A few of Amazon's shuttle routes
A few of Amazon’s shuttle routes

Grist has been running a fantastic series on Seattle. You’ve probably seen one or two of the pieces show up in your social feed, but if not, the whole series is worth your time.

One piece in particular wonders whether Amazon’s SLU campus is the future of the office park. I was nodding vigorously in agreement as I read it (having had similar thoughts in the past), but there was one passage that stuck out:

There are no company buses shuttling workers into and out of the neighborhood. But its main employers (Amazon, Fred Hutchinson Cancer Research Center, Group Health, and a University of Washington biotech and research hub) all pay extra money to have the streetcar line down Westlake Avenue run every 10 minutes during rush hour instead of every 15.

While it is indeed great that SLU employers fund the streetcar, Amazon (and UW) do make use of shuttle vans.  Some of these shuttles simply help employees get from building to building in a sprawling campus (a necessity for large employers spread out over many buildings), while others are more of a last-mile commuter affair, taking employees to campus from Colman Dock or King Street Station. You can view the routes here (or download the app!).

The last-mile nature of Amazon’s commuter shuttles, though, suggests a stopgap measure. If and when our various transit agencies provide adequate service into SLU (the proposed Westlake transit lanes and the Center City Connector are a start), Amazon could presumably cut back.

51 Replies to “Amazon’s Shuttle Fleet”

  1. When can we retire the fraudulent boosterist meme that any statistically-significant percentage of SLU employees ever take the streetcar?

    1. If they’re paying for it, clearly the companies believe otherwise (though whether or not that belief is true is another question)

      1. Oh, silly me. I thought that Grist self-defined as an intellectually rigorous publication whose writers seek to effect positive change through uncompromised application of their journalism degrees. I had no idea that they too were regurgitaters of mindless and easily-disproven corporate groupthink.

      2. Reading the article, d.p., you are correct, Grist is definitely taking that position.

        However I don’t think anything quoted on this blog post says anything other than what the employers paying for the streetcar believe.

        (FWIW I am seriously skeptical that the streetcar has had any real effect–it’s basically empty most of the time and is often slower than walking)

    2. What’s statistically significant, according to you? According to Wiki, in 2013 there was 2,600 rides/weekday. That’s something like ~5% of all SLU employees per day to the best of my estimation.

      1. 1,300 round trips x 20 = 26,000.

        Reports suggest that Amazon alone may be approaching double that number of employees in SLU.

        So no, not even close to 5%.

      2. Surely you don’t mean 52,000 Amazon employees in Seattle today?

        One estimate had Amazon’s headcount at 24k at the end of 2014. There simply isn’t enough office space for significantly more workers at the moment.

        By 2019, Amazon projects to have 10MM square feet of offices according to statements the company made last year. That would be enough for 71k workers, based on standard size/employee ratios, although no guarantee that many would actually be present in the offices.

      3. Good to know where Amazon’s headcount stands.

        I was unable to find an up-to-date precise accounting of aggregate SLU jobs, but I think we can all agree that Amazon is hardly the only employer — or even the only quadruple-digit employer — in the hypothetical SLUT catchment area. The hilariously out-of-date PSRC figure suggested that SLU had about 20,000 jobs before Amazon arrived, and it isn’t like Amazon is the sole entity to have seen significant expansion or to have appeared from whole cloth in the last half decade.

        TL;DR: 1,300 round trips ≠ 5% of anything.

    3. The companies also pay for extra peak runs on the 8, and would presumably do so for other routes if the situation warranted. When the C is extended to SLU and Roosevelt BRT is built, they might be top candidates too. But I imagine nobody sees the current 8 as anything more than a stopgap, so supporting extra runs on it does not mean they’d be opposed to replacing it with a better line.

    4. Reading the article, I don’t think it says anything of the sort. It only says that companies pay to increase the frequency of the streetcar, not that anyone rides it. I think the most misleading part of the article is this:

      The streetcar line opened in 2007, and it was a game-changer: Once it started running, the price per square foot for office space along the line doubled.

