last meter perspective


Last Meter science #1. Package boxes are vending machines. Is that what we really want at the last meter?

Stockholm + London

John Manoochehri

February 5, 2019

Last Meter science #1. Package boxes are vending machines. Is that what we really want at the last meter?

February 5, 2019

The science of individual package box solutions is pretty new, but there's lots of maths and business modelling that can be applied to demonstrate that these kinds of boxes are only likely to be a small part of the solution, and should be added later - in adapted form, as part an ecosystem of solutions - but not now in their current form.

Here's a small sample of the analysis work that leads to BASE2 Last Meter®, developed over many years of design work (including e.g. design of Kiruna's new city with BIG), policy work with the UN and WWF, and teaching/research at KTH among other places. We have some maths and data on this, but we sort of apply it to our own model so we won't release that.


CHEAP STUFF The majority of packages are low value. Logistics is already losing a huge amount of money. Boxes make deliveries more expensive, and technically vastly more complex. Right now, the majority of retail goods aren't locked in boxes in the stores. They are insured, and the average theft rate is about 1.5% (less for cheaper items). Financial engineering alone says it will be cheaper to find insurable non-box solutions for a portion of lower-value goods. But design analysis helps too: surely a private building, locked, is more secure than an open retail store?!

BIG STUFF Across the totality of packages available for delivery, the majority do not fit into the majority of individual boxes. The portion of IKEA's total inventory that fit into the average individual Instabox box is less than 1%. LuxerOne, the leading box maker in the US, is adding larger and larger shared 'overflow closets' to their box solutions to handle the surplus goods that don't fit in small individual boxes. If a) boxes are the wrong size, and b) consumer will accept a single shared box for all their goods together - why have small individual boxes at all?

EXPENSIVE STUFF For the really 'expensive' stuff, maybe only as much as 150 USD, there's likely enough margin, and certainly enough retailer incentive, to do an actual home delivery - i.e. really track people's movements and get it to them with a personal touch, and suck down the costs of re-deliveries. Like new-version Urb-It or the long-gone personal-delivery service MyWays (by DHL).

PERISHABLE STUFF The majority of home deliveries at this time, and for the likely foreseeable future are groceries and hot food. Apart from the fact that most weekly grocery shops won't fit in an individual box, perishable goods and individual boxes don't mix. Just one box with food left by someone, will kill the experience for nearby users with a functioning nose.

So what does this leave left, to justify all the expenditure on boxes that is being requested?

About 20% by value and maybe 40% max by volume of all possible deliveries. Is this enough to justify boxes? Possibly. But when you look at the composition of this remaining volume, the future becomes clearer, and it's hard to see today's package boxes there very prominently.

RETURNS The dynamics of last meter logistics are badly distorted by the presence in the delivery mix of cheap items and perishable items, which have the unique logistical footprint of only ever going in one direction (outbound to consumer). For approx 100% of goods that are worth putting in boxes, these goods can and should be enabled to go in both directions by last meter logistics optimization. Clothes can be tried on, computers can be fixed, seasonal goods can be returned after rental, waste can be handled to enable the circular economy (ping Ragn-Sells and TipTapp).

This means that any solution for these goods should facilitate returns as easily as it facilitates delivery. The idea that the current generation of boxes does this is simply incorrect. The consumer needs to do lots of fiddly things to make returns happen, even if the boxes allow it.

But even if box tech is upgraded to make it easier in theory to do returns, in practice, returns from consumers need to be optimized for inventory management. So the goods need to come back in good condition, in a standard format, without requiring complex actions by the consumer (special packing for example) or warehousing staff.

This means that, most likely, sector-specific solutions will emerge, particularly when the business models that require returns (e.g. clothing rental) start to get more serious. Secure padded boxes are needed for computers. Hanging racks are needed for clothes. Shoe-size boxes are needed for ... shoes. Medsafe systems are needed for medical equipment. Firesafe boxes are needed for waste. Reverse logistics solutions needs to require ± zero effort from the consumer and logistics and warehouse staff, be basically zero-priced, and bring back the inventory in very good and restockable condition (or, if waste, handle it correctly).

Pick your sector and you'll immediately think of a cool logistics solution to send it both ways at the last meter. One day it will happen (you should make it happen!). Standard, fixed, individual boxes are useless for all of this.

