Good UX can saves lives. Good UX can lead to increased sales and fun apps on your phone, but it can also increase charity donations, speed up the response time of emergency services and prevent life-threatening catastrophes by making safety and operation processes easy and fool-proof.

One component of this ‘safe UX design’ that we’ll investigate here, is integrating the ability to halt an experience before it gets any worse.

Backing out of a bad situation can be nerve-wracking, particularly for management. There’s always that nagging doubt that if you power through, it might all be okay. We’re taught from a young age that giving up is cowardly and we should keep trying. If at first you don’t succeed… and all that. This rarely ends well, however.

If you hear a fire alarm, there’s no point hanging around looking for smoke. If you get out as soon as you hear the siren, you can always go straight back in if it turns out to be a false alarm. You lose a few minutes, but you can be sure you won’t lose your life.

My secondary school business studies teacher explained to us at length that refusing to give up was a particularly bad strategy on the stock market. If your stock in a particular company starts to plummet, it can be tempting to hold on until it comes back up, rather than sell it for less than you bought it for; but the longer you hold on while prices plummet, the more you lose. Conversely, if you sell immediately, you minimise your losses. If the shares start to rebound, then you can buy them back for less than you sold them for and watch them go up to where you started.

Andon Cord

This is the exact philosophy that they work by at Amazon. If any employee, even the lowliest muggle, sees an issue with any of their products, they can immediately put a freeze on the product’s listing, preventing any more from being sold until the product has been investigated.

It’s all based on Andon Cord, yet another brilliant production technique created by the parent of modern business practice, Toyota. Named after Japanese paper signal lanterns called ‘Andon’, the cord was literally a rope hanging above the production line that anyone could pull to immediately ‘stop the presses’ and shut down production.

Does this mean that the occasional work experience temp will cost the company a few sales by putting an unnecessary stop on something? Of course. Is it worth it to know you’re never sending out broken products to your customers and tarnishing your own reputation while you wait for someone to get authority from their boss’s boss’s boss’s boss to stop production? Absolutely!

I used to work at a company that sold tickets to Sea World, right around the time Blackfish came out and a public backlash was brewing about the treatment of the animals. I had to form a committee to petition the board to cease sales of the product, which took weeks. All the while, sales were continuing and we were risking a PR nightmare from a situation senior management weren’t even aware of, when they could have been held until a decision was made and a strategy implemented.

Please check back again later

So that’s Amazon and theme parks, but what does Andon Cord mean for your average UX project? Well, at every stage of your user journey, you should build in an Andon Cord that can be pulled to shut everything down when someone sees it’s going wrong; whether that’s an employee noticing it, the customer reporting it or something flagged by an automated monitoring tool.

Whatever the source, when the cord is pulled, it is better to cut the user journey short with an apology, than let their experience run off a cliff and lose them as a customer forever.

A simple screen saying “whoops! It’s all gone wrong, Please check back again later” can avoid a disastrous experience, and with modern tools, you can capture that failed user journey and send an automated email to the customer when your site is back up, offering them a discount as an incentive to come back and try again. Whatever it takes to make sure that you don’t make things worse by not reacting to a problem quickly enough.

Luckily, it’s unlikely to be anything too dramatic that you’re dealing with, but you should always be ready to stop your process, take a step back and fix an issue at any time. The openness to taking a small hit can prevent disaster for both you and your customers.

It’s worth bearing in mind that customers have their own Andon Cord too, it’s popularly known as ‘going to your competitor’s website instead’.


For a free, in-depth guide to running user research in Agile Sprints, read our comprehensive e-book:

Download “How to Run Research in Agile”


Main image by Kelly Sikkema