It's not just in my head that it performs poorly; it genuinely doesn't work as well in my area compared to the alternative and maybe it is because of the top-down approach. Believe me, I would prefer to use it because its interface is much better, but not at the cost of being stuck in traffic. I test it against well-known routes during my Monday-Friday commute, routes I've been driving for years.

The UC Berkeley study you mention in passing is pretty outdated (2019) at this point with how quickly tech and algorithms improve. I would have agreed with that study's findings during that period though. For example, with the adoption of generative AI, I've already noticed improvements in GM just in the past year with navigating parking lots and store entrances. I'm sure GM is using AI to help with traffic and navigating as well.

This more recent study from Vardy (2023) better reflects my personal experience with Here WeGo. Furthermore, this 2024 study discusses how dynamic navigation can improve overall traffic flow if a certain adoption threshold is met, which makes sense since you're basically spreading out the traffic that would otherwise be stuck on the main artery.

Below is just one example between two points in my area that I have to do twice a week during 4p rush hour where I will test Here WeGo vs GM:

​​​​There are three main route options from my kid's school to the rec center for various sports twice a week. I've done this commute twice a week for three years now.

• Route A: 11 minutes with no traffic, 15-30 minutes at 4p, 6.2 miles, primarily uses I-5 (the main artery).
• Route B: 13 minutes with no traffic, 15-17 minutes at 4p, 5.2 miles, uses neighborhood streets (a third artery that became a viable route with locals well before Waze and Google Maps became popular; So well known with locals but not with commuters).
• Route C: 15 minutes with no traffic, 20-22 minutes at 4p, 5.5 miles, uses common surface streets (a secondary artery, also passes by my house).

Both Waze and GM typically suggest routes A or B based on current traffic conditions and have never recommended route C. On the other hand, Here WeGo consistently recommends route C. Additionally, Here WeGo tends to navigate off the main roads of route C onto less congested secondary streets, which have more stop signs, including both two-way and four-way stops - - this kind of goes against the top-down approach. My assumption is that Google Maps' Street View has allowed them to enhance their database and algorithms to account for these stop signs, whereas Here WeGo does not seem to do so.

In the above example, I find myself taking route B about 90% of the time, and occasionally route A depending on the traffic on I-5. I only use route C when a stop at the house is necessary but generally avoid it as I know it consistently adds 5min to this specific commute vs route B.

Similar to the Vardy study, my experience with the ETA on Here WeGo isn't great. While it dynamically adjusts as you're driving, the initial estimate isn't as accurate as GM.

I think Here WeGo is exactly what you state, top-down navigation and that's actually why it consistently fails in my example above. Plus bad algorithm because it often does weird routing outside of main surface streets onto secondary streets. I pay for Mercedes Me and receive the map updates and live traffic so it can't be out of date.

Anyhow, I'm happy it works for your situation, as I have said, it would be my preferred navigation because the implementation is way better (HUD, fullscreen maps on both center and driver displays, AR which I don't really like but my kid enjoys it).

 

To be fair to what you are saying, I don't do that kind of driving. The real issue here is that you have to drive everywhere to get stuff done and worse you drive during commute hours. I've long tired of that insanity. Actually, I've never really done it in the first place. I've always chosen to live somewhere where I have alternatives to the car and the things I need are close by. Most of my weekly errands are walkable. My gym is a 15 minute walk from my house, so that doubles as the warm up. I have two grocery stores within 5-10 walk. My optometrist is a 15 minute walk from my house etc. Back when my wife and I commuted to an office, we both did so by train. The train station is also only a 15 minute walk from our house. For the errands I do have to do by car, I mostly use local knowledge to get there on less travelled roads and I do those outside of commute hours. Admittedly, I do use Google Maps if I must go somewhere during busy times, but mostly I use GM to check in advance how long it would take me and then if I don't feel it, I simply postpone or cancel the trip.

The issue is none of these systems are solving the traffic problem that is choking this country. The solutions to the car problems in this nation seem to be more car problems instead of considering alternatives. Now Musk etc. want everybody to own a self-driving car. What do you think the result of that will be? It won't solve anything. Everybody will just be stuck in worse traffic with even more cars on the road, except those cars will be self-driving. It's modern Sisyphus. I refuse to partake in this clusterf**k. I mostly use navigation systems for longer trips to unfamiliar areas, that are actually enjoyable to do by car and not attempting to save 5 minutes in the daily driving circus.

 

I'm totally with you. Besides the commute of taking my kid to school and the after-school activities, I avoid the gridlock as much as I can as well. The school is within a bikeable distance and I'd be up for biking it, if we lived in Amsterdam, but as you probably know, the SoCal drivers during commutes are a force not to be reckon with. And home schooling is definitely not an option in this household. 🤣

Similar to you, I work from home and have a flexible schedule and all of my conveniences (groceries, gym, medical wellness, etc.) are within a 5min drive or 20min walk. I actually bought a really cheap ebike ($250 so if it gets stolen I won't care that much) and bike to most of my conveniences. I average around 5k-6k miles annually half of that is taking my kid to school.