The growing barrier to entry for Apple is that the ISP's themselves want to rent/sell the router these days. (pretty good chance most ISPs have known the following for years , but some independent look at it. )
" ...
Key findings
- Opensignal data shows much faster performance for Broadband Download Speed on ISP-provided routers than on those third-party ones — around 60% faster for cablecos and telcos and around 45% for overbuilders
- For cablecos, we observe a much greater concentration of low-speed readings among Bring Your Own (BYO) routers, while ISP-provided routers have higher concentrations of download speed readings at faster speeds.
- Opensignal data demonstrates a correlation between a higher share of ISP-provided routers among a cablecos’ customer base and cablecos’ lower levels of churn among non-mover switchers (i.e. those who did not change addresses).
..."
The use of ISP-provided routers links to retaining fixed broadband customers in the U.S. | Opensignal
Robert Wyrzykowski, Micah Sachs and Chris Mills Key findings
There are bad modem/routers from cable vendors in some narrow cases , but, largely folks who buy their own routers/modems tend to squat too long on the devices. [ Usually where the ISPs provider gets into substandard performance zone is where they do the same thing. Using rental fees to move the equipment forward over time versus goose profits higher. ]
The ISPs getting into the Telephone ( VOIP and/or " MVNO+wi-fi calling " ) business it is even bigger hurdle ( especially with 'all-in-one' telecom/router/modem boxes ) .
But 'everything' on Wi-Fi pragmatically means non Apple products.
Similar issue with the Apple clients. Most mobile Apple devices will interact with a much higher number of routers outside of the owner's residence than inside the residence.
A highly competitive 'mesh' product just means even more interaction with non Apple products and deeper 802.11 standards Apple has to implement on the 'base station' side. Managing 'wireless' backhaul. Dealing with optional wired backhaul. etc.
By 2013-2015 players like Eero ( acquired by Amazon 2014) and Plume were springing up along with the other standard router players. Mesh somewhat stopped the complete dominance of 'race to the bottom' on router pricing. Once other players could sell more profitable routers , then they could add lots of features that Apple would have to heavily complete with (that Apple didn't necessarily have).
P.S. by 2013-2015 time frame the TimeMachine backup 'feature' of the Airport Extremes was largely diminished and undifferentiated also.
