Home / Analysis / Research: online retailers unprepared for Christmas surge?

Research: online retailers unprepared for Christmas surge?

Many UK online retailers have just completed their most successful quarter due to the Coronavirus pandemic. Econsultancy reported that 77% of British consumers now do at least part of their online shopping online. This figure is up from 61% in 2019.

Against this backdrop and in the run up to the traditionally busiest period for retailers, Edge Testing Solutions (Edge) set out to test the websites of 258 leading UK online retailers. The testing specialist used industry-standard tools to see how retailers’ websites are holding up to increased traffic by determining factors which lead to better customer experience and conversions, including: load time, navigation tools, image quality and broken links.

Sharon Hamilton, MD of Edge Testing commented on the research: “The findings of the study make for fascinating reading. In these challenging times e-commerce is a vital business model not just for success but also currently, for survival. Key surprises were around performance. A retailer cannot afford to have a slow or unresponsive website – especially as customers are demanding immediate response times and the majority are happy to move on to a competitors’ website if kept waiting.”

Top 20 retail sites results
Edge analysed all the scores from the various different tests deployed. The following table highlights the top 20 retailers who Edge considers are the frontrunners in terms of having fast, intuitive and well-functioning websites. Leading the field is footwear retailer Gabor, scoring a total of 563.2 points out of a possible 600, where 100 points were awarded for fully loaded time, system usability scale, average image weight, functioning links, Yslow score and PageSpeed score.

Out of the top 20, fashion retailers dominated, with 40% gaining a top slot, including Long Tall Sally, Louis Vuitton, Jack & Jones, Burberry, Victoria’s Secrets, The North Face, Whistles and Gant. Technology and home improvement followed with 15% each, including retailers Microsoft, Apple, Novatech and Clas Ohlson, Ikea and Victorian Plumbing.

 

1 Gabor
2 Clas Ohlson
3 Microsoft
4 Apple
5 Zalando
6 Long Tall Sally
7 Thomas Sabo
8 Louis Vuitton
9 Ikea
10 Victorian Plumbing
11 Jack & Jones
12 Thorntons
13 Lloyds Pharmacy
14 Novatech
15 Burberry
16 Beauty Bay
17 Victoria’s Secret
18 The North Face
19 Whistles
20 Gant

Key findings analysis

Considering how important a retailer’s website is to its business success, there were some surprising scores for some of the UK’s most popular and widely used brands. When one considers that a typical customer wants a seamless, straightforward yet enjoyable shopping experience, some retailers appear to be going out of their way to stymie their customers’ online buying desires.

  • Need for speed – Edge analysed which websites responded within acceptable and measurable baselines and which performed poorly, equating to lost customers and lost revenue. With Black Friday and Christmas approaching all retail websites have a strong need for speed.

Loading speed was a key area requiring improvement with 16% of the UK’s leading online retailers taking an average 10 seconds to fully load. Even with multiple visits the average loading time was seven seconds. What happens when peak retail traffic kicks in for the Black Friday and Christmas online shopping periods? At what point will a potential customer’s patience break and a competitor’s website opened instead?

Key figures:

  • 16% (42) of landing pages took on average ten seconds to fully load with three sites taking longer than 20 seconds.
  • 89% of sites were lower than a ‘C’ grade from the Yahoo! speed test
  • Similar results from Google’s speed test – 85% ‘C’ or lower with 51 of the tested websites receiving an ‘F’
  • Average YSlow and PageSpeed scores were 66% and 62%.
  • Average first page load speed was seven seconds.
  • Ease of navigation – Edge examined which retailers had put in the effort to produce a seamless and stress-free shopping experience and which had turned their websites into an obstacle course. Positively, using the System Usability Scale, we found the vast majority (98%) of the sites’ navigation features to be satisfactory, achieving from A-C grades, with the exception of four retailers, all scoring 67.5, a concerning the usability scale score gradings gives them a D grade (poor). Navigation does not appear to be a problem with the vast majority of websites allowing the customers to move freely through the various browse and sales paths. The statement is true as 98% of the websites tested, received a grading of A-C. From our testing, we found that fashion/footwear retailers had the highest website usability scores, with 75% making the top 20 scores.
  • Image quality – poor imaging impacts the brand in question’s credibility, the length of time a customer stays on the website, conversion rates for sales, and the level of repeat visits. Websites that use large, poorly optimised images also typically take longer to load. Only 9% of tested websites displayed images to an ‘A’ grade standard, according to the Website Speed Test tool.
    • 185 sites had average image weights under 0.5 MB
    • Five websites including two of the major supermarkets had image weight scores of four MB and above.

