All posts made in May. 2013:

20:23, Wednesday 15 May., 2013

I've been doing some competitive landscape analysis around connected products/Internet of Things platforms -- I'll write up my thoughts soon. During research I touched on Bluetooth 4, which seems like it could be the connective tissue of a peripheral ecosystem around smartphones just as USB was for peripherals around the PC.

And in this section, I hadn't included Apple's MFi Program in the list (MFi is hardware and certification for iPod, iPhone and iPad.) Greg asked me why. Well, I said, they don't do enough UX integration, and besides, I don't want to give them any ideas. If they did what I think they should do, they would totally own connected products.

But hell! The Big 3 are full of the smartest technologists on the planet!

It's not for lack of ideas that they aren't doing this.

So here's how Apple, or Amazon, or Google could totally become the platform for the future world of connected products, and - with a connected products platform of my own - the thought that one of them might make a move like this is what keeps me up at night.


Starting point: With the Kindle, Amazon have an amazing chip that has global connectivity via 3G. They also have a billing model where the content provider pays for delivery (currently $0.15/MB for deliveries to the US, which explains why you don't get many graphics-heavy books on the Kindle). This kind of billing infrastructure is hard.

What happens: Amazon apply their genius for service oriented architecture (SOA) to Kindle's Whispernet functionality, take advantage of their economies of scale, and provide wireless chips that any developer can use. Just as they SOA'd their storage requirements into S3, and their server farms into EC2 - now both services that are the tarmac of the modern web - they couple this SOA'd hardware connectivity with Amazon Web Services, and create the perfect platform for connected products. Of course Amazon also own an identity system with associated credit cards/payments platform. Plus they really get APIs.

Amazon would own connected products. You wouldn't build on anything else.


Starting point: The emerging smartphone peripheral ecosystem (appcessories and whatnot) is built around Bluetooth 4, the low power wireless standard that Apple have been including in their products since 2011.

What happens: Right now dealing with appcessories on the iPhone sucks (claiming and syncing), so Apple add some minor UX support, adding hardware products to the homescreen with a parallel to Newstand called Nightstand -- a virtual table for physical things. You associate each product with your Apple ID. Then, to solve the problem that connected products need to talk to the web without a smartphone present, they activate the Bluetooth 4 already present in the Apple TV (and maybe add one to the Airport Express), and make it so that any product that can connect via your smartphone can also connect via any Apple TV you've signed in on using the same Apple ID. For bonus points, iCloud is used for the messaging layer, so any data sent via the Apple TV also shows up on your iPhone. Of course Apple owns an identity system with associated credit cards, fully capable of micro-payments and subscriptions.

Apple would own connected products. You wouldn't build on anything else.


Starting point: Android. Motorola.

What happens: Google take cheap cellphone guts - the peace dividend of the smartphone war -and use Motorola to release a development platform that runs Android, rebooting the Android @Home program that was launched back in 2011 with smartphone-controlled lightbulbs. In this new 2013 world of Arduino and Raspberry Pi, hardware is way more accepted... but loads of people already know how to develop for Android. So developers flock to this new platform. You're not locked into Google's hardware, because Android hardware is commoditised down to the CPU, unlike similar offerings from Amazon or Amazon. The UX is provided by Android apps, of course. Google Cloud Messaging is used to link the connected hardware to regular ol' websites that developers build themselves. Websites are easy, and Google trusts the web. The platform is a great combination of open and familiar. Google also owns an identity system, and a payments platform.

(A note: I don't think Google could pull off the Apple model of a peripheral ecosystem built around Bluetooth 4. Google doesn't have enough non-smartphone presence in the home, and Android fragmentation would be a major problem -- especially Samsung's ownership of the front room via the Smart TV platform, which would put the two companies at odds.)

Google would own connected products. You wouldn't build on anything else.

Who I'd back

I wouldn't back any of 'em.

It's true, if any of the Big 3 made a move like this, you'd be dumb to use anything else for your Kickstarter project or new hardware company. It would be great. So many common problems would be solved.

