Uncovering the core difference between good and bad products & services

Image credit: the estate of Heath Robinson

Everything we use in the world around us comes with its own user experience, intended or otherwise.

Often, however, those who produce the artifacts we rely on fail to perceive the user experience they are creating. I’ve always been fascinated by the way in which even very large companies with enormous budgets very often create poor user experiences because of their inability to conceive how the ordinary user will interact with their products or services.

We all know its easier to swallow a gel-cap than to ingest a large chalky tablet with hard edges. That’s why more and more orally-administered products take the form of gel-caps. But there are still a great many products ranging from supplements to antibiotics that are still manufactured in a user-unfriendly format. This is presumably because factors such as manufacturing cost outweigh, at least in the minds of the accountants, the benefits of making products that are easier for the customer to swallow.

We’ve all encountered websites where we’re asked to create a username and password. Astonishingly, it’s still common to find sites that make us play guessing-games by requiring a certain combination of characters and integers without informing us in advance of what these requirements are. Thus we only discover the requirements as error message after error message tells us what we’ve done wrong.

It’s obvious that these sites were never given to an ordinary person to test; instead the engineers tested their own code and to them, having written the code in the first place, it was simply obvious that the password field required at least one upper-case letter and one lower-case letter and one integer and one special character (but not, of course, ! or # or *) and must be between eight and ten characters long.

Providing clear guidance as to the required password format would save users frustration and wasted time, but such things are not always obvious to software engineers because they don’t see the product in the same way that regular users do so they’re blind to its defects.

Amazon is rightly lauded as the company that brought eCommerce into the mainstream. It innovated not only things like one-click purchasing and recommendations but also ensured that the total customer experience was smooth. It’s therefore very odd indeed that their streaming video service is flawed.

I signed up for Amazon Prime and Netflix while living in the USA and as far as these two companies are concerned I’m still a US customer; however, they both detect my IP address and recognize that I’m actually in Switzerland. Netflix thus shows me only choices that, due to content licensing issues, I can actually watch here, which is logical. There’s no point showing me movies or TV series I won’t be able to watch due to legal restrictions. Amazon, however, shows me its entire catalog and it’s only when I actually click to purchase a show or a movie that I’m shown a message saying This content is unavailable in your region.

It would, of course, be more user-friendly to refrain from tantalizing me with things I won’t actually be able to see. Netflix has its own flaws, though: it assumes that because I’m in Switzerland I speak fluent German. Thus all titles, captions, and everything else are rendered in German regardless of the language I select as the audio track. For the sake of amusement I recently watched an old episode of Sherlock in which all the pop-up text messages were rendered in German, as were all visible signs and dates. Doubtless some engineer thought this would be “helpful.”

Things get much worse when disparate elements of a product are produced by people who never experience the entire assemblage. Automobile navigation systems are often superb examples of user-unfriendliness, indeed so much so that personally I’d never buy a German car for precisely this reason. The first generation of Mercedes navigation systems would guide the user to within a mile of the desired address and then shut down, leaving only the comforting message “You are now within one mile of your destination.”

BMW and VW-Audi navigation systems utilize voice commands, which the system frequently fails to recognize, and the wheel-based alternatives are painfully slow to respond and even more laborious to use as one attempts to enter, character by character, one’s destination. Only to discover that 4 Meinhoff Strass isn’t a valid way to enter the address, so one must play guess-the-format until one discovers by process of elimination that one must enter Meinhoff Straß 4. And then select from a menu of options that contains hundreds of entries beginning with (for no rational reason whatsoever) Aberfeld Straß 91.

But the fun with the VW-Audi satnav doesn’t end here. If you do succeed in finally telling the system where you want to go, you’ll be provided with a route that regardless of the criteria you’ve selected (such as Fastest Route or No Secondary Roads) will reliably send you down dirt tracks and demand that you drive hundreds of miles out of your way in order to conform to VW-Audi’s unique notions of how to get from where you are to where you want to be.

