I was toying with doing a whole series on poorly designed Metro maps, but then in the comment thread you guys did a pretty good job of covering that subject.

However, I agree with a point made in the thread that the big missed opportunity in Metro’s system map is the use of color.

In the map, color is primarily used to distinguish who the service provider of each route is.  The problem here is that no one cares who is providing the service.  At most, it might allow you to compute the fare, but anyone able to do that with just a map also knows that you can get that information solely from route numbers.

A better idea can be found on the Utah Transit Authority’s otherwise user-hostile website.  On their system map, green lines represent 15-minute service, blue ones 30 minute headways, and red lines are UTA’s express buses.

The point of a system map is to quickly understand the options for getting around. Most users need the map when they’re in unfamiliar territory, which will generally not be a peak-hour trip.  The current map is nearly useless for this rider, as the routes he actually can take are hopelessly cluttered by ones that won’t run for hours.

For those really interested in the subject, Adam has uploaded an 18-page survey of Western European transit maps, in which we learn that Metro has chosen a “classic style” map, while I’m advocating for a “Scandinavian style” system map.  The report goes on from there, and people in related fields would profit from reading it.

Oran has also played around with presenting frequency information on a system map.  As always, Human Transit has map-related insights here and here[UPDATE: Try this version of Oran’s map instead.]

54 Replies to “Unclear Metro Maps (II)”

  1. I’ve seen that link to Oran’s map before. Is there any way to actually view the map? All I see is “Currently unavailable.”

    1. Sorry I removed the links a while ago but I put one of them back. It’s not really a system map, rather a map of frequent services. I wasn’t too satisfied with it so I did another map, using what I learned in that paper linked above, just for transit in the city of Seattle. This comes closer to likely modifications Metro can do to their own map.

      See (DRAFT & unfinished) Seattle Transit map

      1. Looks fairly paired down from the regional one. But has a lot of good information. Have you thought about neighbourhood maps should look like?

      2. Very good, but I am a little perplexed about why the streetcar gets its own category. I agree with Jarret Walker of Human Transit in that if the streetcar provides frequent service, as defined by you, it should be grouped with the frequent service buses. If it does not, it should be grayed out. Right now, there is no indication of the level of service the streetcar provides.

      3. Because I wanted people to know what to look for. This is a streetcar, not a bus, so look for the tracks, wire and train. I haven’t put the station locations in yet.

        Actually, there is an indication of service for the streetcar. Match the purple line color and streetcar symbol to the same one in the frequency chart below. I incorrectly stated that the streetcar has frequent service to 10 pm. That’s only true for Friday and Saturday. Sunday service ends at 7 pm and other days end at 9 pm.

      4. True, I’ve been bit by that on Sunday evening! I love the map by the way, though personally I’d just drop the OWL routes.

  2. I’m quite interested in Oran’s map but I don’t see any links to the actual map anywhere…

    1. I was able to see the map, but I’m not sure I agree with Oran’s notion of French-style different colors for every frequent route, with everything else reduced to gray lines. I think Martin’s right, a Scandinavian-style map would work well for Seattle. On 65th, the 71 =/= the 64 or 76 =/= the 83.

      1. I could see a frequent route map as a complement to a system map, though. Metro’s system map includes a LIST of frequent routes rather than representing them on a map. Ayi-yi-yi-yi.

  3. I’m not sure re-arranging the chairs on the boat will make a big difference. Drilling deeper into the structural problems of Metro’s routing choices will yield better results, then tackle the colors of the lines issue latter.
    Just look at all the comments about ‘un-readability’ of just one map placed on a printed timetable – and it seemed everyone had their favorite. I called them route “mutations”. Now, imagine trying to consolidate hundreds of those on one map. You see the problem.
    Another, is that with so many routes sharing so few major corridors, the lines into downtown get really packed, like crossing the bridges, or I-5. A trunk and feeder operation would simplify map making, but I’m not convinced it makes riders lives any better or significantly increases ridership, so maybe the current map is about the best we can do.

