Apple Maps vs. Google Maps vs. Waze

Introduction

In early 2017, a conversation with yet another Waze fanboy finally nudged me to start a navigation app experiment. I was skeptical that the Alphabet owned company could meaningfully best its parent’s home grown Google Maps. I was also curious whether Apple Maps had discovered competence since its iOS 6 release.

I thus set out to answer three questions:

  1. Which navigation app estimates the shortest travel time?
  2. How does each app over/underestimate travel times?
  3. Which navigation app actually gets you to your destination most quickly?

This exercise lasted the majority of 2017 and led me to dread almost any car trip due to the self imposed data gathering tasks that came with it. Nonetheless, my wife and I persevered, and I hope this data serves the community well.

Methodology

Ideally, I would have tested all routes with two other drivers, each of us departing from an origin simultaneously to follow a different navigation app’s guidance to the destination. This would have resulted in a direct comparison of the apps’ actual performance for all observations. Alas, I lacked two fellow drivers who thought this was a worthwhile endeavor.

Instead, I recorded 120 observations (i.e. trips) of navigation app estimated and observed performance. For each observation, I randomly selected to follow one of the three navigation apps and recorded:

  • Which app was randomly selected to follow for each trip
  • Estimated driving time for each app (i.e. Apple Maps, Google Maps, and Waze)
  • Departure and arrival time
  • Traffic conditions (i.e. work commute hours or not)
  • Weather conditions (i.e. rain or not)
  • Driving type (i.e. >75% city, >75% highway, or mixed)

Sample Data Log

Navigation App Log

Results

I summarize the results with three simple charts:

Estimated Trip Time: Which navigation app estimates the shortest travel time?

I first looked at estimated trip time. This considers only the upfront estimate from each app and not how long it actually took to drive the proposed route. For purposes of visualizing the data, I indexed performance of Apple Maps and Waze to that of Google Maps.

nav_estimated_trip_time

Relative to Google Maps, Apple Maps estimated trip times were on average 8% longer (i.e worse) and Waze estimates are 3% shorter (i.e. better). These results were largely consistent with what I expected given Waze’s a strong following of users who swear it is the best option among navigation apps. If the estimated trip times consistently predicted actual driving time, Waze would be my preferred navigation app and I believe this is as far as most Waze users get in the navigation app decision process.

Average Error: How does each app over/underestimate travel times?

I next considered average prediction error for each app. For each observation, I calculated the difference between actual observed trip time and estimated trip time, i.e. (Observed Time)/(Estimated Time)-1.

nav_average_error

Average error results were the exact opposite of estimated trip time. Using Apple Maps, I on average arrived  1% faster than initially estimated, versus 2% slower with Google Maps and 11% slower with Waze. In other words, Apple sandbags its estimates so that users on average arrive at the predicted time or slightly sooner. Google and Waze are overly optimistic in their predictions and thus their users arrive later than expected.

Error Adjusted Estimated Trip Time: Which navigation app actually gets you to your destination most quickly?

Finally, I combined estimated trip times with estimation errors to derive error adjusted estimated trip time, i.e. estimate for actual time to get to destination. These results are again indexed to Google Maps for purposes of presentation.

nav_error_adjusted_estimate

Adjusted for prediction errors, not only does Google Maps outperform its competitors, Waze is actually the worst performing of the three.

Results Summary

If you want to get to your destination most quickly, use Google Maps.

If you want an accurate prediction from your navigation app to help you arrive at your destination on time, use Apple Maps.

If thinking you’ll get to your destination quickly helps to ease your commuter anxiety, use Waze.

Closing Thoughts

The performance of the three apps sparks a set of questions regarding incentives and business strategy.

For Apple, Maps is a basic solution for its average user who wants a maps solution out of the box. Apple Maps does not directly drive ad or subscription revenue for Apple so there is less reason for Apple to incentivize iOS users to use Apple Maps over other solutions. However, Apple does care about user experience, and sandbagging trip time estimates so that users arrive at their destination on time results in a great user experience. Hence, I believe that Apple is intentionally conservative with estimated arrival times.

