How Nest’s Tiny Green Light Inspired Revolt

Consumers don’t always see tech product updates as improvements

Thomas Smith
Dec 26, 2019 · 14 min read
Images courtesy of Gado Images.

August 2019, Google made a change to a tiny green light on its Nest security cameras. The company was probably totally unprepared for the complete horror-show that followed.

The episode began innocently enough when Google decided to do some digital housekeeping. Across their brands, they took lots of little steps to shore up their privacy practices. This was in response to legislative changes in Europe, a political climate that increasingly emphasizes data freedom, and other relatively mundane factors.

One of their targets was Nest, Google’s popular home automation company. Google zeroed in on a toggle in the interface for Nest’s home surveillance cameras. By default, a little green light glows on the front of the Nest camera when it’s recording. The toggle allowed users to switch off the light, keeping the camera dark even when it was in use.

Green light on a Nest camera which is actively recording.

Recording without a status light—and thus potentially without peoples' knowledge—was an issue, the company decided. This feature should probably be disabled. Google sent out a friendly email to Nest users updating them about the change.

Then, all hell broke loose.

Users took to social media to lambast, scream at, and digitally crucify Google. Here’s an example post from Google’s forums, from user ch8417:

Stupidest. Move. EVER. Google Stop forcing your left-wing ideas on half the country. Just because YOU want the bad people to know they’re being recorded does NOT mean that *I* do. Why do you think I bought this camera in the first place??????? STOP THIS NOW. Bring back MY privacy.

Believe it or not, many comments were even angrier — and more political. Users were threatening to boycott the Nest brand or dispose of all their existing Nest products in some kind of Bonfire of the Cameras.

Why all this vitriol over a tiny green light?

To understand why Google’s change nearly prompted a user revolt, we have to take a step back and look at what people expect from their products—and how modern tech products are set up to let them down.

Static products

Consider your toaster. Why did you buy it? Why that toaster and not some other toaster?

You probably looked at its features. Maybe it had some snazzy toast optimizing settings. Maybe it came in a pretty color. Maybe you liked the brand. Maybe the product name sounded cool.

Now ask yourself this: When was the last time your toaster spontaneously changed? I don’t mean the ways in which it has broken or worn down, as toasters do. I mean the last time it turned a different color or sprouted new buttons. Or the last time one of its existing buttons disappeared.

Traditional products are static.

Barring some mysterious dark magic (that I’d love to hear more about), it’s unlikely your toaster has ever morphed into a totally new product. It has probably kept the same features it had when you bought it and hasn’t shed or added any new functionality.

After centuries of buying stuff, this is what we’ve been trained to expect. When you buy a product, you consider its features, price point, etc. You make a buying decision, and the product is yours.

Most of the time, if the product later changes, this is not only surprising but illegal. If you bought an annuity and the terms suddenly changed, you’d have the SEC knocking on the seller’s door and a cause of action for a lawsuit. Products are supposed to stay static, and this is what consumers expect.

The idea of a product

Today’s tech products completely violate this long-held expectation. That’s because what you’re buying and placing on your kitchen counter often isn’t actually the product itself — it’s just a conduit to the product, which lives somewhere else.

Consider the Amazon Echo. Alexa doesn’t actually live in there. She lives in a massive computer distributed across thousands of servers in Amazon’s cloud. Your Echo is just some fancy microphones, a Wi-Fi chip, and a couple LEDs.

Alexa doesn’t live in here.

When you buy an Echo, you’re not buying a physical version of Alexa. You’re just buying a little widget that allows you to access her on Amazon’s servers, where she really lives.

It’s as if when you put bread in your toaster, a tiny gnome appeared, took the bread to a restaurant kitchen, toasted in perfectly, and returned it to your toaster, all without you ever realizing it had left.

In a sense, today’s tech products are not really products at all. They’re more the idea of a product.