  4. You don’t have to look far to find a good model. Portland’s TriMet has clear route maps, a very clear system map that indicates route frequency, an incredible trip planner with a map feature, mobile apps that include real-time arrivals and trip planners, etc.

    In Seattle, I only use One Bus Away and Google Transit. When I’m in Portland, I only use the TriMet site and their mobile apps and the experience is much better, even though I’m much less familiar with that city.


      1. Trimet has special branding and maps for what they call “frequent service” routes. It make it instantly evident which routes are all day and which ones are lower quality. While there are something like 14 routes on the list almost 60% of trips are made on these routes. Metro could use this strategy well. Given a million or two you could rebrand all frequent routes. This leverages the service metro already provides, making it easier to riders and especially non riders to better understand and rely on these core routes. Most importantly though it gives Metro something to show off.


        As the FTA administrator said last week. Paint is cheap. Call it a “special bus” and people will ride it.

      2. “almost 60% of trips are made on these routes [those on the frequent-service map]”

        That could be a way to increase public support for a hub-and-spoke system or frequent-bus network. If you highlight the routes that are frequent and advertise their ridership percentage (which is already high: 71/72/73, 358, 43, 44, 48, 49), that could help generate public support to improve them to RapidRide status and add similar service in other corridors.

      3. The 49 will be partly obviated by U-Link and the First Hill Streetcar (possibly resulting in a reduction in service levels as its purpose will be mainly reduced to 10th Ave service), I suspect the 70-series will be moved to terminate at UW station with the 70 probably moving to all-day 7-day service (and possibly eventually obviated by an extended SLUT to the U-District), and the 358 probably is already marked for RapidRide status. The point about the 43, 44, and 48 is heard, though; the 44 may be partly obviated by Link but that’s VERY long-term.

  5. So bottom line…we have all these good ideas for intelligent, clear maps and we know how to make great maps (what with all the amazing talent here just on our blog, not to mention the whole Puget Sound region), however Metro’s issue is that they don’t have money to make these or to create them? I mean if it’s money, can’t people here create these maps for Metro for a much cheaper price? Am I missing something…why can’t WE create these maps for Metro?

    1. This is what I’ve been saying. Ask the GIS students at the UW Geography Department!

      1. You need someone that can update the maps every 3 months. I dont know that is viable for students.

        Metro has money problems but this really is cheap compared providing the actual service. Better maps and good branding for “frequent service” are in my opinion a cheap win. Metro gets to show off without spending a single dollar more on operations.

  6. I’m looking at Portland’s map right now and I don’t find it particularly clear. For example, not all the streets a bus route travels on are listed. What streets does the 51 Vista go down? All you see are squigglies. Also, if I remember correctly, route maps are not included with the schedules in the schedule book. What’s up with that?

    Seattle has many more bus routes than Portland so a Seattle map will be busier than a Portland map. I suggest that the map scale be made significantly larger. Also I would consider issuing separate transit maps for different parts of the city, e.g. the “300 Series North County”, the “100 Series South County”, the “200 Series East County”, the “City of Seattle” one, and I would think a separate one just for the downtown area.

  7. They are awful maps. I just popped over to London for a couple of days and I should have taken picks of that, but I was really excited about these pedestrian/cycle maps. I first saw one near Marble Arch, but I took this one when I was in Lambeth. London does really well:
    At bus stops, London provides a Spider Map of local routes, an areawide map with some details of the spider map but also important destinations in the area, and a region-wide map with all key London services and so on. Metro has always been really poor with signage. Seattle gets the best of it and I suppose the transit centres. Outside of that, one is s*** out of luck. Where London has lower ridership with only the poles, TfL still lists the buses and have triangular wedges with maps that help to make sense to the commuter. The most recent attempt to update signage for users by Metro is completely token and ineffective. With in-house GIS brilliance and a bit of creative research, it shouldn’t be that difficult to put maps together for online and assemble stop information boards in a rational way. To be fair, I shouldn’t throw stones beacause Corás Iompair Éireann is so much worse with signage for its bus and rail services in Ireland. As far as European systems for user guidance, they are a joke.