At the other extreme, Waze (Alphabet) makes money through ads when you use their app. What better way to get people to use your navigation app than by over-promising short trip times when no one takes the time to record data and realize that you under-deliver? If an unsuspecting user opens Apple Maps and sees a 34-minute route and compares that to 30-minutes in Waze, the deed is done. Now Waze has a life-long customer who doesn’t realize they’ve been hoodwinked and Waze can throw at them stupidly annoying ads.

“But wait! Waze leads me down super sneaky secret routes that avoid highway traffic jams.” Yea, so… is that a good thing?

Based on various publications, my experience using Waze, and anecdotes from other drivers, I do believe that Waze guides drivers down more “creative” routes. But the results shared above imply that Waze doesn’t get you to your destination any faster. So, is it better to spend 30-minutes following Waze through suburban neighborhoods and alleys than 27-minutes minutes following Google Maps into a highway traffic jam?

I think some people will say “yes — when I’m moving I feel like I’m making progress even if I don’t get to my destination any faster!” And I get that. Driving is stressful and the feeling of progress may help alleviate some commute-related anxiety. But is that progress worth wear and tear on your car and road infrastructure, additional accident-prone miles, and increased traffic in kid-filled suburban neighborhoods that results from Waze’s alternative routing?

Not for me.

Appendix: Limitations

The high level limitation of this experiment is the “synthetic” way in which I compare app performance by applying average prediction error to estimated travel times. As described above, an ideal experiment would involve three drivers. Furthermore, the results suffer from additional limitations, e.g.:

  • All routes were taken in the San Francisco Bay Area and thus reflect Bay Area traffic, roads, weather, etc.
  • 1/3 of the routes were for my specific SF city commute
  • Actual route times are heavily dependent on driver (my) behavior, e.g. speed, propensity to pass, use of carpool lanes
  • Navigation app routing algorithms constantly evolve, hence this analysis is simply a snapshot in time

Appendix: Data Highlights

  • All routes taken in San Francisco Bay Area
  • 120 observations — random number generator used to select app used for each route
    • Apple Maps: 41
    • Google Maps: 41
    • Waze: 38
  • Driving time:
    • Average: 26 minutes
    • Min: 8 minutes
    • Max: 1 hour, 15 minutes
    • Note: to check whether short routes were somehow skewing results I replicated the analysis above with only routes longer than 20-minutes and found consistent results (i.e. Google Maps performs best on error adjusted-basis)
  • Observations by driving type:
    • >75% City: 68
    • Mixed: 20
    • >75% Highway: 32
  • Observations by traffic conditions:
    • Commute / rush hour: 40
    • Non-commute: 80

73 thoughts on “Apple Maps vs. Google Maps vs. Waze”

  1. Thanks, awesome review.

    Only thing to notice is that you considered only Bay Area. Apple Maps is amazing there… used multiple times, it’s always my choice there. But it’s not that good outside of it and it really horrible outside of US.

    For me, the obvious choice is Google Maps every but in Bay Area.

    Like

  2. While I enjoyed reading your experiment, I believe you forgot an important factor that is where you drove, and drew conclusions a bit too early. I have noticed different levels of accuracy in different locations. For instance, in Paris France, Google Maps is very inaccurate, especially for traffic estimations. In the SF Bay Area on the other hand, I arrive earlier than expected pretty much all the time.

    You don’t describe what actually made the timing differences. Are the suggested routes the same? Or are traffic estimations wrong? What are the traffic estimation providers in your area? Are they the same for each platform?

    I believe that while the data you have is interesting, it is not complete enough to draw the conclusions you came to. In my mind, there are too many unanswered questions that remain to say that either is better than the others in general. That being said, your data might show a local-optimum for your area.

    Liked by 1 person

    1. Thanks for the feedback! Check out the “Limitations” section of the post — geography is a huge limiting factor in the broader applicability of this data. In terms of “what actually made the timing differences” — that’s at least partly a limit of the depth of data gathered. Because I manually collected the inputs I wasn’t willing to capture every variable (e.g. how much slower than usual is the drive due to traffic). That said, when controlling the analysis for whether the route occurred during commute times, a proxy for heavy traffic, the results were consistent.