Likewise, the Google Chromebook — one of the best-selling categories of laptop — has a wimpy processor on its own. The power of the machine is in the services it connects to — Gmail, Google Docs, YouTube, etc. All of these live not on the device itself, but on someone else’s cloud. The laptop is just a conduit to these more powerful, remote computers and software platforms.

In this sense, today’s tech products are not really products at all. They’re more the idea of a product: some physical hardware, some remote computers, and some services, all tied into a semicoherent offering that you buy in a store for $149.99.

And that means these products are constantly changing. Because your widget isn’t the product itself, the company that sold it to you can easily change its features on the fly. All this takes is an update to some software code on a remote server somewhere or a remote update to the device’s firmware.

The services behind it might become more expensive (or cheaper). New capabilities might be added. Or — crucially — capabilities might be taken away. Because they’re constantly in flux, let’s call these new kinds of hybrid hardware/software/service offerings “dynamic products.”

Consumers left behind

The issue with dynamic products is that consumers haven’t gotten the memo about their existence. The vast majority of the time, they expect that the widget they’re buying and sticking in their home is an actual product with specific, concrete features. They don’t want the idea of a product. They want a thing that behaves like the toasters of yore.

Some companies have done a good job leading customers through the transition to dynamic products. Tesla, for example, ships all its cars with the hardware needed for full self-driving. To switch it on, though, you have to pay Tesla some exorbitant sum of money. They’ll do an OTA update to your car’s firmware when you buy it, and boom, the self-driving gizmos will switch on.

The hardware itself isn’t the product — every Tesla owner has it. It’s a combination of the hardware and a service that only Tesla can activate.

Credit: XKCD

It’s kind of like a classic cartoon on XKCD, a web comic popular in the tech sector, where the protagonist wants a USB port installed in their arm. They figure that it won’t work at first, but once the hardware is installed, someone will come along and write software for it, allowing them to control USB devices with their mind.

This version of a dynamic product works for Tesla because they’re straightforward about it. You can buy a Tesla for one price with self-driving switched off, or a higher price with it switched on.

There’s also a clear exchange of value: You pay to activate the service on your existing hardware, and Tesla commits to improving and maintaining it so your Model 3 doesn’t drive into lakes or walls. And they also commit to accepting the liability inherent in self-driving tech, a move for which they deserve to be compensated. Both parties give something up and get something in return.

Most companies, though, have done little to transition consumers away from static products. And in some cases, they’ve even worked hard to drag people to this new paradigm—kicking, screaming, and suing.

The fury of farmers

The best example of consumer resistance to the new paradigm of dynamic products actually comes not from tech, but from agriculture.

Modern farm equipment is incredibly complex and expensive, with GPS guidance, sensors everywhere, and millions of lines of code running on onboard computers. And it has to work flawlessly, often at specific, critical times of the year. A combine harvester that fails days before the harvest could cost a farmer their livelihood.

For that reason, farmers have spent generations learning to repair their own equipment. But as this equipment has become more and more tech-driven, that’s meant both fixing physical issues and software bugs.

As reported in Vice, big companies like John Deere have wised up. They don’t want to lose lucrative repair monopolies or the revenue from licensing authorized service providers. For the better part of a decade, they’ve been working to restrict farmers from making even simple repairs to their own equipment, wrapping their products in end-user license agreements that prohibit tinkering.

Tractors have come a long way. (Image courtesy Gado Images archive)

When you buy a John Deere tractor today, you’re getting a dynamic product. It’s a physical tractor and the legal right to use the software installed on it. And this right is restricted; the software is there for your use, but if you violate the EULA (for instance, by making your own repairs) John Deere could conceivably sue you or brick your tractor.

Farmers are understandably pissed about this. And they have started fighting back—and winning. Under a provision of the Digital Millennium Copyright Act that allows for backup and limited modification of software, farmers are now able to hack their own equipment in many cases to keep it running. And their fury over being locked out of their own equipment has catalyzed into the Right to Repair movement, now championed by presidential candidates, including Elizabeth Warren.