    1. I made a spider map for Ballard.

      Metro focused only on bus stop signage. That’s only a part of the information system. They should’ve taken a look at the entire system of how they present information to the user, which includes schedules, maps and route branding.

  8. The link to the 18 page survey doesn’t seem to be working. Any other references for this?

  9. I don’t think the maps are unclear. I don’t hear that complaint anywhere except on this blog. But then this blog has a “greener grass” tendency.

    No, the map is just fine the way it is.

    1. That’s not the case at all, plenty of people complain about it. There’s nothing wrong striving for a better service. I was pissed last time I was in Seattle. I was in the Pioneer Square area looking for the correct stop back to Tacoma and by the time I’d gone to 5 stops, I watched the bus depart and had to wait another half hour. There wasn’t a map to indicate the clusters. This happens to people all the time who AREN’T regular riders like tourists, people not commuting but going outside of their normal routes, and rare users. But sure, we’re all whinging pricks who KNOW it’s greener on the otherside. Never had a mess-up once in London and I don’t even live there.

      1. You are blaming the map for your lack of planning. You could have figured out what bus stop to wait at for your return trip from your home computer, even before you left, or you could have asked the bus driver during your trip in, or you could have called ST, etc. So please don’t blame the map for your lack of preparation.

      2. Online resources are useless if you don’t have access to them, which concerns tourists and people who don’t have internet access, can’t use or don’t know how to use a computer.

        Not everyone has a cell phone. Calling in takes too much time, especially during busy periods, where you have to wait just to talk with someone when you could spend less than a minute reading a well designed map and figure it out.

        Bus drivers are not encyclopedias (maybe some are), they don’t know everything about the system, usually only routes that they have driven. Asking them distracts them from operating the bus safely and asking at stops wastes everyone else’s time.

        All those resources are out of the way. They are not presented to the user directly. Having people call or use the web (if they can) instead is not an excuse for not providing good information right at the bus stop, visible to all.

        The map is broken. Try using it to figure out a trip without the above resources.

      3. Oh come on that’s just stupid. I figured it would be the same place I got off. Apparently, it doesn’t work like that. How was I supposed to know that? Oh right, I was too stupid to use the internet or be a worldly regular rider. My fault, my fault, it couldn’t possibly be because I thought the system in the area might be logical just like at Tacoma station: get on and off at the SAME place. Sorry, some of are complete retards. You know what, feck off. This, is exactly why proper maps in the area should be put up to designate stop groupings or a more rational regional mapping system or even local routes! Other places get this right, Seattle doesn’t. I will not take responsibility for missing my bus because the system is supposed to be friendly for pedestrians who may not have time or access before their departure. What you are saying is that only regular riders matter. The rest of us can just feck off. Meanwhile, redoing the whole thing would encourage people to venture on other routes or utilise transit. But sure, you’ve made it clear in previous threads that you don’t care about rider usability, so why would you here?

    2. I think Martin’s point above wasn’t that “the maps are unclear” so much as they don’t contain all the useful information they could for potential riders.

  10. The Metro map is like psychedelic or kaleidescopic art. Only the nerdy can understand it. Others are blown away and limit their transit use because of its frickin incomprehensibility. Seattle’s car-related business interests appreciate Metro for making transit harder to use.

  11. I find the Metro maps of the downtown area to be much better, although not perfect. Passengers frequently ask which routes to take when I’m downtown and since I haven’t driven many Seattle routes, I’m frequently as clueless as they are. I refer people to the downtown maps when I know they are going somewhere in that area. The problem comes when somebody asks for a point outside of that map at which point we’re in full agreement that the big map is pretty difficult to use.

