By John Gruber
WorkOS Radar:
Protect your app against AI bots, free-tier abuse, and brute-force attacks.
Just after New Year’s some sort of underground cable screw-up resulted in our home, along with an irregular swath of our neighborhood, losing electricity for 26 hours. We don’t lose power often, and when we do, the outages are usually brief, but 26 hours felt pretty long — especially with the outside temperature below freezing and daylight hours near their calendric nadir. The icing on this particular outage’s frustration cake was that our power company, PECO1, seemingly had no idea what exactly was wrong or how long it might take to fix.
The power went out around 10:30 am on January 2, and soon thereafter PECO was estimating that power would be restored by 2 pm. Then it was 4 pm, then it was briefly 2 pm again (despite the actual time then being after 2 pm — which is when I got the sinking feeling I should get the flashlights out), then they were claiming there were no known outages in our area, until eventually they just stopped providing any estimates at all of when our power might return. I’d have given PECO some credit for honesty if they’d simply replaced the estimated time for power restoration with the shrug emoji.
I was following along with these updates and checking the outage map from my iPhone, on PECO’s website. Which website I wasn’t at all familiar with, because our power really doesn’t go out very often, and my wife takes care of the bill. PECO’s is one of the worst websites I’ve ever had the misfortune to need to use. Among its faults:
Basically, PECO’s mobile website feels like it was developed using and exported from Microsoft Excel. You might say, “Well that makes no sense, because you’ve never been able to build or export websites using Excel.” To which I’d respond, “Yes, exactly.”
So, every time I wanted to see if there was an updated estimate on our power being restored, it took at least a minute or two of waiting for pages to load, signing back in (which was always slow), and poking around their inscrutable site navigation. The website did prompt me, occasionally, to install their mobile app, but I was like “Fuck that, it’s probably just their website in a wrapper.”
It was a cold and dark night, but our power was restored the next day just after noon,2 and it stayed restored, so I metaphorically dusted my hands and thought to myself, “I hope I never need to use that fucking website ever again.”
Last night, our power went out again. This time, thankfully, it was only out for about 80 minutes. When the outage hit, before even once trying PECO’s cursed website, I went to the App Store and installed their iPhone app. It was a revelation. PECO’s iOS app is everything their website is not: fast, well-organized, and, blessedly, keeps you signed in.3
I’d go so far as to describe PECO’s website, at least as experienced from a phone, as utterly incompetent. I’d describe their native iOS app as — I can’t believe I’m going to use this word — good. It’s hard to believe the website and app are from the same company.
This makes no sense to me. A utility company is the sort of thing where I’d expect most people would use them via the web, even from their phones. Who’d think to install an app from their electric company on their phone? But it’s a night and day difference. I feel like a chump for having suffered through the previous 26-hour outage obsessively checking their terrible, slow-loading (I just can’t emphasize how fucking slow it is), broken website when this app was available.
There’s absolutely no reason the mobile web experience shouldn’t be fast, reliable, well-designed, and keep you logged in. If one of the two should suck, it should be the app that sucks and the website that works well. You shouldn’t be expected to carry around a bundle of software from your utility company in your pocket. But it’s the other way around. I suspect that my instinctive belief that a service company or utility should focus its customer service efforts on the web first, and native apps second, is every bit as outdated as my stubborn belief that invite ought not be used as a noun. (Invitation is sitting right there.)
I won’t hold up this one experience as a sign that the web is dying, but it sure seems to be languishing, especially for mobile devices.4 And the notion that mobile web apps are closing the gap with native apps is laughable. The gulf between them is widening, not narrowing.
They were The Philadelphia Electric Company for over a century before changing their official name in 1994. They should have kept the old name rather than rebrand, despite the fact that no one in Philly had ever called them anything but “PECO” for decades. Nobody here ever called Veterans Stadium anything other than “The Vet”, but it would’ve been stupid as hell to officially rename the late great concrete masterpiece of early 1970s brutalism to its nickname. ↩︎︎
In what I’d hold up as yet another proof of Murphy’s Law, the power came back on while I was mostly done with a shower that wasn’t cold, per se, but certainly wasn’t warm, let alone properly hot. ↩︎︎
While writing this column, I installed PECO’s Android app on my Pixel 4 and gave it a whirl. It shares a visual design with the iOS app — I strongly suspect they’re made from a shared code base and one of the various cross-platform frameworks. But where the iPhone app is fast (or at least fast enough), the Android app is slow. But I can’t say how much of that is from the app and how much because my Pixel 4 is five years old. But I also tried the iOS app on my iPhone 12 (four years old), and it felt snappy there too. ↩︎︎
It’s kind of weird that there are now zillions of supposedly technically sophisticated people who, when they use the term “desktop app”, are referring to websites. I’ve personally mostly thought about this usage as a sign of the decline of native Mac apps. But it’s also a sign of the decline of building websites meant to be used on mobile phones. I think maybe what we’re seeing is not that the web, overall, is dying, but the mobile web is. ↩︎︎
Previous: | Nvidia, the New King of Keynotes |