The green light of mutiny

Seen in the broader context of static versus dynamic products, the anger over Google’s tiny green light makes a lot more sense. When consumers purchased their Nest cameras, they expected them to have a certain set of features, like a toaster or a traditional tractor. In short, these consumers were expecting to receive a static product.

For many, the feature allowing them to switch off the little green light may have been a big selling point. Beyond creepy snooping, this feature is helpful if you’re filming through a reflective window, using the camera as a baby monitor, or just trying to secure your home without alerting burglars to the camera’s presence.

Now here’s a big company coming along and telling them their product was actually dynamic. It could be changed on the fly, and its feature set would soon be changing without their advanced knowledge and without their consent.

Add in that this was a hardware change to a physical thing (albeit a small one) and that these cameras occupy an intimate place in the home, and it was likely too much for some consumers to handle. The result was a backlash, an outpouring of anger, and a passionate, if short-lived, consumer mutiny.

Dynamic products are your friends

So are dynamic products always bad? Should companies focus on supplying static things with a predefined and unchanging feature set?

No. In many cases, dynamic products are a wonderful thing.

I bought my Amazon Echo Dot three years ago. Since that time, Amazon’s Alexa service has improved tremendously. Its speech recognition is more accurate, it has more third-party skills, and it’s better integrated into Amazon’s ecosystem.

My ancient Echo Dot is dynamic and still works fine.

Through the magic of dynamic products, my three-year-old Dot has mostly kept up with these positive changes. Sure, it’s not as snazzy looking as the newest Dot. But because it ties into the same Alexa service as the newer models—and because that service lives somewhere in the cloud and not on the device itself—I’ve been able to keep up with the newest tech advances without having to buy new hardware.

The magic of dynamic products is that they can grow, change, and improve even after you buy them.

Even in the Nest ecosystem, the same effects apply. Google is constantly introducing updates and improvements to its Nest cameras. Since I bought my original camera, the monthly monitoring cost has dropped, the app has improved, and it’s sprouted all kinds of A.I. capabilities, like people detection and a feature that knows if my dog is barking.

This is the magic of dynamic products. Because they’re not tied as strongly to physical widgets, they can grow, change, and improve even after you buy them. The software and services that are a big part of their core value are subject to iteration, improvement, and updates in a way that static hardware is not.

This even has positive implications for product longevity and sustainability. Since my Echo is constantly making itself better, I’m not locked into a two-year replacement cycle, like with many older tech gadgets. And that cuts down on the obsolete hardware I have to throw out — definitely a net gain for the environment and my wallet.

More to the point, consumers rarely take up arms over the addition of new features to an existing device. When Google added A.I. capabilities to its Nest products, you didn’t have customers taking to forums and writing “You JERKS, you gave me person detection? Stupidest. Move. Ever.”

People love it when products become more useful, and that’s a big selling point of dynamic products. Barring the aforementioned dark magic, your toaster is unlikely to compete in this area.

How to do dynamic right

So what can companies who sell dynamic products learn from the green light incident?

First, changes to existing devices should not be taken lightly. Switching on a green light doesn’t seem like much. And from a privacy standpoint, it probably does make sense.

But any change to a device has the potential to rile up consumers, especially those who are used to the static product paradigm. Companies should treat any change to an existing product — especially the removal of a feature — as a big deal and give customers time to process it and respond before changing up the feature set.

Second, consent is key. One of the reasons for the Nest mutiny is that it came from up on high, and consumers had no ability to opt out. Instead of involuntarily switching off the feature, how about switching it off, but including an option to virtually sign a disclaimer and switch it back on?

In this case, for example, the light could have defaulted to the “on” setting after the change. But consumers could have been given the option to accept a TOS change, with text like “By enabling this feature, you agree that your use of the Nest camera will not violate any privacy laws in your jurisdiction.” Customers who didn’t care about the light could leave it on. Those who did care could sign the new TOS and switch it back off.