I could say with the above analysis, we already know that individual boxes will not work for the majority of retail goods on the basis of sheer physical dynamics. But there's so much more to say.


MERCHANDISING + TOUCH Consider that up to 20% of the total cost of goods sold in some categories is spent on advertising and merchandising, including in-store presentation - this does not even include money spent to create deeper experiential touch and consumer value (e.g. Apple Store Genius Bar). Then ask anyone spending that money if a viable replacement for that is this: the brownbox-steelbox-blackbox of cardboard packaging, steel delivery boxes, and the 'blackbox' of not knowing the experience your consumer is getting at product handoff.

Anyone that cares about their product despises the poor quality of last meter handoff enabled by the brownbox-steelbox-blackbox ruckus - but they perhaps aren't saying it yet because they are panicking about declining in-store revenues and will try just about anything to survive. Give them any cost-effective solution that brings back the power of merchandising, brand profile and consumer touch and quality - and they will dump boxes like a stone. For this reason alone - as any student of retail and merchandising and branding (and anyone who loves products, even people who love packaging! (like me)) will tell you - brownbox-steelbox-blackbox is not the future.

BIG MATH Individual standardized box solutions are proposed as helpful on the basis of, essentially, a holy miracle of matching package volumes and sizes to available boxes with minimal failure rates. Modelling this into the future, with big but not very complicated math, suggests the holy miracle, even if it works right now in some cases, is very temporary.

There's a lot of quantifiable variables, and almost all of them go in the wrong direction. The main variables are usage rate, stash time, mismatch rate, failure rate, and value rate.

Usage rate is the overarching metric, indicating how many packages over a fixed unit of time are successfully served by the boxes; stash time is the average time a package stays in a box; mismatch rate is the number of times a given box location has no solution for a package of a certain size (or other feature required, e.g. refrigeration); failure rate is the average amount of technical failures of the box system; and value rate is the total value of packages served over a fixed unit of time. (These things have different names in different research; as I say last meter science is only just starting out! There are also lots of other variables, but some of them I don't want to share.)

When you crunch these variables under pretty simple assumptions about growth of the number and diversity of packages, it becomes evident that the balance of usage rate and value rate is just headed downwards, as costs increase. That's not good.

SMART MONEY As retail logistics grows from its current tiny percentage of total retail to a larger percentage, the people spending the money on this - retailers - will start to be smarter about their money, rather than just panicking about the decline of stores. They will observe the mismatch of declining usage rate/value rate vs costs, sure, but they will also ask are they getting the right prices in the first place.

If logistics and box actors are using fixed pricing for box use - no matter what is in them, whether it be 5 dollar gewgaw from Wish or a 500 dollar necklace - this is a huge penalty on higher-value retailers. How so a penalty and not a reward? It's a penalty because proximity to the consumer should be an extremely valuable prize. And the cheap value folks are paying far too little for the privilege, even if the boxes work.

The main reason the use-value metrics go down for boxes, over time, is that the boxes are clogged up with shite that even consumers don't care that much about. Smart money will always want a market solution, and retailers are smart. Box allocation is not a market-priced solution, it's something optimized for the logistics sector which is not yet pricing correctly by box value. This pricing flaw, in addition to the general depressing awfulness of brownbox-steelbox-blackbox, is why retailers will move to sector-specific solutions of which they control basically everything: quality, returns/inventory management, and pricing.

And so, I could say with the above analysis, the donkey is dead and I am merely flogging it. But there's so much more to say. And this time - it's really serious!


Aside from the physics and money of boxes, in the bigger picture, there are just some big immovable facts at play that make it very hard to understand how boxes are supposed to solve all packages from now into the future.

TECHNICAL LOCK-IN In case anyone hadn't noticed, we live in a technological age. There are few certainties about technical pathways, but two that are pretty solid are:

  • things change fast so be ready for the future today
  • standards rule, and open standards rule supreme.

Today's box solutions are disastrous on these grounds. All the software, hardware, parts, skillsets for manufacture, installation and maintenance, are custom. This leads to two kinds of lock-in: obsolescence, and risk. Already boxes are obsolete because they don't have NFC capability, which is the obvious (only) future for consumer locks at large scale (unlock with mobile phone). And right now, if boxes software is hacked, or there's some problem in the supply chain, it won't be a very quick or cheap fix, since it's all custom stuff.