      Like

    2. Nevertheless this study was biased on on route, that was a great act that Brings up this threads. In SF the acccuracy of Apple is great, but as you said, on other cities as Paris, Lisbon, Mexico City , São Paulo and Milan I experienced that Waze performed better, especially due to the alternative routes. sneaking peak at the time only and not at the traffic of the route instead is, for me, the best way to decide which app I will use. Since I got to take roads with a lot of turbulence these days, Waze is REACTING better when bad things happens AFTER I left. Cheers and again, thanks all for the insights.

      Like

  3. I’ve been using Waze a while now. Not because of any perceived greater efficacy at route planning or it’s ability to take me on a magical mystery tour of urban residential neighburhoods, but the user generated feedback. Nice to know where those speed cameras have been parked today or where that truck has broken down on a freeway, as reported by others who saw them there recently. I’ve even contributed to the database that drives this. This social satnav feature is the one that drew me and holds me.

    One note I’ve made on the arrival estimates though seems to get no coverage in your piece is the dynamic nature of traffic. In my experience leaving for work before 7am the Waze estimate will be some 5 or so minutes optimistic. Leaving a half hour later it is more likely near 10 minutes too short. Now my commute is near 40minutes ideally and I would be unsurprised to discover that if I could instantaneously traverse the road segments of the proffered route (as an algorithm will) I would get the travel time that Waze generates. In the real world though, 30 minutes after leaving home as I approach the central city the traffic has increased and is moving more slowly. My travel time is blowing out a little. By now Waze has also adjusted it’s arrival time estimate based on how much travel it sees ahead. Invariably, by the time I’m approaching my destination it’s estimate is pretty good and as I noted likely about 5 minutes later than it said in my driveway at home.

    The other apps may be modelling historic traffic trends over time on each road segment in the route to get a time estimate when I’m expected to pass through it, or they may just be adding a fudge factor to allow for travel time. Either way, the effects of traffic lights, construction, and other random factors can easily add 5 minutes to a commute, so it’s hard to see +/-5 min as anything other than a given on any estimate I see and that if it’s within a few minutes either way as inspired coding!

    As I say, just a thought.

    Like

    1. Thanks for the feedback! I did not mention the social aspect of Waze and I agree that it almost certainly is valuable to many Waze users.

      While I did not attempt to explain why Google Maps appears to perform better, I agree with your thinking that better modeling of historical traffic data is a likely contributor.

      Like

      1. I think the social responsiveness is the key feature of Waze. I don’t even use it regularly, but I do for road trips.

        Like

    2. Yes, it’s been my experience that Waze is much quicker to change routes, adjust times, etc., when compared to Google Maps. Sometimes that approach gets you there faster. Sometimes, taking the historically average route ends up being best (which is what Google Maps tends to suggest). So in certain circumstances, like a car wreck, undocumented construction, etc., Waze seems to give more up-to-date info, but following its instructions doesn’t always yield a better outcome, because conditions change and sometimes it leads you into worse traffic than just taking the average route (or it asks you to make some kind of crazy left turn at place where making such a turn is very difficult because of the speed and density of the traffic. So you kind of have to use your judgment.

      I’ve also found it interesting that Waze doesn’t seem to give everybody the same directions from the same place. A friend and I were in the car and I had Waze going on my Android phone and she had it going on her iPhone. Waze suggested slightly different routes on the two phones. Don’t know if it’s trying to balance load among surface street routes, or whether it’s giving different routes to try and test which one is better.

      Kind of seems like Waze is a test-bed to help Google continue to improve the Google Maps algorithms. They know the Waze user is likely more of an enthusiast and willing to try different routes in hopes of gaining a benefit. To the extent that your data is correct, we Waze users maybe getting used by Google more than we’re benefitting from faster travel times.

      But I still like Waze. Over time, I’ve learned how to make my own judgements sometimes, and I like having the more immediate information. I also like being able to include a GPS speedometer on the Waze interface, which Google Maps doesn’t have. Helps me stay clear on what the speed limit is.