In 2019 I drove a brand-new Audi S3 from Lake Como to Lausanne using both the Audi system and a supplemental Garmin system. The Audi system was hilarious, attempting to send me down dirt roads at every opportunity and seemingly intent on finding the worst-possible option at every point along the way. Finally, when the Garmin system told me I was only 117 kilometers from Lausanne, the Audi system was insisting I turn around, drive back toward Milan, and then proceed according to its “superior” guidance algorithm. This helpful direction would have added a mere 208 kilometers to my journey.

Oh, and the Audi system also assumed I’d be permitted to drive at an average speed of well over 220 kph despite Switzerland’s maximum 110kph limit.

Image credit: the author

While Germany automobile manufacturers are stunningly clueless when it comes to software, French manufacturers are little better. A handful of years ago I rented a top-of-the-line Peugeot for a drive from Oxford to Newcastle. The navigation in this vehicle had the “user-friendly” feature of auto-zoom based on the speed of travel. As soon as I got onto the motorway and reached 70mph, the map zoomed out to its greatest extent (there was no manual over-ride) which resulted in a very amusing problem. Some software engineer at Peugeot must have decided, perhaps after one too many lunchtime pastis, that the navigation system should helpfully display all petrol stations and Peugeot dealerships along the route, and that this amazingly helpful display should not be something the user can turn off.

The result was that the 7-inch display was entirely covered with petrol station symbols among which were scattered a dozen Peugeot dealership symbols. It was literally impossible to see any part of the map under these “helpful” displays. Fortunately, my Garmin wasn’t so “helpful” and I was able to see my route with ease. And the Garmin took me precisely to the desired address instead of turning itself off half a mile away like the Peugeot system.

While software obviously opens up lots of opportunities for “helpful behavior” we find similar problems with more mundane products.

During my time living in Marin County just north of San Francisco, I’d always commence my gym workouts with half an hour or so on an exercise bike. When I started going to my favorite gym the bikes were excellent. Alas, they wore out and were replaced by the management. The new bikes looked lovely: they had clearly been shaped by a proud graduate of the Precious Darlings School of Design & Self-Indulgence. The new bikes were swoopy, graceful, and clearly futuristic.

Unfortunately, they were also totally unsuited to the human body. The (non-adjustable) saddle position was totally wrong in relation to the pedals, which meant that damage to knees occurred quite rapidly. In addition, soreness of the lower back was induced by (non-adjustable) handlebar placement. In short, these bikes were an ergonomic catastrophe and it was obvious that no one had every been asked to use a prototype in advance of the manufacturing order being placed.

Poor design isn’t confined to civilian products. The British Ministry of Defense has a long and proud tradition of providing British troops with hilariously useless kit. Many years ago the manpad Blowpipe G2A anti-aircraft system was a great example: assuming the operator had two heads and three arms, it could just about be made to work. In the 1982 Falklands conflict, more than 100 Blowpipe rockets were discharged against attacking Argentine Airforce planes; not a single rocket got close to its target. Conversely the one US Stinger that was fired destroyed its target easily.

The British soldier also got to enjoy the original SA80, which was so poorly designed and so badly made that there was literally no environment in which it could function for more than a few rounds before jamming. For some reason, since the end of World War II, British engineering has largely followed the principle of “well, it doesn’t work, but we didn’t expect it to anyway, so that’ll do.”

Image credit: War Is Boring

And today, despite many retrofits, the SA80A3 battle rifle is still not very good.

Conversely the Desert Tech MDR, which is also a bullpup assault-style rifle of the same approximate dimensions and weight, is an example of superb design. It’s fully ambidextrous and the ejector port can be swapped from one side to the other in less than two seconds. All casings are ejected forward, so as to avoid hitting anyone who’s alongside. There are two separate levers to drop a spent magazine, both within easy reach yet positioned to avoid accidental contact. And the priming handle is likewise found on both sides of the upper so as to facilitate fully ambidextrous stoppage drills.