      That is bullshit. It was simply a matter of time before that area became attractive to businesses. Allen knew it, and the Commons plan was simply an attempt at a cool park along with a jump start in business. But once that died, Allen wanted to speed things up. Biotech — once assumed to be the big driver in the area — had slowed (there are a couple biotech businesses, but that is it). But then it was his old business — software — that came to the rescue. If you look at employment predictions, software jobs have been at the top for the last twenty years (along with medical jobs). In other words, people predicted that there would be huge numbers of software jobs in the future. But the dot com years were terrible for software, even for companies that had little to do with dot com businesses (companies like IBM and Microsoft got hammered). So when the industry recovered, it recovered in huge numbers. At the same time, traffic in the area was getting worse. This meant that downtown real estate (and that includes South Lake Union) became a lot more valuable (since it is easier to get there using public transit). It may be hard to get to the last mile in South Lake Union, but at least the first ten miles are relatively easy. As it is, the last mile is not that hard because it is usually less than a mile (and relatively flat). That is why there are areas that are clearly not influenced by the streetcar (like the Cascade neighborhood) where property values have soared. Likewise the area much closer to the heart of downtown (e. g. 7th and Blanchard) an area that used to be dominated by old motels, has changed dramatically. Of course it has. Seattle is just going through a huge economic boom, fueled by software, and anything close to downtown is bound to go in value. The streetcar had nothing to do with it (nor did it spur development in Fremont, or Ballard, or Lake City, or …).

    5. A “statistically-significant” percentage, however we choose to define that, will take the streetcar once the center city connector is completed (assuming the actual project is completed as currently planned, never a wise assumption in Seattle) and turns what is essentially the world’s longest-running pilot project into actual transportation. The time improvements to the existing line paired with a seven-minute trip through downtown and pioneer square to connect with Sounder, the ID station, and First Hill are why none of the anti-streetcar folks here can debunk SDOT’s basic premise: $30 million in local dollars for roughly 20,000 riders a day—all while connecting and making capital improvements to our existing lines—is a legit investment.

      1. Whatever.

        1st Ave is still a roundabout way from anywhere to anywhere, no matter how many lanes you take.

        Add in the slowness with which the streetcar takes its multiple turns, the fundamental impossibility of giving it priority across 2nd, 3rd, and 5th, and the high likelihood of box-blocking at Stewart and at Jackson, and SDOT’s time estimation is worthless.

        The streetcar won’t be “statistically significant” in our lifetimes.

      2. >> $30 million in local dollars for roughly 20,000 riders a day—all while connecting and making capital improvements to our existing lines—is a legit investment.

        You could do the same with a bus. For less cost and just as much speed — maybe more, because you could go up hills or avoid obstacles. Meanwhile, our buses carry more than these streetcars. So sorry if I’m not impressed with a “legit investment” because most of it was a waste. We could have done the same thing for a lot cheaper with buses.

  2. Improving connections via Westlake into SLU obviously makes sense from a network prescriptive, but why should Metro or any agency take over Amazon’s last-mile shuttle routes? The agency’s purpose is to deal with the transportation needs of an entire region, not just focus on a single business’s last mine needs already in close proximity to nearly every transit service available in the region. Looking at that map provided, Amazon employees can take existing 2-seat rides or walk to cover that entire area. Amazon Campus to Coleman Dock? 16/66 to SLUT. King Street? SLUT to the bus tunnel. Airport? SLU to Link. UW? Microsoft? SLUT to 545. Unless, of course, the argument here is more about one-seat rides.

    1. You ask a good question, so a little background check of South Lake Union is a good place to start.
      https://en.wikipedia.org/wiki/South_Lake_Union,_Seattle
      This area has been disconnected from the CBD and efforts by Paul Allen and others to annex this area into natural growth patterns of the CBD have failed over the years. I worked on the Seattle Commons effort in the 90’s to no avail, and now that Amazon and others have taken the bull by the horns, it’s growing as it has been destined to do since the cows came home.
      Unfortunately, transit has been left behind, catering to the old school political interests of the Downtown Association and political elite with a token, albeit dysfunctional, streetcar plopped down the middle of a street to mascarae as public transit.
      To have abandoned this area from a HCT station of its own, in favor of mushing north to Alaska on our ‘spineless transit fantasy journey’ is a public disgrace. It didn’t have to work out this way!
      Private shuttles are what’s left in the void of crap-planning for the future.