Image quality provided interesting feedback. Low image weight is good because large images slow down web pages creating less than optimal user experience. Those with low image weight have optimised their images either by using a plugin or script, which in turn speeds up the load time of the page.

The top 20 retailers with the lowest image scoring all use high quality and multiple images to sell their goods and services. However, so did the bottom 20 which included two supermarkets, two high street chemists, a book retailer and several popular high street clothing stores. Weighty images hit load times hard and when you multiply this by many thousands of users that’s a lot of people sat in limbo waiting to shop.

  • Functional vs non- functional links – a surprising number of websites do not perform ongoing housekeeping with ‘dead links’ to obsolete pages, other brand websites and external websites. This can lead to customer frustration as they are guided down time-wasting blind alleys. So we measured the quantity of working and non-working links for each website tested. On average, 3% of links on sites were not functioning however on 7% of sites, more than 10% of links were non-functioning. Ensuring a website remains relevant and operating efficiently is achieved through good website management and housekeeping. This means ensuring all links function correctly and that obsolete links are removed promptly. Only six websites scored a perfect 100% including Ikea and Pets at Home with another 155 scoring 99%. At the other end of the scale one of the three worst offenders was a major supermarket with a score of 65.3%. One of the primary global online retailers operating in the UK also only managed to score 86.1%.
  • Security – of the 258 websites tested only five totally blocked all the measurement tools employed by Edge with an additional two blocking all but one of our toolsets. From a security aspect, we were able to deploy all of our tools against most websites – the tools deployed were not intrusive or malicious however our activities did reflect the excellent cyber resistance of a small number of the retailers’ websites tested. The remaining 95% of websites allowed our tools to analyse their websites and this may be a potential cause for concern.

Daily repeated tests were conducted on the websites from 11 – 18 August 2020. For each test, Edge identified the top 20 retailers and this information is in the full report.

Sharon Hamilton, MD of Edge Testing concludes: “During a time where consumers are shifting to online shopping more than ever before, Edge’s research reveals that retailers still have room for improvement, with slow speed, poor image quality and broken links threatening user experience. Many of the poorer performers were well known brands who cannot afford to let a poor performing website impact forthcoming sales.”

In light of the findings, Edge recommends that retailers undertake:

  1. Load and performance review of their e-commerce capability which would include analysing:
  • Current website performance capability (to provide a benchmark)
  • Measuring scalability – most retailers should have the necessary mechanisms in place to auto-scale their websites however auto-scale effectiveness also depends on the overall architecture of the web supporting systems
  • Web site load balancers plus all other business logic sitting behind the servers
  • Ascertaining ability to manage forecasted traffic spikes (incorporating forecasted traffic for seasonal retail campaigns plus taking into consideration the increased traffic COVID-19 has driven to e-commerce)
  • Identifying supporting server capability / capacity
  • Speed and agility – the ability to change the web site contents with the latest value propositions and services without impacting performance
  • Identifying what is slowing down the website’s performance (e.g. weighty images, ineffective caching approach, poor coding techniques, lack of compression techniques)
  • Apply the necessary changes and refinements to optimise their websites performance capability.
  1. Separate review and potentially performance test of order fulfilment – many retailers may have highly effective and well maintained websites however the weak link may be the internal or external order fulfilment mechanism. Edge works with several order fulfilment companies who now find themselves with a drastically increased level of demand through the impact of COVID-19 on e-commerce. One company stated that ‘COVID-19 has upped the ante so every week is now like Christmas of old so what will Christmas 2020 be like?’ With companies such as Yodel announcing 3,000 new jobs ready to cater for the Christmas 2020 buying surge the supporting order fulfilment systems must also be able to manage previously unseen levels of orders.
  2. Housekeeping initiative to minimise the number of broken links – this would be a straightforward exercise to undertake and there are various tools on the market which can speed up the review process. A well maintained website performs faster and provides a smoother shopping experience for the customer.
  3. Review of the websites pen test capability to withstand malware and cyber attacks – pen testing should be considered an ongoing activity rather than a set of snapshot tests. As websites are amended, expanded and include additional linkages, new cyber risks and weaknesses can be introduced. Therefore ongoing Pen Testing as a business as usual process should be undertaken.

Download the full research here.

Read the latest edition of PCR’s monthly magazine below:

 

Check Also

Research reveals automation key to future of cybersecurity but younger generations fear jobs being replaced by tech

Despite 88% of cybersecurity professionals believing automation will make their jobs easier, younger staffers are more concerned …