TECHNICAL COMPLEXITY There's another truth about the technological age, but not about technology itself: it's that people want things to be simpler not more complex, as technology itself gets more sophisticated. This is one reason why NFC is going to be (already is) essential in any access system for boxes, doors, whatever - consumers can just unlock with their phone - but it's also a general point about technical systems. If you don't need a technical solution, don't use one. In particular, any technical system, particularly one involving security, that is not securely but universally linked to others, will die in the cold on its own. Parakey, Latch, Fidesmo, even Mastercard and others are on the right track here. It's not clear the package boxes are.

Consumers are going tire of fiddling with box interfaces and technical systems very quickly - as unfair as that may seem since it's them who is (in theory) being helped. And retailers will work this out soon enough, and will seek simpler and more universal solutions and partners.

SECURITY + CONTROL Sadly, this is not just a technological age, it's a socially complex age. Trust is going up in some ways (AirBnB), but declining in others (terrorism) - but in both cases overall, the need for security and for responsible people and institutions to keep control - in any situation - is only going up.

Individual boxes inside residences or offices buildings, or other public locations, have an extremely problematic security profile. If someone - heaven forbid this, but it's entirely conceivable - wanted to cause harm to an individual or group, or the public, it would be simple to send a package to box solution in a building or a public location. Less dramatically, boxes are also maybe an interesting 'solution' for drug dealers. Fire and chemical safety is also worth considering - most box makers are offering somewhat secure solutions, but it's not clear that the fire and chemical rating of all the contents of the boxes is matched to the boxes themselves.

So, unless the logistics companies are screening every package as they do in airports, the responsibility for safety in this instance falls on the shoulders of the owners of the locations on which the boxes are placed. This is a story yet to be written. But it's very hard to see that the location hosts of box solutions will not want to have some control over the contents of boxes that are situated in the spaces they are responsible for - if there is no other guarantee the contents are safe. Right now, boxes are not close to offering that.


Boxes are expensive. Amazon Hub is priced at 10-20,000 USD per unit! So who pays and who benefits? Right now, it seems like the owners of the space in which the boxes are positioned are required to pay for them. What benefit do they get? It's not exactly clear, but let's assume that what they 'get' is massively improved flow and quality of packages into their building.

Technically that would be impressive, and it would be convenient for users. (Although see above for why analysis and math suggest it's not the future.) But if boxes succeed, that means that Amazon (or in other contexts the retailers and box companies), for example, has basically parked part of a retail store in someone's building or train station or store - and is making tons of money from it.

Real estate is not always the first to be smart, but they always always catch up eventually. And so - in the very best, magic-thinking, case - where massive increase in retail logistics is facilitated by boxes, real estate owners are extremely unlikely to either pay for boxes, or to allow the use of those boxes in their spaces to go unmonetized to their benefit.

The future in this regard has actually already been written. What is a vending machine other than an 'individual package box solution' - that is a) paid for by the retailers, and b) pays the real estate owner where the vending machine is stationed? Likewise: Coca Cola pays for/subsidizes fridges in retail stores to facilitate sales. Why should retailers not offer a similar approach for box solutions at scale in real estate? Or, negatively: why should boxes be installed, even if they are paid for by retailers, if they make no money for real estate owners?

Box companies are thus probably selling to the wrong people. Retailers get value from them, and so should pay the cost, and share the benefits. Real estate owners will soon enough work out that the Amazon's package boxes are just "Amazon vending machines" ... which real estate not only does not get money from, it is supposed to pay for!

But then again - is this really the best use of the space anyway? If space use is priced correctly, maybe packages aren't at all what will be supported at the last meter. After all, of all the things coming to offices and homes - if things (and experiences!) with the highest value don't even use boxes, because they always go straight to the end user - maybe something should be done with the space available - and priced accordingly - that supports them instead? If money talks, who knows, maybe boxes will walk.


So this is a short summary of a fraction of last meter science. Most of the good stuff is not even here, because we use it to guide the Last Meter® solution and story which we are gradually sharing, and still developing. But this is a hopefully a useful input to discussion and decision-making for anyone thinking about individual box solutions.