      Beyond that, while I’m sure Apple Maps is on par with Google/Waze as far as Bay Area map data is concerned, my sense is that it’s not nearly as good as Google/Waze in many other places. I know you raise that in your caveats, but it’s an important thing to consider imho. Results may vary in other locales.

      Like

    3. This is why I use Waze too (Chicago area). My own observations from using all 3 apps in this area is they are all about the same (as already mentioned).

      Like

  4. I would honestly like to see a full data log. Your seem to take a very straight forward approach, and I would be interested in looking at it from a deeper statistical perspective. There is likely a good bit of data to extract here that could explain when and how each app performs in different circumstances, which isn’t going to be apparent from the graphs you’ve generated.

    Like

  5. Would be interesting to benchmark Here maps against these numbers. I know Navtec used to provide the best maps, presuming Nokia, and then Daimler AMG kept this up it might provide a good alternative.

    Like

  6. In my experience, google estimates assume that you drive like an utter maniac the entire way and have no delay at any traffic lights. It think they just take the speed limit. For predictions I use here maps which do good traffic data. When driving, my Garmin does a very good job of estimating arrival times. Main issue is that it to easily assumes that nearby side roads will be clear when a motorway is congested and avoidance suggestions tend to be worse than just sitting out the motorway queue.

    Like

      1. Thank you so much for sharing your data!

        I also drive in the Bay Area daily for work, but have virtually opposite results (I will conduct a more formal investigation and report back, though, maybe I’m skewing things anecdotally).

        Most of my drives are almost exclusively on 101, during rush hour (btw 8 and 9 am, and 5 and 6:30 pm), SF to Redwood City, and back. Waze *always* overestimates my arrival time by 5-10 minutes, on a 45 min. commute. Now, I fully acknowledge that I’m the asshole in the sports car who jumps between lanes in traffic, and goes as fast as I can when there is no cars in front of me (up to 100 mph, if I can, which is also why I value Waze’s other features, like the police alert). So my results are probably not those of the average driver.

        Like

  7. Thank you for the effort, it was a great read.

    I have a question about predictions. Your post is on HN as well and I left a similar comment over there, so I will just quote the relevant part:

    – at 5 in the morning Google Maps will route me through a busy town 500 km down the road because there is no traffic now.

    – at 17:00, from the same starting point, it will avoid the same city, because of high traffic. This high traffic is irrelevant to me as I will hit the town in 7 hours, when there will be nobody on the roads.

    Have you seen a difference in this regard between the apps? Is any of them deciding on a road which takes into account what will happen a few hours from now and not base the decision on current traffic far away from me (and which will drastically change over the day)?

    Like

  8. Very nice experiment which made me think. Do you have any data about distance you have travelled? I wonder what it would results look like if we had data from many locations and compare in two separate categories: city and highways.

    Like

  9. Thank you so much to you and your wife for taking the time over the course of a year to do this research and share it publicly for free! Its people like you that make the world a better place and allow us to question and verify instead of just assume.
    I have had an ongoing conversation with a friend of mine regarding this very topic and the fact that Google/Alphabet bought Waze so many years ago from an Israeli company for $1 Billion in order to make their own Google maps directions service better. Thank you and i hope this gets you the genuine credit you deserve and possibly requests to do other unbiased product comparisons 🙂

    Like

  10. Nice post! I suggest you to show boxplots of the results inatead of mean value barcharts. This woud allow to quickly assess the significance of the deviations versus variability.

    Like

    1. That’s fair, and I did not consider any of the “social” Waze features. I will however say that in the Bay Area the ratio of false positive police reports to actual police presence must be at least 10:1. It makes me wonder whether the police place add fake reports to the app to get people to slow down…

      Like

      1. That is interesting on the false positive police reports because where I live (city of about 500k surrounded by rural areas), I would estimate that it is the opposite. The amount of money has saved me by avoiding speeding tickets has to be in the thousands of dollars.

        Like

  11. I love that you’re trying to see if the light in the refrigerator actually goes off when you close the door. I would think it significant how many times the app changed it’s path after the trip was actually begun.

