We presented the Bowen Craggs Google Analytics Benchmark 2017 in a web meeting this week. Here, in the first of two posts (second to follow next week), Andrew Rigby gives the headline findings, including data for referral traffic, device usage and landing pages.
The Bowen Craggs Google Analytics Benchmark has been running since 2014, collating and comparing data from a range of corporate websites across a variety of sectors, and representing a range of activity.
Although the constituent websites may have changed, what we have measured has not, so the data is comparable and shows trends over time.
The 2017 benchmark covers the period between May 2016 and April 2017, collecting Google Analytics data from 24 corporate websites.
The growth of mobile, organic and social traffic
We noted some clear trends:
The growth of organic traffic to corporate sites – with half of the visits to the sites in the benchmark coming via unpaid search – means search engine performance is increasingly important. Visits by users referred to corporate sites from other sites are dropping
Users referred to corporate sites from social networks still represent a small proportion of overall traffic at 2.5 per cent, but some companies are starting to see significant social traffic. Facebook is the largest single social source, but those users are the most likely of social referrals to bounce and have the shortest, shallowest visits
Mobile traffic to corporate sites is generally increasing and now accounts for an average of one in four visits, but so is the variation across sites in the benchmark: for some sites it is one in three visits
The length and depth of the average site visit has dropped slightly. The average visit included 2.42 pages and lasted 2m 1 second, compared to 2.6 pages per session and 2 minutes 12 seconds in 2016
The home page is still very important on corporate sites and the most common entry point with 27 per cent of traffic, but over the years the benchmark has been running there is a clear increase in the number of pages on which users enter sites. Understanding the variety of entry points, and where users then go on a particular site, can be very helpful to web managers in ensuring smooth user journeys.
Nobody is average, so context is key
The busiest site had 37 million visits per year; the quietest 530,000, and an average of 7 million.
Yet these headline figures highlight perhaps the single biggest takeaway from the benchmark: the variation in user behaviour between sites. The companies at the extreme ends of the visits range obviously affect the average; many sites in the benchmark attract in the region of 3 million visits per year.
It is important, then, to not just compare the traffic on your site against an average – although that may be useful – but to also drill down into the detail of the comparison. A site with 4 million visits may be below the average, but in fact sits above the median as it has attracted more visits than many of the companies in the benchmark.
But even then – as many of you will be thinking as you read this – simply comparing figures is often meaningless without the context to understand them.
In next week’s commentary we’ll look at the importance of putting web analytics data in context, and how you can draw company-specific insight from it.
– Andrew Rigby
You can be part of the analytics benchmark, and enjoy a host of other benefits, by joining The Bowen Craggs Club, an exclusive network for the most engaged online corporate communications professionals. It is aimed at individuals and companies who believe in the need for world-leading corporate web estates. Although most group members work in Fortune 500 corporations, we welcome senior managers from public sector and non-governmental organizations with responsibility for large web presences.
For more information on the Club, visit our website, or contact Lisa Hayward, firstname.lastname@example.org.
To discuss our measurement services, including how we can help with analytics and visitor surveys, please contact Dan Drury email@example.com or see our website.