    1. Is anyone on this blog work in the King County Metro Transit department and if not, are any of them aware of this blog or our concerns (which mirror the community)?

      1. I know of a few that read it, but sometimes turning the ship requires the order of the captain, not just some crew members. Are you listening Kevin?

    1. Other places (LA, for example) differentiate by transit agency because the numbering systems are incompatible, with many systems with different numbers in the same place. It’s important to distinguish between, say, Big Blue Bus 4 and Metro 4. The only comparable problem here is near UW-Bothell and Aurora Village where CT local routes cross the county line, or where a couple of Metro routes stray into Snohomish County, or where some Pierce Transit routes stray into Federal Way, and all the CT routes are in the 100s (in the north county) and I don’t think the PT routes are incompatible either. I think it’s a) a sign of continued parochialism among agencies and b) a reflection of Sound Transit’s hope of being the backbone to the other agencies’ systems. It works for ST’s Regional Transit Map Book (though it’s near useless with regards to Metro), but it’s not necessary for service just within King County.

      Also, something I forgot to mention in the other thread. Pick up one of the night-owl timetables. Look at the downtown map (not online for some reason – I can only vouch for the 80-series, not the 280). Tell me you can decipher it with a straight face.

      1. Ah, that’s because the snow routes rendered the downtown map redundant anyway. But for example, what’s the exact route the 85 takes downtown? I’ll give you a hint: it’s NOT straight up and down 1st Ave in both directions.

  12. After seeing Oran’s frequent service map of Seattle I realized why Metro creates its system map as it does. Its for the same reason airline system maps (found at the back of every in-flight magazine) show individual curved lines for every possible city pair, including those of their partner airlines, even if many of those routes include “connections”: the map gives the impression that this airline can take you everywhere in the world. It is so illegible that you can’t really connect the curved lines and see where they fly to.

    The frequent service map for Metro points out just how little frequent service is provided. Metro’s system map creates the illusion that there are hundreds of bus routes covering every nook and cranny of the county. But its not informative enough to plan an actual trip.

    1. I’ve been generous with the representation, some routes don’t have frequent service in the evening or on weekends.

      Outside Seattle, it’s more pathetic.

  13. I’ve read most of the linked article on map choices, and equipped with that information, I think Metro would get more use out of the complexity-management maps. Metro has more than a couple dozen routes, making a whole-system map (even of the Scandinavian type) hopelessly dense.

    Stop-specific route maps would be a great start: show a map of everywhere I can go from this stop without connecting, and mark on each of those routes some set of major connection points. In theory these could be algorithmically generated for any stop in the city, and posted as desired.

    Beyond that, as Stephen Fesler points out, a spider/octopus map can cover many use cases. I can keep a map of the buses from my neighborhood, and use it when making simple trips. For more complex trips, a spider map gives a broader picture than the map for just one stop.

    Thinking further, a project to produce spider maps for each of the neighborhoods would align well with a more general project of walkability neighborhood maps.

    1. Metro could also use the same map as now, but reform the colors. Right now there is one line that could potentially represent numerous agencies, at least for bus lines (this is most apparent near the county lines where CT and PT local routes leak into King County). You can add more colors without making the map hopelessly dense if you use one line to represent all of them.

  14. Interesting metrics would be:

    1. What number/percentage of riders take the same route, and use the same stops everyday.

    1a. What percentage of these people are taking a non-optimal route, because they don’t have the information/map/web access to inform them.

    2. What number/percentage of riders are taking a unique trip…like a job interview…where they had to plan, or research the ride which they may never, or rarely take again.

    2a. What means did they use to plan: map, computer, phone, friend.

  15. Some comments/suggestions about Oran’s map.

    1) I believe that the 5 continues frequent service all the way to Northgate way, not just greenwood AVE
    2) I believe that the 16, 12, and 3/4 in the Central District are also frequent service corridors.
    3) The 30/31 run frequent service in Walingford
    4) It would be good to note the existence of the 30 on Westlake Ave
    5) It would be good to add the 510/511, the 545/255, and the 550/554 to this map.
    6) You should have a different color for peak only and night only routes
    7) It would help to have a “almost frequent service” color code for routes like the 15/18, and 60.