I am duty-bound as an entrepreneur to support other innovative solutions, and so I welcome everything that helps in some way. And of course anything is possible. Betamax was beaten by VHS.

But as an architect, scientist, and (hopefully) businessman, I also have to share that I truly believe that the world of box solutions is accelerating into a giant wall of financial physics, money money money, and big facts. Netflix killed Blockbuster, even BluRay couldn't stop the march of streaming.

Ultimately, individual box solutions are helpful right now. But they are helpful in the way that sandwiches and blankets were helpful when the Titanic hit an iceberg. They aren't lifeboats, and the ship of retail logistics is going to sink faster; retail panic will accelerate when it's clear that boxes can only handle a tiny fraction of flow, for a lower quality experience than stores, for an increased price-to-user.

Individual package box solutions are really an output of the current logistics world, of undifferentiated package flow that is massively accelerating. To solve this, you need to think holistically, and work step-by-step. That's the work of BASE2 and that's the plan with Last Meter®. We don't have all the solutions today - in fact we explicitly say we cannot help solve all categories of last meter optimization in one step - but at least we aren't driving into a giant wall.

To learn more - particularly if you are a real estate owner or manager of any sort, or a retailer, logistics , or on-demand service company, or even (although now you hate me) a box company - check out And stay tuned for the launch of, where stories of #lastmetercool #lastmetercrazy will be shared.

The future, to leave this on a practical and positive note, will probably have the following kinds of last meter logistical solutions, differentiated by type:

1 Low-security, low-value, high-volume packages: Groceries, hot food and low-value/insurable stuff will be left in low-security droppoints inside locked buildings, maybe with camera monitoring. BASE2 Last Meter® calls these droppoints. This will facilitate massive logistics efficiency and clear out the whole system a bit - financial engineering via insurance will help a lot, and retailer/logistics-subsidized building-access upgrades.

2 Mid-value packages: Undifferentiated packages with medium-value content will be left in locked locations inside buildings. These will be non-individual shared boxes (closets) or rooms, maybe with camera or more sophisticated monitoring. BASE2 Last Meter® calls these storepoints.

3 Sector-specific solutions: Different categories of retail will collaborate as sub-sectors, to invest in technical solutions that help them. Groceries, drinks, tech, hot food, clothes, shoes, furniture, medicine, and basically everything else - all these want and need specialized solutions. BASE2 Last Meter® calls these (although they basically don't exist except for the ones BASE2 and few other retailers/architects have designed) also storepoints - just better ones. It's actually perfectly possible some of the sector-specific solutions will look like - or be - individual boxes! But they will be designed with a very precise correlation to the size, value, and security specifications of the contents - and the suppliers will like have a more precise relationship, via a platform like Last Meter®, with both real estate owners and retailers, to help to correlate their technical solution to those spatial, value, and technical requirements.

4 Large items: Very big things will need large spaces close to parking and elevators. These are also storepoints - just big ones.

5 In-home: For the most expensive stuff, which might not even be that expensive, the solution will be nothing other than a person representing the brand you are buying from (not a logistics dude or dudette or random on-demand person who doesn't care and doesn't know anything about it) bringing you stuff when you are home with a big ol' smile, and with lots of care and knowledge about you getting what you want (e.g. will help you set it up).

There are lots of other details and features to Last Meter®, and most of the last meter solutions that will be needed aren't built yet. Sorry.

What is exciting though is that the box companies, if they are open to it, will be able to build this ecosystem of (unbuilt) solutions rapidly - once they perceive the risks in the current path, and detect the signals behind the panic of the retailers - and start tuning out the urgency of 'solving' for vast undifferentiated package flow.

And finally, as an architect, having just helped to design the world's first Last Meter® enabled new-build property - to be announced soon - here's a whole other reason - beyond the above - why property developers are not so keen to include boxes in their fresh new designs.

Package boxes are fugly. Are we going to design the last meter of future properties around steel boxes? Nah.

We'll do it right from first principles - with math, engineering, digital-age economics, and data-driven design - and it'll feel like packages and all incoming services are just integrated seamlessly into buildings, like internet is integrated to computers, and apps are integrated to phones.

Welcome to Last Meter®. This not a rehearsal. The future is here.


#lastmeter science is an occasional briefing series from BASE2.