Boom—liability avoided and customers placated. And since Google knows customers’ addresses, they could even ship every customer who re-enabled the feature a “Protected by Nest Cameras” sticker to put on their door. This would ensure that customers disclose the presence of a camera, and it would be free advertising for Google. Two birds with one stone.

When removing a feature, dynamic product companies should always consider whether there’s a way to leave the feature in, with enhanced liability protections for themselves, or some other option that balances their own needs with those of their customers.

Finally, when it comes to hardware products, companies should tread softly. Very softly. The idea of physical products in your home that connect to massive, unknowable, off-site infrastructure is still something consumers are coming to terms with. And they’re understandably jumpy.

Earlier this year, Apple and Amazon faced a major backlash when it was revealed that outside contractors were reviewing conversations recorded by their voice assistants. In the tech world, the fact that this was happening seemed obvious — how else do you train a deep learning system if not through human review of data? But to consumers, it was a revelation, and not a positive one. People felt violated, and experts started warning against using the devices in private areas of one’s home.

It was a bit like Mark Zuckerberg testifying to Congress, who seemed to have been shocked to learn that the company used personal data to sell ads. You mean you weren’t providing a hugely valuable service, for free, for no reason at all? Zuckerberg’s patient explanation that “this is the Internet’s business model” also felt obvious in the tech world but was probably quite shocking for most of the population. Tech needs to be aware of this disconnect and do more to actively bridge the gap in consumer understanding.

A final word: Trust

In some ways, the level of trust that consumers have been willing to place in dynamic products is remarkable. If you told someone in the 1980s that in a few decades we would all place web-connected microphones and cameras in our homes, tell advertising companies the intimate details of our lives (and send them pictures of all the food we eat via Instagram), and carry GPS-enabled tracking devices (cellphones) with us everywhere we go, they would either laugh or plotz. Today, it’s just the world we live in.

Screens and microphones are everywhere in our homes

Again, consumer acceptance of dynamic products is a huge net gain to society. Voice assistants, cellphones, and even surveillance cameras are improving faster, costing less, and lasting longer than ever before. And tractors are much better today than they were in the 1980s, too, by the way.

But if companies want to continue to receive this level of trust and buy-in from consumers, they need to make sure consumers fully understand the trade-offs they’re making and feel comfortable with them. Google’s green light incident is instructive and provides some valuable lessons to dynamic product companies.

Ultimately, it’s good this happened with a little green light and not something more important. No company likes angry consumers. But it’s much better to discover these kinds of latent customer concerns through a change to a tiny green light, rather than a change to a self-driving function that sends cars hurtling into embankments — or even a privacy scare like the one Amazon and Apple are facing.

Google is a smart company. They’ll learn from the green light mutiny and apply those lessons to future feature updates. And other dynamic product companies can learn from it too.

If you’re changing the feature set of your product, make sure customers feel fully in the loop, give them the chance to provide consent wherever possible, and be especially cautious about changes to hardware. People have trusted you enough to allow your device into their homes, and it’s essential to be a good steward of that trust, both for your own company and the industry as a whole.

And if you’re a Nest user and you’re still upset about the green light, I have two words for you: gaffer tape.


Helping designers thrive.

Thomas Smith

Written by

Co-Founder & CEO of Gado Images. I write, speak and consult about tech, privacy, AI and photography.



Helping designers thrive. A Medium publication about UX/UI design.

Thomas Smith

Written by

Co-Founder & CEO of Gado Images. I write, speak and consult about tech, privacy, AI and photography.



Helping designers thrive. A Medium publication about UX/UI design.

Medium is an open platform where 170 million readers come to find insightful and dynamic thinking. Here, expert and undiscovered voices alike dive into the heart of any topic and bring new ideas to the surface. Learn more

Follow the writers, publications, and topics that matter to you, and you’ll see them on your homepage and in your inbox. Explore

If you have a story to tell, knowledge to share, or a perspective to offer — welcome home. It’s easy and free to post your thinking on any topic. Write on Medium

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store