      1. Whose idea was the streetcar? Hint: the same person you said has been trying “to annex this area into natural growth patterns of the CBD”. Furthermore, “Transit has been left behind.” By the same person? The SLU business community hasn’t asked for anything beyond the streetcar. They doubtless support the Westlake and Eastlake “rapid streetcar” proposals, but they haven’t been prominent in promoting them. I guess that means they want a car-centric neighborhood with a token streetcar.

    2. “why should Metro or any agency take over Amazon’s last-mile shuttle routes?”

      It’s a balance. Can Metro provide citywide routes that meet Amazon’s needs without unduly subsidizing Amazon at the expense of the rest of the network? The best network, as we know, is a frequent grid, but Seattle has geographic barriers which are especially acute around SLU, and low-density areas which detract from a strict grid’s ridership. Seattle should provide infrastructure allowing reasonably fast routes on the grid streets of Westlake, Denny Way, 45th, 40th, etc. And it should have land use that fills in the low-density areas between urban village nodes to facilitate more frequent transit. Both of those will be long-term challenges. Beyond that, what should be reasonable transit between the ferry terminal and Amazon, or Ballard or Lake City or Greenlake or Bellevue and Amazon? A fast route to the ferry terminal or a one-seat ride to Lake City may be beyond a citywide network’s priorities, so that’s where shuttles would be inevitable.

      1. A grid might be fine for trips along the grid; but if people are coming off the ferry at the dock, they already had at least one transfer at the other end of the ferry route. One transfer at each end of the ferry and another one for the grid might become onerous. Major transfer points like ferry docks, heavily used commuter rail stations and widely spaced light rail stops usually contribute more passengers than an ordinary bus stop (otherwise they wouldn’t be major); so it might at least be of somewhat higher priority than everywhere-to-everywhere routes.

    3. Metro shouldn’t take over Amazon shuttle routes, but I doubt anyone takes the 16 or 66 and transfers to the streetcar to go to SLU from Colman Dock. Between the transfer penalty and the couple blocks you have to walk to make the transfer you’re better off walking to 3rd Ave from Colman Dock or walking from the 16/66 to your destination.

  3. Does anyone else see the idiocy of having spent tens of billions of dollars over the last 3 decades on transit, and yet, even in the most dense, and supposedly transit-rich area of the state, private companies still need to run their own shuttles? Are we simply pouring my income tax dollars down an empty sieve, know as Seattle…

    1. What income tax dollars? Washington State has no income tax and what Federal dollars are spent on transit in this state are but a rounding error in the yearly budget of DoD programs like the F-35.

      If you don’t want to pay for transit buy everything taxable outside the Metro and ST taxing districts or online.

    2. Well, at least WSDOT is building a recovery tunnel to ‘mine’ some of those dollars being poured down the rat hole. The should break into the glory hole of lost taxes in a few years.

    3. There’s no income tax in Washington State, broski.

      And SLU is not ‘transit rich’ for most definitions of the word. The buses that go there are limited to the 7*s, the 8 and the 40. There’s the streetcar… and that’s pretty much it.

      An area like, say, 3rd and University are transit rich, and you don’t see the businesses down there running private shuttles.

