Amazon CFO ‘doesn’t see a way to do same-day delivery’ | The Verge:

When asked directly about the possibility of offering same-day delivery, Szkutak hedged a bit, saying that “we’re always trying to get closer to customers… That’s not new, it’s been something we’ve been doing for years.” However, he then freely offered that “we don’t see a way to do same-day delivery on a broad scale economically.” That’s as close to a flat-out “no” on same-day delivery as we’ve ever seen from Amazon, so we’re not going to be holding our breath for that to change anytime soon

My reaction is different. I think there’s a lot of wiggle room in that statement. Define “broad scale”.

I agree with him, it’s probably not economic to do same-day delivery in, say, Billings Montana. What if Amazon decides to take, for instance, the top ten metro areas and build a service for them? That’s not “broad scale” in that geographically, it’s only covering a small percentage of the US — but it touches a non-trivial number of potential users. Could you build a service for, say, New York City, Los Angeles, San Francisco and Silicon Valley, Seattle, Chicago, Atlanta — pick some big, fairly dense populations and think it through.

Anyone who remembers Webvan here in the Bay Area knows it’s possible, if you don’t mess up your logistics chain as totally as Webvan did. Consider this….

There have been reports on and off for about a year about a huge distribution center being built in Patterson California for some mystery company. Everyone “knew” that company was amazon, but it’s only recently that’s been admitted formally. So Amazon is going to have one million square feet of automated warehouse coming to the greater Bay Area. it’s due to open soon after the sales tax change occurs; Amazon couldn’t put a business presence inside the state borders without triggering that. Now that they’ve agreed to do it anyway, they don’t have to stick their warehouse out in Nevada.

(interesting side question: is Amazon taking advantage of ‘losing’ the sales tax war by moving distribution to Patterson? Or did Amazon strategically give up that fight so they could start implementing a business model that takes advantage of big distribution centers within the state of California? interesting question…)

A quick look at driving times from Patterson: San Francisco, San Jose and Sacramento are all almost exactly 90 minutes away. Fresno is two hours. So in theory, if you wanted to order something that lived in that warehouse, and you were willing to pay for expedited delivery, it wouldn’t be too hard for Amazon to promise same day delivery. Order by 10AM, on your porch by 6PM. What price? Good question. But where I doubt you’d pay $20 to get toothpaste delivered this evening, what about a flat screen TV?

Webvan did this a decade ago; Webvan died because they did it BADLY, not because it was a bad model. What Webvan needed was solid distribution and warehouse logistics. Guess what Amazon has?

So think about a service with significant limitations: I’d target the top 10 population markets in the country, say, and limit delivery to 90 minutes driving from the distribution center. you limit the items to a subset of inventory that is (a) in reasonably high demand, (b) in that warehouse, and [c] shippable without a lot of hassle. These same-day priority items get picked and shipped out from the main distribution center 3-4 times a day on big trucks to smaller redistribution centers. You could cover the majority of the Bay Area eligible for these deliveries with three: South San Francisco, Milpitas or Fremont, and near Sacramento (say, Elk Grove). Big trucks pile in, and incoming orders get pulled off and stuffed onto delivery vans that run out and do a round of deliveries, then circle back and grab the next load. These models are also proven, it’s a combination of what UPS does today with what Pizza Hut does. So none of this is inventing new models, it’s scale and execution.

Take it a step further: you now have buildings handling sub-regions where you are packing up for final delivery. What if you shifted some of your inventory here? What if you stocked your 5,000 most in demand (for same day) items in those regional hubs? Those hubs could take those items and have them on your porch in four hours. For the right price, of course. This is the Pizza Hut model, but for WIFI routers and hard drives instead of pepperoni.

So without a lot of innovation (instead, it’s all about capital, ability to scale and knowing how to execute logistics — all of which Amazon has), Amazon could put a key subset of product items four hours away from a lot of people, and a much larger inventory within same-say delivery capability.

they may only be able to offer it to 30% of the population (which I think excuses them from “broad scale”), but that’s still a lot of people, and a lot of revenue opportunity. And over time you expand to new metro areas and refine the model and the product mix.

Will they do this? I dunno. But it sure seems like the kind of thing Amazon CAN do, and that almost nobody but Amazon could do. And the business and logistics models it’s based on are all known and proven (including how NOT to do it like Webvan did it).

Could you build a price point that’d make this pay? I think so, but I haven’t done the math. But it seems to me there’d be enough demand that $20-25 for a small box delivery, $45 for a large box and a discount for 2nd-etc boxes in the same delivery could be made to work — and that enough people would go for it to pay for it.

And then, of course, I think to myself: what if one of the ways Amazon builds out this infrastructure is to buy UPS? If they did that, a big chunk of what they would need is already in place….

So when I hear Amazon saying “no, we don’t see why we’d do that”, I just think about how many times Steve Jobs said something like that. And with Amazon, like Apple, I always hear “… until we’re ready to.” at the end of those sentences…