The race is on for eCommerce sites to deliver the fastest website speeds for consumers, but the speed metrics that companies rely on for speed optimization are misleading. The industry has been focusing on speed metrics that are more relevant for the traditional, multi-page website rather than those relevant for modern websites, like Single-Page Applications (SPAs) / Progressive Web Apps (PWAs). Specifically — first page loads have been overemphasized. And this emphasis is misguiding developers and product managers at eCommerce companies of all sizes.
Consider this: The average eCommerce browsing session lasts just over three minutes and spans 6 pages, with an average of seven second load time for each page. This means that consumers are spending 20% of their session waiting for pages to load. Shaving a few seconds of the first-page load alone won’t impact the wait time much, but optimizing the first-page load and cutting browsing transitions to sub-second levels can make a huge impact on business and reduce wait time down to a mere 5%.
Browsing speed matters
This does not mean to say first-page loads don’t matter — they certainly do as they impact your website’s bounce rates, and can affect your pool of potential buyers. In fact, research shows that 53% of visitors will abandon a website that takes longer than three seconds to load. But the first-page isn’t the entire journey, and wait time for subsequent pages matters, too. Every one-second delay in page load times cuts conversions by 7%, and with six pages per session there are six pages to optimize across the path-to-purchase. Optimizing for first-page loads and subsequent page loads will help eCommerce companies boost customer experience and overall revenue.
Since late 2018, website speed impacts the entire digital strategy. Google’s ‘Speed Update’ ties mobile site speed to SEO stature and ROI from SEM, with faster sites ranking higher and bidding lower than slower sites for the same keywords. Therefore, mobile site speed impacts your company’s visibility, traffic, conversions, and revenue.
Ever since the ranking algorithm change rolled out, eCommerce professionals have been relying on Lighthouse and other tools to measure page-load speeds. Unfortunately, these tools don’t measure browsing transitions, i.e. the time it takes for content to become visible on a subsequent page in a session. And you cannot optimize what you do not measure.
Measuring browsing speed
While eCommerce progressive web apps deliver the fastest possible experience, the popular tools to measure site speed do not measure them correctly. PWAs are essentially SPAs, and most speed measurement tools are geared to measuring multi-page websites that rely on network-based metrics, such as the number of assets per page. Some of the technology that enables the amazing speeds of PWAs, including intelligent caching and content prefetching, actually add assets to the page, meaning that most speed measurement tools will estimate longer load times than are actually delivered.
Layer0's CTO, Ishan Anand, phrased it best, saying that it’s time for a mindset shift in terms of website speed. First, from an emphasis on first-page loads to one that includes first-page and subsequent page loads. Second, it’s time to transition from network-based metrics to perceptual metrics, or metrics that are based on how fast consumers perceive your website to be.
Speed Index, First Contentful Paint, and First Paint time are a few examples of perception-based metrics that determine how long it takes for content to be displayed to consumers. If this is slow, users see the site as janky - and feel antsy to get to the next page. In fact, Ericsson researchers have found that slow load times create the same psychological discomfort for users as watching a horror movie, or waiting in line at a store. This is far from optimal, as shoppers go online for convenience and ease - not for long wait time.
There is one tool that does measure browsing transitions: WebPageTest. This tool measures browsing speed at key points in the customer journey. You can find detailed info about setting up WebPageTest to measure browsing transitions along with five tips to improve browsing speed, in this article.
Speedy journeys and nearly instant websites don't only benefit customers, they benefit you via incremental lifts to conversions, SEO ranking & traffic, and ROI from SEM.
Venus Fashion is a proven case study. Venus went from only 15.25% of site pages loading in under a second to 72.75%, with a median load time of 320 ms on Layer0. The result: A 24% increase in conversion rate. Check out the incredible speed on your phone: https://www.venus.com/.
It’s clear, sites that don’t load fast will rank lower, giving lightning-fast sites the opportunity to rise. Consumers attention won’t last if your site has them waiting - even for just a couple seconds, and ignoring this will cost you loyal customers and revenue.
Don’t lose your customers to your competitors; it’s time to rethink the way you optimize for site speed. Measuring and optimizing for first-page loads alone is not enough. Providing consumers with an instant customer journey from landing through checkout is your best bet for rising above the noise, extending your reach, and increasing conversions and eCommerce revenue. But to do this you need to change the way you think about website speed.
A shift from focusing on first-page loads to optimizing for every page load is required, as is a transition from relying on network-based speed metrics to perceptual metrics, such as First Contentful Paint.