    4. Bailo is lucky that he lives outside the Prop 1 area so his sales tax and car tabs didn’t go up for transit.

    5. Yes, John, many people here complain about this almost daily. They feel like the transit agencies have spent way too much money on suburban projects (like light rail to the airport) and way too little in the city. As many have pointed out, this is no good for anyone. The suburban commuter is forced to take these little shuttles once they get into the city. Spending money in the city benefits more people. Those in the suburbs are better off taking express buses or commuter rail, which then connect into a subway (or light rail or whatever you want to call it) in the city, which makes frequent stops along the way, making these buses superfluous. So, basically, we should have the much of the following by now (since it should have been built in this order):

      1) UW to Downtown light rail (with stops along the way, like First Hill). This would also connect with 520.
      2) UW to Ballard light rail.
      3) Metro 8 subway (which would also connect with express buses from I-90).
      4) East Link
      5) South Link to a similar (new) station somewhere south of SeaTac.
      6) North Link to Mountlake Terrace (where a huge transit center would interact with buses from the north).

      Unfortunately, many politicians have pushed light rail as the answer to every problem (like suburban connectivity) when it clearly isn’t. That is basically why we have the system we have now.

      1. Many politicians think suburban connectivity is the only problem. What suburbanites experience is congestion on the freeways, so they think that’s where transit is the most valuable.

      2. I’d say North Link would be a higher priority than going south and the segment to Northgate would be just below UW-Ballard (though to be fair when Northgate was planned Ballard was still a pretty sleepy part of Seattle)

        Similarly It wasn’t apparent the 8 corridor would be such a win in the mid to late 90’s.

        Also given the suck location of the MT transit center going as far as Lynwood makes some sense (difficult access from anywhere but I-5)

      3. I just hope that articles like this won’t be used to justify Ballard getting a streetcar down Leary in ST 3.

      4. @Mike — The suburbs aren’t getting much for their money. Again, once they get here, they can’t get around. Buses are annoyingly slow from the suburbs, but still pretty fast compared to driving. Yet still they drive. They drive to First Hill (because the transfer is annoying). They drive to South Lake Union (because the transfer is annoying). They drive to Fremont or Ballard (because .. well, you get the idea). All these people drive in from the suburbs because, while the first part of their drive is terrible (Lynnwood or Shoreline or Federal Way or Burien or Bellevue to Seattle) the last part (Seattle to the other neighborhood) is so much faster than the transfer that it makes it worth it.

        For the folks coming from the south, Link speeds up their commute into the city by five minutes over a bus. But it still take another fifteen (or more) to get to that other spot. When all is added up, driving makes sense, and so they drive. More suburban rail isn’t going to change the equation. The only thing that will change things is if we have better transit in the city.

        @Chris — Wrong on all points. Greater Ballard has always had significant numbers of people — it has always made sense for light rail (given the geography). The 8 cuts through lower Queen Anne and the Central Area — areas that have been some of the most densely populated areas in Seattle for a very long time. The Mountlake Terrace spot has very easy access from the freeway, and that is all that is needed for a sparsely populated suburb like Lynnwood (or any other suburb to the north or south). There just isn’t a high concentration of residents anywhere to expect big numbers of walk up riders (Lake City contains a highest concentration of people than any census block in Snohomish County). In other words, everyone will arrive by bus anyway, so going on HOV lanes to the bus stop makes a lot of sense.

      5. “The suburbs aren’t getting much for their money.”

        I was talking generally, not any specific proposal. A lot of people base their votes on general feelings, not on the details of a proposal.

        “Again, once they get here, they can’t get around.”

        Most suburbanites don’t go to Seattle. They work in another suburb or their own city. When they do go to Seattle it’s a few times a year for ballgames or Bumbershoot or other activities. There’s a large number of people who commute to Seattle, but that’s most visible because they’re all going to one place, not because they’re the majority of suburbanites.

    6. Seattle transit tunnel declared complete after $455 million
      Cost for Central Link and Tacoma Link: $1.7 billion
      Sounder cost what? $300 million in capital upgrades and equipment?

      So what were the other “tens of billions” spend on?

      1. Almost 2 billion for U-link, but given the ridership it will have, its worth every penny.

        It would have been worth a lot more if we hadn’t lost the First Hill stop though.