    Like

  12. If I were King of the GPS world, my app would learn the streets I prefer to use. It is always interesting to watch how the app gets me to my house in my neighborhood. If I am constantly forcing my disembodied driving companion to recalculate its path due to my own predilection for a certain route, how bout learning my route?

    Like

    1. Yes so this was one of the most frustrating parts of the experiment. I believe that in some locations and under certain conditions I can outperform all the apps. I’m particular, in my neighborhood I know the traffic light patterns quite well and can hit a series of “greens.” However, for the sake of keeping the experiment unbiased, I always followed the apps’ primary route. It was a huge bummer to follow the app into a traffic jam when I very well knew that an alternate route would save me a few minutes.

      Like

  13. Oddly, I was wondering the same thing recently, and while I don’t have multiple trips or take as scientific of an approach, I did test Apple Maps vs. Waze on a ~ 20-minute drive in NJ.

    My wife and I happened to have arrived at the house of a friend separately. The drive home was during relatively low traffic conditions (a Sunday evening), and we were curious about the best route home.

    She used Waze, I used AM, the routes did differ slightly with her getting off the highway before me to take less traveled roads.

    She reached a bottleneck point slightly ahead of me (she was the first car at a light, I pulled up 5 cars back). If she had made that light and I missed it, she would have arrived about ~1 minute ahead of me. As it is, about a 10-second difference in our arrival times due to the other cars.

    Liked by 1 person

  14. Very interesting, I followed a similar methodology studying the accuracy of walk-times for Google Maps. Traffic signals mean that Google Maps underestimate walk times by nearly 20% in urban areas.

    Like

    1. The cynic in me wants to respond with “what about Yahoo Answers?” But given unlimited resources (and patience) it would have been ideal to include several other options, including MapQuest.

      Like

  15. There may be an opportunity to further this study going the speed limit on all roads (35mph in 35mph zone), vs going 7-10 over the speed limit (some theoretical limit in which police do not bother pulling you over because you are still driving safely), vs going the “speed of traffic” (going a speed so that you are passing as many cars are are passing you).

    That does make things more complicated…

    But, if Waze takes you down 30-40mph back roads, and you can safely drive 10-20% faster than the speed limit, then perhaps their “optimistic” measurement could actually be somewhat accurate, but only for a certain type of driver.

    Like

  16. My thing is, always use waze… but use your brain.. if ways says use 101… in the bay area… really rethink that Idea.
    Once you get to where your going by car and still have to walk blocks to the address your going… I.e having a company you have to go to that’s in f-ing san francisco.. switch to google maps for walking directions.
    waze only understands car’s if you try to use it for walking you’ll go insane.

    Like

  17. FWIW, I’ve noticed that Waze has gotten steadily worse in London re: predicted time vs actual. I think the traffic algorithm underweights the difficulties of turning across the street at busy times (or at least underweights how long it takes for a column of Waze using traffic to turn across a busy street). It’s also really bad at understanding that maybe I didn’t take that turning b/c of something I saw and no I don’t have time to stop and report the blockage.

    I’ll note in passing that like you, I find in areas I know all the apps have a significantly limited understanding of traffic patterns compared to a human.

    Like

    1. True, coming back from Lake Tahoe to Bay Area on Monday in heavy traffic, I noticed Apple Maps 1) couldn’t anticipate that the Monday holiday was going to create longer than usual travel times and 2) I noticed that many of the orange & red congestion highlights on the route would disappear as my car started going through such sections at 40mph w/o any congestion.

      Like

  18. I have been wondering about this question myself, with Tesla’s navigation as one of the apps I’d like to include. (I presume you had good reasons such as not having a Tesla.)

    From your experience, what would you recommend for someone doing their own similar experiment?

    Like

    1. Yep, no Tesla.

      The biggest issue for me was remembering to record arrival time when I got to my destination. This seems simple but an added wrinkle is that I recorded the time when I arrived at my destination, not when I parked my car. Arriving at a destination often means negotiating pedestrian and other traffic that can be quite distracting, so having a solid system for remembering to record arrivals is crucial. What worked for me was keeping my log book within my field of view by the car’s arm rest.