    There is still some more work that needs to be done on this map, but I hope we can see this map, and one more major routes map for whole region, rather than all those agency specific maps we see right now, as soon as possible. It would really help with the whole “transit is confusing” reason for so many not to ride.

    1. Great comments/suggestions, thanks.

      I used the fabulous OneBusAway to check individual stop schedules. All of the corridors on the map are based off Metro’s Frequency of Service chart.

      1) No, frequent service ends at Greenwood & 85th St.
      2) 16 doesn’t go to Central District. Every other 12 trip ends at 15th & Madison, leaving 30 minute service to Interlaken. 3/4 frequent service corridor ends at 23rd where the routes go their separate ways.
      3) Cool, I didn’t know that. I checked the schedules and on average they are about 15 min apart from 6 am to 6 pm on weekdays and from 9 am to 6 pm on Saturday. The afternoon times on the 30 are kind of screwy and Saturday has a 13/17 min spacing but I’ll accept it.
      4) The map is unfinished and I forgot.
      5) I am going to put suburban routes in but it’s not a priority.
      6) Yes, I plan on doing that. That’s why they are separate categories in the legend.
      7) You mean ones that run every 20 minutes? I’ll think about it.

      1. The 16 is another “almost frequent service” corridor – every 20 minutes in the midday, according to Metro’s larger PDF (linked to from Oran’s link above). It joins the 1, 15, 18, and 60, and that’s it. (Metro’s PDF claims the 73 as well, but I don’t know where that comes from.) I’m working on my own Scandinavian-style system map with thick yellow lines representing 15-minute service and thin yellow lines representing Metro’s standard on its Downtown map – 20-minute service Mon-Sat, 30 minutes evenings and Sundays – only to find that criteria is merely an out to include the 16, as it, the 15, and the 18 are the only qualifying routes in the whole system. (I might still keep it, if only because it fixes the cheat on Oran’s map where the 15 and 18 are shown as still providing frequent service up to Market St so they can connect to the 44.)

        If anyone wants me to keep working on it, I will, though it’ll take me a while and won’t be as good as anything Oran would put together, unless I switched to my color-route-numbers idea above (then I’d just edit Metro’s existing map directly, though that poses a copyright risk). (My limited application options are part of the problem – I’m using DeLorme Street Atlas USA 2007 or 2008, solely because Google Maps doesn’t allow me to change the print area, even between portrait or landscape, so I’m stuck with the size of the screen or 8 1/2 x 11, but the tradeoff is dealing with Street Atlas’ often obsolete info [it still doesn’t have Shoreline as its own city!] or inexplicably missing streets, and I can’t keep track of what routes lines represent by naming them or Street Atlas will render them on the map in a method that’s not helpful to the final product.)

        (If I do stick with the Scandinavian map, right now I’m using blue for routes that go downtown – or whose timetables show what routes you can connect to to go downtown, as an arbitrary divider between routes that provide actual service versus routes like the 75 that are more point-to-point – purple for routes that don’t, light green for all-day express routes, dark green for peak-only express routes, red-orange for the SLUT, teal for Link, and black with railroad tie hash marks for Sounder, with red being held in reserve for RapidRide, at which point the color for the Seattle Streetcar will probably become more orangey. Problem: I only have a peak-only color for express routes, but what about peak-only routes like the 35 that don’t have an express component, or that only connect to an express route, like the 110? What about express routes with the same number as local ones, like the 7X? What about school routes, owl routes, DART routes?)

      1. There are two early morning trips that start at 85th and aren’t marked as express. Metro should just make them express buses so its map doesn’t show some local trips as taking the 85th loop.

Comments are closed.