But I'd be sad. We'd be stuck with a platform that met our imaginations only of today. It wouldn't evolve; big companies are too slow.

We're only going to discover the weird and wonderful opportunities of connected products once we've rolled our sleeves up and got our hands dirty. How are connected products going to change our homes, our offices, our cities, our social lives? Who knows. It'll take years to find out. And at that point, maybe we can have a dominant platform. That'll be fine. Until then there's BERG Cloud and a dozen others to help figure it out. There will be more. Let a thousand flowers bloom!


A weblog by Matt Webb.

Korbo, Lorbo, Jeetbo.

You're probably looking for my email address or the syndication feed.

You can get updates to this blog on Twitter: follow @intrcnnctd.

I'm @genmon on Twitter. Also find me on Instagram and LinkedIn.

14:25, Tuesday 14 May.

How To Price Your Hardware Product, Marc Barros:

The mistake most hardware startups make is they don't charge enough because they don't think of the problems they will encounter at scale. They don't calculate the real cost to deliver their product to a customer's door, they leave no margin to sell through retail down the road when opportunities arise, and they can't easily raise the price after it has been set.

Covers some good points that you need to take into account, beyond your profit margin:

All points that are easy to forget when you're looking at the bill of materials for whatever the core component is.

Here's one of Barros' examples using top-down pricing: $200 retail means you get $101.80 from your customer. A product cost of $58.10 means you have a margin of $43.70, or 42.9%. He recommends shooting for a margin of 50%. All reasonable, sensible, I like his summary for this: Don't be afraid to charge more. Long term, your loyal customers will thank you for staying in business. You're not thanking your customers in any way if your low margins mean you have to skimp on customer service, or developing improvements to the product they've invested in.

To my mind, there are two disruptions that make this take on pricing difficult.


I think about Kickstarter hardware projects in two categories. There are those made for love not money. (And that's cool -- hardware products, like any creative act, can be made for 1,000 true fans with the potential - but not requirement - to break through into the mainstream. I love it.) Then there are those where Kickstarter is about getting mindshare, learnings, and the infrastructure to build the products that come after this one -- there's no profit requirement. That's cool too: In an established company, products sit underwater for a long time before they break even.

These projects are low margin, funded by love and future expectations, and - because Kickstarter is also a great distribution platform - they don't need to build in retail margin. Consequently the prices are lower than equivalent non-Kickstarter projects.


I use Amazon as a proxy for the shifting sands of new business models. The Kindle is sold at cost, or below: It's all touchscreen, PCBs, and battery. Where do Amazon make their money? Well, nowhere yet... they're a notoriously low margin, long term view company. But once they make $3/month additional sales, the Kindle Fire moves into profit. But think about this... if the $159 was sold with the same markup suggested by Barros, we'd see a RRP of $547. Insane.

This isn't new. Cellphones have been subsidised by carriers for years, their high up-front offset against monthly bills. Car financing is common. DFS functions more like a credit company then a sofa store.

But it's becoming more common in the hardware world as subscription relationships become more accepted -- and more necessary. When products connect to the cloud, the cost structure changes once again. On the one hand, there are ongoing network costs which have to be paid by someone. You can do that with a cut of transactions on the platform, by absorbing the network cost upfront in the RRP, or with user-pays subscription.

We're finding product categories dominated by one business model or another. It's hard to enter a subscription-dominated category with a straight-forward retail model. Your product will look too expensive.

It's not as easy as it once was.

Enough product companies are operating at zero margin, or on some alternate business model, that pricing hardware is no longer as simple as making sure you have the right margin.

15:19, Thursday 9 May.

Open in my browser right now:

  1. Planar Choerographies. We're all familiar with stable orbits in a two body system: it's how the earth goes round the sun. The earth describes a big circle, the sun a little one, and both are centred on their mutual centre of gravity. It turns out there are stable orbits for n-bodies too, and they're lovely. I wonder what it would be like to live on a planet in a seven on a butterfly solar system.

