A study into the impact of web page speed on user experience and conversions that is still cited frequently, is one done by Akamai. Even though it dates back to 2009, the findings are still relevant today, and perhaps even more so, now that smartphone usage is well and truly entrenched in our daily lives.
- 47 percent of consumers expect a web page to load in two seconds or less. The study also found that 40 percent of consumers will wait no more than three seconds for a web page to render before abandoning the site.
- 52 percent of online shoppers stated that quick page loading is important to their site loyalty. 14 percent will begin shopping at another site, and 23 percent will stop shopping or walk away from their computer.
- 79 percent of online shoppers who experience a dissatisfying visit are less likely to buy from that site again, and 64 percent would simply purchase from another online store.
And an analysis of more than 900,000 pages on mobile – done by Google in early 2017 – showed that there is still a lot of opportunity for improving. 70 percent of the pages analysed took nearly seven seconds for the visual content above the fold to display on the screen, and more than 10 seconds to fully load all visual content above and below the fold. This led the same team to train a deep neural network to predict the impact of page speed on bounce rates, with results that should concern any site owner/operator:
Page Speed and SEO
However you look at it, page speed is an SEO problem. At the very least, Google (by far the most popular search engine) uses site speed – how quickly a website responds to web requests – as a ranking signal. At the moment, this signal looks at the speed of the desktop version of your website, but once Google’s mobile-first index officially rolls-out, it will likely switch to specifically looking at page speed on mobile.
— Jennifer Slegg (@jenstar) April 10, 2015
But page speed also influences other factors that could affect how your site ranks, such as:
- Click-through rate – users who have previously visited your site – and found the experience unsatisfactory – will be less likely to visit your site again, no matter how many times it shows up in search results.
- Direct and repeat traffic – any unpleasant experience on a particular site will also impact the amount of direct and repeat traffic the site attracts.
- Bounce rate – a high bounce rate could suggest low quality content, lack of relevance, or a poor user experience. And slow page speed is definitely one component of a poor user experience.
- Dwell time – an offshoot of bounce rate, users spending more time on your site when coming from a Google search is seen as a good indicator of quality.
- Crawl rate – Google does their best to ensure any Googlebot activity doesn’t disrupt the experience of users visiting the site. One way they do this is by continually assessing the number of simultaneous parallel connections Googlebot may use to crawl the site, as well as the time it has to wait between the fetches, and adjusting the crawl rate accordingly. If your site responds quickly, the crawl rate goes up, while decreasing on sites with a slow response time. On smaller websites this shouldn’t be a problem, but on larger, frequently updated websites, having a low crawl rate can result in pages not being indexed at all, or not quickly enough, leading to a lower site rank. Google illustrates this with a few choice FAQs, including:
Q: Does site speed affect my crawl budget? How about errors?
A: Making a site faster improves the users’ experience while also increasing crawl rate. For Googlebot a speedy site is a sign of healthy servers, so it can get more content over the same number of connections. On the flip side, a significant number of 5xx errors or connection timeouts signal the opposite, and crawling slows down.
We recommend paying attention to the Crawl Errors report in Search Console and keeping the number of server errors low.
So I guess there are two aspects here when you look at server speed. On the one hand there’s the kind of perceived speed in the browser, in the time it takes to render a page, and that is something that is definitely a ranking factor, it’s probably not the biggest ranking factor. And usually we try to differentiate between sites that are really slow, and sites that are kind of normal. So just optimizing on a millisecond basis is not going to affect anything in the search results.
But obviously, the faster you make the site, the more people are going to stay on your site, the more they are going to do on your site, the more they tend to recommend it to other people, so indirectly it is a factor.
which is confirmed by the Google team’s deep neural network experiment discussed in the opening paragraphs of this article.
How to Test Page Speed
John Mueller mentioned the perceived speed of a page in the browser, but for site owners and operators, this is hardly an effective measure of page speed. For something more reliable, you will need to use one of several tools available:
- Test My Site – a Google tool, specifically looking at the speed of your mobile site. With mobile page speed possibly being a ranking factor once the mobile-first index rolls out, this is a very important tool to use.
for now this is still just a plan so there's not much we can say, but generally aim for great scores on the pagespeed testing tool
— Gary "鯨理" Illyes (@methode) June 1, 2016
- PageSpeed Insights – similar to the Test My Site tool, but giving insights into page speed on both mobile and desktop. As per Gary Illyes, any result in green should be good, with any optimisations shown on the results page being suggestions to how you could improve page speed further.
- WebPageTest – the site that powers Google’s own Test My Site, but more customisable in terms of types of tests to run, and with deeper insights into the results. These include the number of requests made to your server, and the time each request takes.
- Pingdom – like WebPageTest, Pingdom lets you choose where to run the test from, but with fewer locations. The layout of the results is more visually appealing than that of WebPageTest, but there is no option to specifically test mobile page speed.
How to Optimise Page Speed
The value in using the Google page speed testing tools is that the results include a list of suggestions you can take to improve overall page speed. These may include:
- Avoid landing page redirects
- Enable compression
- Leverage browser caching
- Minify CSS
- Minify HTML
- Optimize images
- Prioritize visible content
- Reduce server response time
Reducing server response time is perhaps the most difficult suggestion to follow. There are multiple factors that influence server response time, and your ability to alter any of these depends on your web host, and whether or not your site is hosted on a shared server, a dedicated server, or a virtual private server (VPS). Other factors include whether or not you use a content delivery network (CDN), and your web host and site’s ability to support HTTP/2.
It is possible to implement all the other optimisation suggestions and still have a site that underperforms in page speed tests, all because of server response time. If this is the case, and you have also optimised your server as best you can, it might be time to consider a new host, or hosting plan.