  4. The reason for the private shuttles comes down to three things:
    1) Avoid the time sink by serving every intermediate bus stop between SLU and King St. Station/Colman Dock
    2) Provide Amazon employees the premium experience of a one-seat ride and a guaranteed seat (vs. likely having to stand on a public bus)
    3) As long as Sounder and ferries continue to exist, a lot of Amazon employees are going to be making connections at King St. Station and Colman Dock, even if bus routes get rejiggered in the future to add direct service to SLU from more neighborhoods.

    On the other hand, one big asset that the public buses have that Amazon buses don’t is dedicated transit pathways through downtown. If the streetcar gets its own lane on Westlake and extends down 1st Ave., again in its own lane, regular transit will likely actually be faster than the Amazon shuttles. Even today, regular transit might win, simply because the shear number of buses traversing downtown means almost zero wait time.

    1. I continue to doubt that the Connector streetcar will be a fast way from anywhere to anywhere, even with dedicated lanes. It simply suffers too many zig-zags, crosses too many Avenues where it will never see priority, and is guaranteed to see its box routinely blocked both where it enters and leaves 1st.

      People shouldn’t get their hopes up that it offers any kind of panacea.

      1. While the streetcar may still be slower than a bike in getting from Colman Dock to Amazon, during rush hour, it is doubtful that an Amazon shuttle would be able to do any better – at that time, the dedicated right-of-way the streetcar will get is worth more than the time penalty of serving the intermediate stops, especially with off-board fare payment.

  5. These shuttles, operated by SP+, are a plague upon the streets of SLU. They constantly run stop signs, refuse to signal when turning or changing lanes, drive like madmen and worst of all, park in bike and car lanes to pick up or drop off.

    There’s one building, the 1918 8th Ave building, where they will stop in the right traffic lane and let people out onto the bike lane on a regular basis.

    Complaints fall on deaf ears and it’ll be a cold day in hell before the City does anything. Now I understand the complaints of San Francisco people with the private Google shuttles.

    And yet, I have had ZERO complaints with the Microsoft Connect shuttles. They are always courteous, follow traffic laws and best of all, never block traffic. Maybe the difference is that the Microsoft logo is emblazoned all over the Connect shuttles, where as the SP+ are glorified, unmarked kidnapper vans.

    1. Most of the Microsoft Shuttle routes are out on the Eastside. The buildings there mostly have places for the shuttles to stop without blocking traffic.

      As you say having the Microsoft logo on the side of the Microsoft shuttles means the powers that be are concerned about the image the vans present to the public.

      I suspect the difference in behavior between the Microsoft and Amazon shuttle drivers is reflective of the corporate mindset somewhat as well.

      1. That was EXACTLY my thought… Sounds rather reflective of the corporate cultures and their employees.

      2. Most Connector stops are definitely in the city, rather than the eastside. The buses (unlike google’s in SF) don’t stop at any public bus stops with the exception of OTC which is the only public bus stop in the entire state a private bus operator can pick up/drop off.

  6. Shuttles really aren’t necessary to get around the campus. How lazy are people? The walking time from Terry/Mercer to 8th/Stewart is only 16 minutes according to Google Maps. That’s basically the furthest distance between most of the offices in the core campus areas. Unless these shuttles run every 5 minutes it would be faster to walk.

    Yes, now there are some far-removed buildings, but as in any job, most workers don’t usually have to meet with people outside their departments very often.

    I get the ferry dock / Sounder connections, and to connect any far-flung buildings, but walking is a better way to get around SLU-Denny Triangle.

    1. The shuttle is theoretically faster than walking when going between the Denny Triangle and Mercer buildings, but in practice I never found them to be any benefit.

      There’s still a shuttle that goes between the SLU campus and a building down by the Lowe’s on Rainier Ave, but it only runs during commute hours – the one time I had a meeting down there I ended up needing to bum a ride off a coworker.

      Amazon’s shuttle system made a lot more sense as a daytime shuttle when they were spread between Union Station, Beacon Hill, and Rainier Ave.

    2. Amazon is bringing in a lot of people from parts of the country where people scoff at walking more than a quarter mile and are deathly afraid of rain. The shuttles are catering to them. Some eventually figure out that downtown Seattle and SLU are walkable (in theory; there’s a lot of sidewalk closures in the past two years and many more to come) and decide to brave the elements.