      Beyond that, I recommend setting up the “randomization” ahead of time so that you don’t end up unconsciously biasing your test. I created the log pages before doing the first test and used a random number generator to fill in the “which app am I using for this test” column.

      Like

  19. I find Waze too dangerous to use. It will frequently choose unprotected left turns across six lanes of traffic. I end up taking alternative routes so I don’t get killed, which take longer. I did actually get in an accident once while using waze route so my feelings were justified. Did you stick to the recommended routes in all cases?

    Like

    1. I did stick to the recommended route in all cases and that was super painful in cases when I knew with certainty that I could outperform the suggested route. This was most common within a mile of my home where I know the stop light patterns well, but also around the Bay Bridge in San Francisco which has a couple sneaky on ramps when driving westbound.

      Like

  20. Nice. Makes me want to do my own logging. Most of my driving is in mixed suburban/rural driving in the US East Coast. Google Maps is the best on accurate estimated time.

    In rural settings I usually have a choice between a direct route on local roads or very roundabout route on freeways. I much prefer the latter; selecting “Avoid highways/Avoid tolls” gets me extremely good routes via Google. Google is also best in navigating weird traffic patterns; I’ve gone on unnecessary detours when I’ve been in a wrong lane for a crucial turnoff because Waze didn’t give that level of detail coming into an exchange or intersection.

    And although I only use it once a year for Bike to Work Day, Google Map’s biking directions are inspired and have taken me down direct and underused roads I never knew existed.

    I mostly use Waze in known commutes with driving directions turned off. It’s traffic warnings, police alerts, and speed limit bleeps are helpful there. But yes, time estimates are terrible. I usually forget Apple Maps exists. I don’t find it gives me interesting local directions.

    Like

  21. Thanks for the report. Waze has one thing the other two don’t have – the reporting ability. Speed traps, accidents, potholes etc…. make it an awesome choice. Now it has voice control and that is so useful as well. Over here in Hawaii – Waze is mostly accurate. You are right, some of them tend to under-estimate travel times.

    Like

  22. Thank you for this. I’ve wondered about some of these same things. I’ve a theory that the apps don’t deal well with left turn waits. Do they track enough cars for long enough or account for a car going nearly zero for several minutes (if you wait through two or three cycles).

    Another observation: I have some preferred routes (mostly for scenic or impatience at long signals) and notice that when I don’t follow the recommended route, the estimated time doesn’t change (maybe the time difference is too small) or it will up the estimate but then later end up at the original estimate. Of course, the inverse is true. No real data.

    I liked Waze because it took me through neighborhoods or streets I might never take. But I pretty much stopped using Waze because of the multi-lane unprotected left turns.

    The other thing I wonder about (and another posted mentioned this) is whether they are looking at current conditions or predicting based on norms for the time when you’ll actually be passing through that region. Another problem is car pool lanes. We sometimes use them, but ASFAIK none of the apps I use ask whether I’ll be using one.

    Lot’s of challenges to getting the predictions right. I hope more people take up this challenge.

    Like

      1. Yes there is. Navigation>Reduce difficult intersections. It’s probably been more than a year since I looked at the options. Good to know. I’ve now set it.

        Like

  23. One problem with Waze in the bay area is that as more people use it the ‘creative factor’ can really backfire. I frequently have to cross the San Mateo bridge and in the all too common event of a wreck Waze is very quick to route you onto side streets and occasionally to the Dumbarton, but in my experience, generally, by the time you make those maneuvers so has ‘everyone’ else and the alternate routes are a quagmire. Unless the wreck is really bad, or you are already close to an alternate route, I find it faster to head toward the wreck, since once it is cleared the traffic will pick up beyond the wreck. Apple and Google maps generally make the right call (stay the course), even if they don’t anticipate the delays very well. Also, in my usage anyway, none of the maps do very well at anticipating when the carpool lanes switch on and off, which also makes a big difference in congestion levels.

    Like

  24. I only use waze because it has speed limit alerts. None of the other maps compared has this (in Australia), If google maps had speed limit alerts (in Australia) I would switch over to google maps.