  2. Wired interview with Bill Gates. Saved 5 million lives, and he's funny? Dammit.

  3. The pitch deck Buffer used to raise $500,000. Great pitch deck. A simple story, well told.

  4. Chris Dixon on hardware startups. A big factor in why hardware is possible now? The peace dividend of the smartphone war. (Chris Anderson.) Chris Dixon lists a few points to keep in mind: Manufacturing (no Amazon Web Services for production); defensibility (no network effects); planning (it's not agile); B2C vs B2B (attention vs margins).

    I'm gonna add four other points of differentiation from software. One is distribution, both attention and fixing shipping things. Is hard. Incumbents win. Second is funding: margins are lower, you have working capital tied up in stock, the pipeline is slower. Third is complexity. Connected products (and that's my concern) have mechanical parts, embedded software, connectivity/protocols, and cloud software. These need to move in sync, and it's hard to tell what takes the lead. The fourth point is business model -- the business model of products is already moving into flux. It's about to go chaotic.

    Further reading: Indiepocalypse (Andy Biao): For hundreds of years, publishers across every industry - book publishers, record labels, film studios, videogame publishers - solved problems for artists in four major ways: being, Funding, Production, Marketing, Distribution. And the internet is disrupting all four of these simultaneously.

    The way I think about this is the "fat middle." In each industry - say, news - we've had the dominant head (New York Times) and long tail (round robin newsletters). In music? Dominant head of stadium tours and U2, and the long tail of bar gigs. The internet's flattened the curve, and a fat middle has arisen. In news, major blogs: Engadget, the Verge, etc. Music: see all of YouTube.

    So... a fat middle of hardware? Yup. It's happening. Cool.

Now I can close my tabs.

16:15, Saturday 4 May.

Recent press on BERG Cloud, the new Dev Kits and Little Printer:

And one slightly older piece:

09:51, Thursday 2 May.

The good thing about rolling your own blogging system is that you're in control of your data and your destiny.

The bad thing is that you have to live with all the ridiculous choices made by you-with-13-fewer-years-experience.

So every time I have anything to say, there's a several hour (day, week, month) long throat clearing process where I have to check what the syntax for the blog posts is, see whether the rendering and publishing code still works, and get the whole thing working on my laptop again.

You'd think that this barrier to entry would result in me only posting when I had something really, really worthwhile to say. Where my desire for public exposition was so strong that it would carry me through all the pain and hurdles.

No. I write posts when I'm procrastinating, or when I'm at an airport. Today I'm procrastinating.

Update: Why on earth does my blog template put a little dot after "May" in the date?? Apparently me-some-years-ago is a lazy coder who can't be bothered to truncate correctly. This is lazy: strftime("%b. %Y")

Continue reading...

The 8 latest posts are named Filtered for washing machines, Connected products trip up the incumbents, Filtered for nematodes and Uniqlo, Red, yellow, green, bice, plunket, plaid, Coffee morning three, Filtered for storytelling, We Didn't Start the Fire Pedia, and Filtered for making and alienation.
Read them.


2014 December, November. 2013 June, May. 2012 July, May, April, March, February, January. 2011 May, March, February, January. 2010 December, January. 2009 February. 2008 December, November, September, August, July, June, May, April, March, February, January. 2007 December, November, October, September, July, June, May, March, February, January. 2006 December, November, October, September, August, July, June, May, April, March, February, January. 2005 December, November, October, September, August, July, June, May, April, March, February, January. 2004 December, November, October, September, August, July, June, May, April. 2003 December, November, October, September, August, July, June, May, April, March, February, January. 2002 December, November, October, September, August, July, June, May, April, March, February, January. 2001 December, November, October, September, August, July, June, May, April, March, February, January. 2000 December, November, October, September, August, July, June, May, April, March, February.

Interconnected is copyright 2000—2014 Matt Webb.