    3. So, first, yes people are that lazy, or want to work while on it or it’s raining. Second, not everyone can walk all that far (or if they can, it might eat up their energy.) I in practice walk between buildings unless it’s pouring. But I’m quite able and like walking.

    4. Ah, yes. Because I am capable of walking from 8th and Stewart to Terry and Republican in 15 minutes, everybody else can too, regardless of their disablilites, their schedule, or whether or not they’re carrying things from place to place. So I will pass judgement against everyone who, for reasons of their own, rides a shuttle instead of walking.

  7. Lord, what a ridiculous way to start a comment list for a posting on the need for reserved transit lanes. Signal pre-empt too, especially ability to extend green signal so a bus doesn’t have to make two long stops every time it hits a red light right in front of a zone right across the street.

    Which transit will need by the mile CBD-wide unless we want to do a transit history thing and bring back the legendary “Wall of Buses” when DSTT joint-operations end.

    For South Lake Union right now, a short conversation between KC Metro, Mayor Murray, Jeff Bezos, and several hundred Amazon-vetted passengers should get a block or so reserved lane and serious signal action at east end of the Waterfront. Where traffic holds line fifteen minutes late at PM rush.

    An extra transit only signal northbound at Whole Foods could get a lot of rubber tires out of the way so the streetcar can enter and clear the zone without having to hold for a half dozen lights.

    Anybody who wants to can knock themselves out paving over the tracks. Will mean more employment taking skill saws to it when Ballard line starts.

    Mark Dublin

  8. I just wanted to call out that the streetcar is not actually slower than walking for everyone. Imagine you don’t walk all that fast, or you have kids with you, etc. When I worked in a building near Mercer a few years ago, I actually did comparisons. I’d walk one day from the Westlake terminus, varying my behavior (speed, obedience to walk signs). I’d note when the trolley would arrive at Westlake (my theoretic wait time), then see on the other end if it beat me or was close to caught up. Depending on how fast I walked and how many lights I jaywalked (the number of crosswalks on Westlake is absurd), waiting anywhere from 2-8 minutes would be worth it. Now it’s been a while and traffic has changed but I repeatedly read people saying you can walk faster which is only true for some. Then add in some people not wanting to walk in the rain, carrying bags, have children or just preferring reading to walking and it’s clearly of benefit to many.

      1. That wasn’t intended as a throat-jump.

        It’s just that if the streetcar were such a useful or preferential tool for “many”, then it would have “many” riding on it.

        We have the sums. It simply does not.

        Listen, I have no particular desire to walk the ugly southern stretch of Westlake either, braving the construction gauntlet and/or weather. Therefore, if I am headed to SLU, and not coming from a direction that make the 40 (or even the Dexter buses) an infinitely easier and more direct path to do so, I am as likely as anyone to take an “opportunistic” hop onto a streetcar that happens to be coming.

        That is how the streetcar appears to attain most of its incredibly modest ridership. The slower-moving parties you describe are still most likely to drive, unimpressed by the connectivity the streetcar has offered. And among those that do choose transit, there’s a numerically-supported chance that they’re already choosing one of SLU’s bus routes (on the basis that they actually go places).

        Your speed and hypothetical-utility comparisons are not wrong, if your starting point is precisely Westlake and Stewart at a given point in time. But we now have years of actual ridership data, and the “many” simply don’t find it useful enough to show up.

      2. Given the comments about the SLUT one must wonder if and when the Broadway streetcar starts running, will we be making the same comments about low ridership? Given the SDOT makeup of Broadway with a dedicated bike lane, one also ask how long after the Streetcar starts running will there be a need to remove something from Broadway for there to be any traffic flow. It could be moving the bike lanes to 12th.

        It’s really too bad that we lost the only streetcar that was really fun and that was the one on the waterfront with the Australian cars. That line was also a tourist attraction too!

Comments are closed.