    Like

  25. As a slightly different data point, I drive the same stretch (Los Angeles area, some highway some local) almost every day, but at different hours.
    For a long time I would look at the Apple Maps suggestion, but then bypass it to take local over the regions where I knew (and could see that) the highway was congested at that time.

    But then, out of curiosity I decided to follow Apple’s suggestion regardless, and my conclusion is that the Apple suggestions ARE faster than taking local. The comments re Waze are true — depending on how autonomous your car is, driving local may be much more pleasant than stop and go traffic, but let’s not confuse that with getting to your destination faster.

    IMHO (at least for me) the best overall solution is to keep Apple maps alive, but when highway looks bad, to veer off into constantly different parts of local, and then rely on Maps once you are on local roads. That way, sure, the trip is a little longer, but you get to learn more and more of your environment, and to see things that you never even knew existed. In my experience Apple is as accurate at time estimation for local as for highway, so they’re presumably somehow factoring in traffic light times. (Note that I drive at the speed limit, not faster, not slower.)

    One other general point is that when I’m a passenger in the car I frequently compare Apple to Google on my phone and (at least in LA) the two seem to me much of a muchness. Both occasionally get it wrong in claiming (or not pointing out) congestion ahead, and there’s no obvious pattern to why, or which is more accurate.

    Finally enough with the generic “Yeah, but Google is better outside the US”. Outside the US it’s a very big world. I’ve no doubt there are some places where Google’s maps are more accurate but over a range of places I’ve been to (eg Myanmar, Thailand, Taiwan, China, South Africa) I’ve not seen this alleged Google superiority. Now when I travel all I have used maps for is seeing what’s where — not navigation or traffic — so those might be superior, but if that’s the claim, be specific.

    The one place Google is undeniably superior to Apple is search (inside and outside the US). Beyond that, Google’s ability to download some map areas for local storage is EXTREMELY useful for traveling (and it’s pathetic that Apple still offers nothing similar) marred (as usual) by a ghastly user experience in trying to select the downloaded regions, to see what downloaded regions exist, and in the handling of the downloaded region updates.

    But Apple seems to do a somewhat better job (neither are especially good…) at handling the complications of map display for countries with non-Roman alphabets. Apple tends to (sometimes with some work…) allow you to see, eg, the name of a street in both Roman and Thai script, whereas with Google I pretty much always find I only have access to the local script version, not a Romanized version.

    Like

  26. Yes, but did Apple Maps actually find the correct destination? On my last 3 or 4 attempts here in the UK, Apple Maps had a 100% record of picking completely the wrong destination, often hundreds of miles from where I was trying to get to! Needless to say I’ve not pursued my testing with it to see whether when it does find the right destination whether it can pick a good route to it or not.

    Like

    1. I was pleasantly surprised with Apple Maps finding the correct destination. Based on early Apple Maps horror stories I expected the occasional snafu but only ran into one issue where Apple Maps didn’t know about a construction zone that prevented left turns.

      Like

  27. Nice study, although since it is just one driver I think its definitely skewed to your driving style (as you pointed out in the limitations)– probably google maps is most suited/similar to your individual driving behaviours/style. I live in the SF Bay Area as well like you, but I consistently beat the prediction Waze makes by 3-4 minutes usually (although dependent on the length of the trip– the longer the trip, the more I typically beat it by) which is far different from the +11% prediction error you get on Waze.
    I’d be interested in joining your experiment and collecting data in the same format to see how it performs 🙂
    I also agree with the last commenter that the unprotected multi-lane left turns that Waze suggests sometimes are one of the worst issues with it. My favourite feature in the Google Maps app is the lane-display data and it baffles me why they don’t have it in Waze given its their parent company and Waze in general loves data + functionality over any sort of aesthetic design.

    Like

  28. Interesting observations. In Malaysia, where I live, Apple maps rely on Foursquare data in city areas, and there it’s quite usable, especially as I’ve noticed that Apple maps often give “smarter” routes. But Apple maps totally lose out to google maps outside of the city areas here, perhaps due to its reliance of Foursquare data too. Google maps simple have most of the locations in its map, even in remote locations, whereas outside of the city Apple maps only show roads and nothing else.

    Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s