The MDR is a rifle aimed primarily at the civilian and contractor markets, made by people who thought carefully about customer requirements rather than about how much margin they could extract from a poorly-written Ministry of Defense procurement contract.

Tarrying a moment longer with the hapless Brits, the new British Army’s light battle tank (really an APC) is rumored to exhibit a major problem: the hatch-open lever position when the driver’s hatch is open intrudes into the hatchway, making extraction of an injured driver nearly impossible. Evidently the British just can’t overcome the impulse to mess things up. No doubt the manufacturer of the vehicle will argue that the solution is to redesign the British soldier so that the new Body-Shape Mark II can cope with the intruding lever.

The entire indigenous British automotive industry collapsed under the weight of very similar “bodge engineering” attitudes. The iconic Rover SD1 famously wore through front tires with a prodigious appetite. The reason was that the front subframe on the right was nearly one inch longer than the subframe on the left, which made it impossible to align the front wheels correctly. But Rover couldn’t be bothered to fix the problem so every single one of the hundreds of thousands of SD1s produced during its eleven-year life had exactly the same issue.

Image credit: Magzter

Added to this charming quirk, most vehicles left the factory already rusting under the poorly-applied paintwork, the electrics were so shoddy than lights failed with monotonous regularity and alternators fried batteries by over-charging them. Trim fit was atrocious, even by the low standards of the day, and pretty much every car delivered had at least one glaring defect. Oh, and the tooling for engine production had worn out, so pistons slapped around in the cylinders from day one. The Rover SD1 was British Leyland’s showcase car; other models were far, far worse.

Since then, under the influence of quality-obsessed Japanese manufacturers, automobiles everywhere have improved immeasurably. They are now safer, more comfortable, more reliable, and more well-finished than all but the most luxurious hand-built machines of forty years ago. No doubt twenty years hence, their navigation systems will have been debugged also. But for those who don’t want to wait that long for German and French producers to catch up, Japanese, Korean, and (surprisingly) US navigation systems are all infinitely superior both in terms of ease-of-use as well as in terms of route planning.

Japanese automobiles also avoid annoying their users. Anyone remember the fad for “safety alarms” back in the 1980s and 1990s? I rented a Chevrolet in 1992 that had six separate warning noises, all of which went off simultaneously under certain conditions, creating an impossible cacophony that left the user totally confused. Even under optimal conditions (parking brake on, engine off and key out of the ignition, lights off, windows rolled up) there were still two warning chimes that were activated whenever the driver’s door was opened. It’s difficult to understand the mentality of the product team that thought these “helpful” noises would result in a better user experience.

I want to finish with an example that shows how design isn’t just about ergonomics or clarity of information provision or the avoidance of irritating noises.

When Automated Teller Machines (ATMs) first came into use, I was eager to try one out. I inserted my bank card, punched in the passcode, and told the machine how much I wanted to withdraw. I immediately saw the mistake:

The first ATMs gave you your money first.

So a great many people wandered off, happily stuffing their crisp new bills into their wallets or purses, oblivious to the fact their card was still in the machine. Astonishingly it took nearly 15 years for the vendors of ATMs to make the simple and obvious alteration: require people to withdraw the card before dispensing the cash and receipt. One tiny change in the process flow made all the difference in the world.

Process is just as much a part of the user experience as anything tactile, visual, or auditory.

The take-away from all of these examples is very simple: before you release your product, whatever it may be, always (always!) ask people who know nothing whatsoever about it to try it out for a while. Watch them closely to see what happens. Listen to their feedback.

Then ignore everyone in your organization who tells you that the negative feedback “doesn’t matter” and fix the problems before you release your product or service into the marketplace.

It really is that simple.

Anyone who enjoys my articles here on Medium may be interested in my books Why Democracy Failed and The Praying Ape, both available from Amazon.

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