by Austin Smith
I attended LMA and LMC’s Elevate! conference in Chicago last week to discuss Alley’s recent design work for The Dallas Morning News with Mike Orren, their Chief Product Officer, in front of an audience of local media executives. Our conversation ranged widely into what makes news websites work well, and at one point I veered into a metaphor that I often use to discuss how to make tough choices about page speed while designing a news website: Airplanes.
One of the most successful elements of their new site is how much faster it is than their old site. We approach page speed as a design problem — sometimes even an organizational design problem — that leverages technology rather than a technology problem that treats design as an externality. We also believe that great creativity can arise from the toughest constraints. We’re lucky to have partners in Mike and his colleagues who really believe that speed matters most and are willing to do the hard work to make tradeoffs.
Digital news products are some of the toughest on the internet because every page load has to serve so many different purposes. Serve the article, but also four ad units, a recirculation unit that ideally is personalized to the reader, some internal marketing, and maybe even a video. And of course, this all has to happen fast — within a couple of seconds.
Light web pages load fast; light airplanes fly quickly. But news websites can’t eliminate editorial content or monetization units any more than airlines can eliminate seats or doors, so we have to find other ways. Some tradeoffs are simple enough; airlines can use lightweight composite materials; we can use system fonts on news websites. Interestingly, airlines mostly annoy their customers when they try to save weight — like eliminating in-flight entertainment or charging more for checked bags — but news websites tend to annoy their customers by adding weight. More ad units, more video, more auto-loading next articles.
When airlines configure their new planes for service, they do so with a strict sense of the tradeoffs they are willing to make. Every flight has a weight budget which is influenced by heat, altitude, and runway length, among various other factors. So every airplane has to fit the normal parameters of the toughest flight it could be assigned to — the longest route with the highest runway in the hottest climate.
Those responsible for designing the customer experience on an airliner are thus fighting three hard truths:
- Heavy planes can’t get off the ground in the first place, so there’s a natural limit to the number of seats and the materials used.
- The customer experience should still be comfortable and feel trustworthy to most passengers, most of the time.
- But airlines have to pay for the customer experience, the fuel, and the labor, and they’ll lose money if they can’t sell a seat on the same route as a competitor for about the same price unless their product is highly differentiated. And even then, they lose out on casual buyers.
Sound familiar to news websites? We’re up against the same stuff:
- Sites that load too much extraneous stuff will take too long to load and the customer will lose interest and hit the back button.
- The site should still look good and feel trustworthy to most readers, most of the time.
- But news organizations have to pay for the customer experience, the web servers, and the labor, and they’ll lose money if they have to ask readers to pay at all unless their product is highly differentiated. And even then, they lose out on casual readers.
Obviously there are lots of differences, but only one really matters — what’s at stake if your airplane or news site is too heavy. At best, heavy airplanes can’t fly. At worst, they crash. Both outcomes are terrible for airlines and must be avoided at almost any cost.
The outcome for news organizations is also bad for business but certainly less lethal. Still, looking good and feeling trustworthy won’t matter if your users bounce quickly, and this has been true since at least 1997. It’s not like we don’t talk about page speed as an industry — in fact, we talk about it a lot. But stakeholders often talk about page speed as a KPI to optimize, not a hard and fast rule to live by. A blank webpage has a Lighthouse score of 100 and loads in milliseconds. Pick a score, say 80, and a number of seconds, say 1.5 — and measure everything you add to the blank page against those targets. The consequences are minimal for news websites in that a page which scores 78 and loads in 1.6 seconds barely feels slower, but it’s a slippery slope and thus requires diligent monitoring over time.
If you treat your page speed like airlines treat their planes’ weight limits, your readers will thank you by sticking around long enough to engage with your content. It’s not a silver bullet, but it’s a start.
Want to fly? Let us know.
Originally published at https://alley.co on September 26, 2019.